Abidi, S, Piccardi, M, Tsang, WH & Williams, M-A 2019, 'Well-M³N: A Maximum-Margin Approach to Unsupervised Structured Prediction', IEEE Transactions on Emerging Topics in Computational Intelligence, vol. 3, no. 6, pp. 427-439.
View/Download from: Publisher's site
View description>>
Unsupervised structured prediction is of fundamental importance for the clustering and classification of unannotated structured data. To date, its most common approach still relies on the use of structural probabilistic models and the expectation-maximization (EM) algorithm. Conversely, structural maximum-margin approaches, despite their extensive success in supervised and semi-supervised classification, have not raised equivalent attention in the unsupervised case. For this reason, in this paper we propose a novel approach that extends the maximum-margin Markov networks (M3N) to an unsupervised training framework. The main contributions of our extension are new formulations for the feature map and loss function of M3N that decouple the labels from the measurements and support multiple ground-truth training. Experiments on two challenging segmentation datasets have achieved competitive accuracy and generalization compared to other unsupervised algorithms such as k-means, EM and unsupervised structural SVM, and comparable performance to a contemporary deep learning-based approach.
Alam, M, Wu, T, Xu, X, He, X, Tsang, K & Rayes, A 2019, 'Editorial: Dependable wireless industrial communications', Transactions on Emerging Telecommunications Technologies, vol. 30, no. 11.
View/Download from: Publisher's site
Amiri, M, Tofigh, F, Shariati, N, Lipman, J & Abolhasan, M 2019, 'Miniature tri-wideband Sierpinski-Minkowski fractals metamaterial perfect absorber', IET Microwaves, Antennas and Propagation, vol. 13, no. 7, pp. 991-996.
View/Download from: Publisher's site
View description>>
© The Institution of Engineering and Technology 2019. With rapidly growing adoption of wireless technologies, requirements for the design of a miniature wideband multiresonators are increasing. In this study, a compact fractal-based metamaterial structure with lumped resistors is described. The structure of the authors proposed absorber is a combination of Sierpinski curve and Minkowski fractal. The new combination provides larger capacitance and inductance in the system enabling perfect absorption at lower frequencies. The final structure with dimensions of 20 × 20 × 1.6 mm3 and an air gap of 12.5 mm provides three main resonances at frequencies of 2.1, 5.1, and 12.8 GHz with bandwidth (absorption ratio over 90%) of 840 MHz, 1.05 GHz, and 910 MHz, respectively.
Ansari, M, Zhu, H, Shariati, N & Guo, YJ 2019, 'Compact Planar Beamforming Array with Endfire Radiating Elements for 5G Applications', IEEE Transactions on Antennas and Propagation, vol. 67, no. 11, pp. 6859-6869.
View/Download from: Publisher's site
View description>>
In this paper, a compact 4×6 Butler matrix (BM) based on microstrip lines is designed and applied to a linear antenna array. The proposed design creates four beams in four different directions within the 27.5 and 28.5 GHz band. One of the advantages of this BM is a reduction in the size of the beamforming network (BFN). In order to attain this objective, the basic microstrip-based 4×4 BM is designed, and then modified to a 4×6 BM through a dual-substrate structure to avoid crossing lines using microstrip-to-slotline transitions. The BFN is cascaded with a six-element linear antenna array with endfire radiating elements. The array can be conveniently integrated into the BFN. The resulting design benefits from low-loss characteristics, ease of realization, and low fabrication cost. The array is fabricated and tested, and the experimental results are in good agreement with the simulated ones. The multi-beam antenna size is 5.6 λ × 4.6 λ including feed lines and feed network, while the new BM design is only 3.5λ0 × 1.4λ0 , which is almost half as large as the traditional one. The measured radiation patterns show that the beams cover roughly a spatial range of 90° with a peak active gain of 11 dBi.
Arockia Baskaran, AGR, Nanda, P, Nepal, S & He, X 2019, 'Testbed Evaluation of Lightweight Authentication Protocol(LAUP) fo r6LoWPAN wireless sensor networks', Concurrency and Computation: Practice and Experience, vol. 31, no. 23.
View/Download from: Publisher's site
Awwad, S, Tarvade, S, Piccardi, M & Gattas, DJ 2019, 'The use of privacy-protected computer vision to measure the quality of healthcare worker hand hygiene', International Journal for Quality in Health Care, vol. 31, no. 1, pp. 36-42.
View/Download from: Publisher's site
View description>>
(i) To demonstrate the feasibility of automated, direct observation and collection of hand hygiene data, (ii) to develop computer visual methods capable of reporting compliance with moment 1 (the performance of hand hygiene before touching a patient) and (iii) to report the diagnostic accuracy of automated, direct observation of moment 1.Observation of simulated hand hygiene encounters between a healthcare worker and a patient.Computer laboratory in a university.Healthy volunteers.Sensitivity and specificity of automatic detection of the first moment of hand hygiene.We captured video and depth images using a Kinect camera and developed computer visual methods to automatically detect the use of alcohol-based hand rub (ABHR), rubbing together of hands and subsequent contact of the patient by the healthcare worker using depth imagery.We acquired images from 18 different simulated hand hygiene encounters where the healthcare worker complied with the first moment of hand hygiene, and 8 encounters where they did not. The diagnostic accuracy of determining that ABHR was dispensed and that the patient was touched was excellent (sensitivity 100%, specificity 100%). The diagnostic accuracy of determining that the hands were rubbed together after dispensing ABHR was good (sensitivity 83%, specificity 88%).We have demonstrated that it is possible to automate the direct observation of hand hygiene performance in a simulated clinical setting. We used cheap, widely available consumer technology and depth imagery which potentially increases clinical application and decreases privacy concerns.
Bah, AO, Qin, PY, Ziolkowski, RW, Guo, YJ & Bird, TS 2019, 'A Wideband Low-Profile Tightly Coupled Antenna Array with a Very High Figure of Merit', IEEE Transactions on Antennas and Propagation, vol. 67, no. 4, pp. 2332-2343.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A wideband, low-profile, tightly coupled antenna array with a simple feed network is presented. The dipole and feed networks in each unit cell are printed on both sides of a single RT/Duroid 6010 substrate with a relative dielectric constant of 10.2. The feed network, composed of meandered impedance transformer and balun sections, is designed based on Klopfenstein tapered microstrip lines. The wide-angle impedance matching is empowered by a novel wideband metasurface superstrate. For the optimum design, scanning to 70° along the E-plane is obtained together with a very high array figure of merit P A = 2.84. The H-plane scan extends to 55°. The broadside impedance bandwidth is 5.5:1 (0.80-4.38) GHz with an active voltage standing-wave ratio value ≤2. The overall height of the array above the ground plane is 0.088λ L, where λ L is the wavelength at the lowest frequency of operation. A prototype was fabricated and tested to confirm the design concepts.
Basnet, S, He, Y, Dutkiewicz, E & Jayawickrama, BA 2019, 'Resource Allocation in Moving and Fixed General Authorized Access Users in Spectrum Access System', IEEE Access, vol. 7, pp. 107863-107873.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Spectrum access system (SAS) is a spectrum sharing framework proposed to share the spectrum between the incumbent users and the citizen broadband radio service devices, i.e. Priority access users and general authorized access (GAA) users. In this paper, we propose an interfering angle based method for the joint resource (channel and transmit power) allocation problem to the mobile and fixed GAA users. With mobile GAA users, the set of GAA users that can hear each other will change at different time instants making the resource allocation problem more challenging. The resource allocation of fixed and mobile GAA users is done considering coexistence with priority users, as well as coexistence between mobile and fixed GAA users. For the conflict-free resource allocation to fixed and mobile GAA users, we propose to use the maximum allowed transmit power for the beams of fixed GAA users that lie within the interference range of mobile GAA users. The simulation results show improved capacity from our proposed method while satisfying a predetermined interference constraint.
Bautista, MG, Zhu, H, Zhu, X, Yang, Y, Sun, Y & Dutkiewicz, E 2019, 'Compact Millimeter-Wave Bandpass Filters Using Quasi-Lumped Elements in 0.13-mu m (Bi)-CMOS Technology for 5G Wireless Systems', IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, vol. 67, no. 7, pp. 3064-3073.
View/Download from: Publisher's site
Ben, X, Gong, C, Zhang, P, Jia, X, Wu, Q & Meng, W 2019, 'Coupled Patch Alignment for Matching Cross-View Gaits', IEEE Transactions on Image Processing, vol. 28, no. 6, pp. 3142-3157.
View/Download from: Publisher's site
View description>>
© 1992-2012 IEEE. Gait recognition has attracted growing attention in recent years, as the gait of humans has a strong discriminative ability even under low resolution at a distance. Unfortunately, the performance of gait recognition can be largely affected by view change. To address this problem, we propose a coupled patch alignment (CPA) algorithm that effectively matches a pair of gaits across different views. To realize CPA, we first build a certain amount of patches, and each of them is made up of a sample as well as its intra-class and inter-class nearest neighbors. Then, we design an objective function for each patch to balance the cross-view intra-class compactness and the cross-view inter-class separability. Finally, all the local-independent patches are combined to render a unified objective function. Theoretically, we show that the proposed CPA has a close relationship with canonical correlation analysis. Algorithmically, we extend CPA to 'multi-dimensional patch alignment' that can handle an arbitrary number of views. Comprehensive experiments on CASIA(B), USF, and OU-ISIR gait databases firmly demonstrate the effectiveness of our methods over other existing popular methods in terms of cross-view gait recognition.
Bou Ghantous, G & Gill, A 2019, 'An Agile-DevOps Reference Architecture for Teaching Enterprise Agile', International Journal of Learning, Teaching and Educational Research, vol. 18, no. 7, pp. 128-144.
View/Download from: Publisher's site
View description>>
DevOps emerged as an important extension to support the Agile development for frequent and continuous software delivery. The adoption of Agile-DevOps for large scale enterprise agility depends on the most important human capability such as people competency and experience. Hence, academic education and professional training is key to the successful adoption of Agile-DevOps approach. Thus, education and training providers need to teach Agile-DevOps. However, the challenge is: how to establish and simulate an effective Agile-DevOps technology environment for teaching Enterprise Agile? This paper introduces the integrated Adaptive Enterprise Project Management (AEPM) and DevOps Reference Architecture (DRA) approach for adopting and teaching the Agile-DevOps with the help of a teaching case study from the University of Technology - Sydney (UTS), Australia. These learnings can be utilised by educators to develop and teach practice-oriented Agile-DevOps for software engineering courses. Furthermore, the experience and observations can be employed by researchers and practitioners aiming to integrate Agile-DevOps at the large enterprise scale.
Cai, Z, Yang, Y, Tang, X, Li, Z, Lu, D & Liu, Y 2019, 'Ultralow phase-noise differential oscillator using quarter stepped-impedance resonator', IEEE Microwave and Wireless Components Letters, vol. 29, no. 12, pp. 806-809.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this letter, an ultralow phase-noise differential oscillator is proposed using a balanced feedback filter based on quarter stepped-impedance resonators (QSIRs). Phase-noise performance of the designed oscillators can be significantly improved by taking advantage of high peak group delay, improved stopband suppression, and differential configuration of the proposed feedback filter. To verify the hypothesis and compare the phase-noise performance among single-ended and differential oscillators, both the prototypes are designed, fabricated, and measured. The measured results show that the designed single-ended and differential oscillators are operating at 1.953 GHz. The differential oscillator shows not only a good balanced output power of 8.29 dBm with a measured high suppression of 40.07 dB on the second harmonic but also a superior low phase-noise performance of −130.19 dBc/Hz at a frequency offset of 100 kHz. To the best of our knowledge, this is the best phase-noise performance among the state-of-the-art works designed on planar hybrid integrated circuits at a similar frequency range.
Canning, J 2019, 'Optical hoovering on plasmonic rinks', MRS Communications, vol. 9, no. 3, pp. 1072-1078.
View/Download from: Publisher's site
View description>>
© The Author(s) 2019. Excitation of surface waves on conducting materials provides a near resistance-free interface capable of a material glissade either by plasmon forces or by optical beam tractors. Analogous to an ice hockey rink, as proof-of-principle plasmon-assisted optical traction, or hoovering, of water drops on a gold surface is demonstrated. Changes in the contact angle provide a novel, low-cost nanoscale method of quantifying observable and potentially tunable changes. Variability in thresholds and movement, including jumps, is observed and can be explained by the presence of significant roughness, measured by scanning electron microscopy, with water tension. The demonstration opens a path to directly integrate various optical and plasmonic traction technologies. Implications of the phenomena and ways of improving transport and potential applications spanning configurable microfluidics, antennas, tunable lenses, diagnostics, sensing, and active Kerr and other devices are discussed.
Chen, SL, Karmokar, DK, Li, Z, Qin, PY, Ziolkowski, RW & Guo, Y 2019, 'Continuous Beam Scanning at a Fixed Frequency with a Composite Right-/Left-Handed Leaky-Wave Antenna Operating over a Wide Frequency Band', IEEE Transactions on Antennas and Propagation, vol. 67, no. 12, pp. 7272-7284.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. Fixed-frequency beam scanning leaky-wave antennas (LWAs) that can scan their main beam over a specific frequency band are highly desired for future wireless communication systems. A composite right-/left-handed (CRLH) LWA is developed in this article that facilitates fixed-frequency continuous beam scanning over a wide operational frequency band. A variation of a simple single-layer nonreconfigurable frequency-based continuous beam scanning CRLH LWA is considered first. Its dispersion properties are approximately investigated using an equivalent circuit model. It is reported how two groups of varactor diodes can be incorporated into its basic circuit model to electronically control its dispersion behavior. An optimized reconfigurable CRLH LWA with practical dc biasing lines is then realized from this nonreconfigurable design. Fixed-frequency continuous beam scanning, from backward to forward directions through broadside, is reported over a wide operational frequency band. Simulations predict that the antenna can operate from 4.75 to 5.25 GHz with the main beam being continuously scannable at each frequency point. A prototype was fabricated, assembled, and tested. The measured results confirm its simulated performance characteristics.
Chen, SL, Karmokar, DK, Li, Z, Qin, PY, Ziolkowski, RW & Guo, YJ 2019, 'Circular-polarized substrate-integrated-waveguide leaky-wave antenna with wide-angle and consistent-gain continuous beam scanning', IEEE Transactions on Antennas and Propagation, vol. 67, no. 7, pp. 4418-4428.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Circularly polarized (CP) antennas are in high demand for use in future wireless communications. To advance the development of CP substrate-integrated-waveguide (SIW) leaky-wave antennas (LWAs) with the intent to meet this demand, a novel benzene-ring-shaped slot-loaded LWA with partially reflecting wall (PRW) vias is investigated and verified to realize wide-angle continuous beam scanning with consistent gain. The dispersion features of slot-loaded SIW LWAs with PRW vias are theoretically explored using an equivalent circuit model. The CP radiation feature is investigated numerically utilizing the E- A nd H-field distributions of an initial design and its equivalent magnetic currents. The results of these studies are used to demonstrate that improved CP performance over a wide-angle scan range can be attained with a change from a standard slot shape to a benzene-ring-shaped slot. The resulting benzene-ring-shaped slot-loaded CP SIW LWA was optimized, fabricated, and measured. The measured results verify that a CP beam was continuously scanned through a wide angle from backward to forward directions with a consistent gain. The prototype exhibits a continuous 97.1° CP beam scan with a gain variation between 8 and 11.3 dBic when the source frequency is swept from 9.35 to 11.75 GHz.
Chen, X, Kong, X, Xu, M, Sandrasegaran, K & Zheng, J 2019, 'Road Vehicle Detection and Classification Using Magnetic Field Measurement', IEEE ACCESS, vol. 7, pp. 52622-52633.
View/Download from: Publisher's site
Chen, X, Ni, W, Chen, T, Collings, I, Wang, X, Liu, RP & Giannakis, GB 2019, 'Multi-Timescale Online Optimization of Network Function Virtualization for Service Chaining', IEEE Transactions on Mobile Computing, vol. 18, no. 12, pp. 2899-2912.
View/Download from: Publisher's site
View description>>
IEEE Network Function Virtualization (NFV) can cost-efficiently provide network services by running different virtual network functions (VNFs) at different virtual machines (VMs) in a correct order. This can result in strong couplings between the decisions of the VMs on the placement and operations of VNFs. This paper presents a new fully decentralized online approach for optimal placement and operations of VNFs. Building on a new stochastic dual gradient method, our approach decouples the real-time decisions of VMs, asymptotically minimizes the time-average cost of NFV, and stabilizes the backlogs of network services with a cost-backlog tradeoff of [1], for any 0. Our approach can be relaxed into multiple timescales to have VNFs (re)placed at a larger timescale and hence alleviate service interruptions. While proved to preserve the asymptotic optimality, the larger timescale can slow down the optimal placement of VNFs. A learn-and-adapt strategy is further designed to speed the placement up with an improved tradeoff. Numerical results show that the proposed method is able to reduce the time-average cost of NFV by 23% and reduce the queue length (or delay) by 74%, as compared to existing benchmarks.
Cheng, H, Zhang, J, Wu, Q & An, P 2019, 'A computational model for stereoscopic visual saliency prediction', IEEE Transactions on Multimedia, vol. 21, no. 3, pp. 678-689.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. Depth information plays an important role in human vision as it provides additional cues that distinguish objects from their backgrounds. This paper explores depth information for analyzing stereoscopic saliency and presents a computational model that predicts stereoscopic visual saliency based on three aspects of human vision: 1) the pop-out effect; 2) comfort zones; and 3) background effects. Through an analysis of these three phenomena, we find that most of the stereoscopic saliency region can be explained. Our model comprises three modules, each describing one aspect of saliency distribution, and a control function that can be used to adjust the three models independently. The relationship between the three models is not mutually exclusive. One, two, or three phenomena may appear in one image. Therefore, to accurately determine which phenomena the image conforms to, we have devised a selection strategy that chooses the appropriate combination of models based on the content of the image. Our approach is implemented within a framework based on the multifeature analysis. The framework considers surrounding regions, color/depth contrast, and points of interest. The selection strategy can improve the performance of the framework. A series of experiments on two recent eye-tracking datasets shows that our proposed method outperforms several state-of-the-art saliency models.
Cheng, Q, Shi, Z, Nguyen, D & Dutkiewicz, E 2019, 'Sensing OFDM Signal: A Deep Learning Approach', IEEE Transactions on Communications (TCOM), vol. 67, no. 11, pp. 7785-7798.
View/Download from: Publisher's site
Cheng, Z, Xiong, G, Liu, Y, Zhang, T, Tian, J & Guo, YJ 2019, 'High-efficiency Doherty power amplifier with wide OPBO range for base station systems', IET MICROWAVES ANTENNAS & PROPAGATION, vol. 13, no. 7, pp. 926-929.
View/Download from: Publisher's site
Chu, L, Shi, J & Braun, R 2019, 'The equivalent Young's modulus prediction for vacancy defected graphene under shear stress', Physica E: Low-Dimensional Systems and Nanostructures, vol. 110, pp. 115-122.
View/Download from: Publisher's site
View description>>
© 2019 Elsevier B.V. The uncertain and unavoidable vacancy defects in graphene have the inevitable influence in the extraordinary intrinsic in-plane strength. In this paper, the equivalent Young's modulus is derived from the strain energy as an important factor to evaluate the stiffness of the entire graphene based on the mechanical molecular theory. The location of vacancy defects in graphene is discussed in the regular deterministic and uncertain patterns. In terms of the boundary condition, shear stress is loaded in armchair and zigzag edges, respectively. The results show that the center concentrated vacancy defects evidently deteriorate the elastic stiffness under shear stress. The influences of periodic and regular vacancy defects are sensitive to the boundary condition. By the Monte Carlo based finite element method, vacancy defects are dispersed randomly and propagated. The results of the equivalent Young's modulus are compared with the original values in pristine graphene. The interval and mean values of Young's modulus, total strain and energy density are also provided and discussed. Compared with the results of graphene with vacancy defects under uniaxial tension, the enhancement effects of vacancy defects are less evident in the graphene under shear stress.
Chu, Y, Fu, X, Luo, Y, Canning, J, Tian, Y, Cook, K, Zhang, J & Peng, G-D 2019, 'Silica optical fiber drawn from 3D printed preforms.', Optics letters, vol. 44, no. 21, pp. 5358-5361.
View/Download from: Publisher's site
View description>>
Silica optical fiber was drawn from a three-dimensional printed preform. Both single mode and multimode fibers are reported. The results demonstrate additive manufacturing of glass optical fibers and its potential to disrupt traditional optical fiber fabrication. It opens up fiber designs for novel applications hitherto not possible.
Cui, PF, Zhang, JA, Lu, WJ, Guo, YJ & Zhu, H 2019, 'Statistical Sparse Channel Modeling for Measured and Simulated Wireless Temporal Channels', IEEE Transactions on Wireless Communications, vol. 18, no. 12, pp. 5868-5881.
View/Download from: Publisher's site
View description>>
© 2002-2012 IEEE. Time-domain wireless channels are generally modeled by Tapped Delay Line (TDL) model and its variants. These models are not effective for channel representation and estimation when the number of multipath taps is large. Compressive sensing (CS) provides a powerful tool for sparse channel modeling and estimation. Most of the research has been focusing on sparse channel estimation, while sparse channel modeling (SCM) is rarely considered for centimetre-wave channels. In this paper, we investigate statistical sparse channel modeling, using both measured and simulated channels over a frequency range of 6 to 8.5 GHz. We first introduce the triple equilibrium principle to explore the trade-off between sparsity, modeling accuracy, and algorithm complexity in SCM, and provide a methodology for characterizing the sparsity of time-domain channels using single-measurement-vector compressive sensing algorithms. Using mainly the selected wavelet dictionary and various CS reconstruction (aka recovery) algorithms, we then present comprehensive statistical sparse channel models, including channel sparsity, magnitude decaying profile, sparse coefficient distribution and atomic index distribution. Connections between the parameters of conventional TDL and sparse channel models are mathematically established. We also propose three methods for generating simulated channels from the developed sparse channel models, which validates their effectiveness.
Dai, M, Cheng, S, He, X & Wang, D 2019, 'Object tracking in the presence of shaking motions', Neural Computing and Applications, vol. 31, pp. 5917-5934.
View/Download from: Publisher's site
View description>>
© 2018 The Natural Computing Applications Forum Visual tracking can be particularly interpreted as a process of searching for targets and optimizing the searching. In this paper, we present a novel tracker framework for tracking shaking targets. We formulate the underlying geometrical relevance between a search scope and a target displacement. A uniform sampling among the search scopes is implemented by sliding windows. To alleviate any possible redundant matching, we propose a double-template structure comprising of initial and previous tracking results. The element-wise similarities between a template and its candidates are calculated by jointly using kernel functions which provide a better outlier rejection property. The STC algorithm is used to improve the tracking results by maximizing a confidence map incorporating temporal and spatial context cues about the tracked targets. For better adaptation to appearance variations, we employ a linear interpolation to update the context prior probability of the STC method. Both qualitative and quantitative evaluations are performed on all sequences that contain shaking motions and are selected from the OTB-50 challenging benchmark. The proposed approach is compared with and outperforms 12 state-of-the-art tracking methods on the selected sequences while running on MATLAB without code optimization. We have also performed further experiments on the whole OTB-50 and VOT 2015 datasets. Although the most of sequences in these two datasets do not contain motion blur that this paper is focusing on, the results of our method are still favorable compared with all of the state-of-the-art approaches.
Ding, G, Zhang, S, Khan, S, Tang, Z, Zhang, J & Porikli, F 2019, 'Feature Affinity-Based Pseudo Labeling for Semi-Supervised Person Re-Identification', IEEE Transactions on Multimedia, vol. 21, no. 11, pp. 2891-2902.
View/Download from: Publisher's site
View description>>
© 1999-2012 IEEE. Vision-based person re-identification aims to match a person's identity across multiple images, which is a fundamental task in multimedia content analysis and retrieval. Deep neural networks have recently manifested great potential in this task. However, a major bottleneck of existing supervised deep networks is their reliance on a large amount of annotated training data. Manual labeling for person identities in large-scale surveillance camera systems is quite challenging and incurs significant costs. Some recent studies adopt generative model outputs as training data augmentation. To more effectively use these synthetic data for an improved feature learning and re-identification performance, this paper proposes a novel feature affinity-based pseudo labeling method with two possible label encodings. To the best of our knowledge, this is the first study that employs pseudo-labeling by measuring the affinity of unlabeled samples with the underlying clusters of labeled data samples using the intermediate feature representations from deep networks. We propose training the network with the joint supervision of cross-entropy loss together with a center regularization term, which not only ensures discriminative feature representation learning but also simultaneously predicts pseudo-labels for unlabeled data. We show that both label encodings can be learned in a unified manner and help improve the overall performance. Our extensive experiments on three person re-identification datasets: Market-1501, DukeMTMC-reID, and CUHK03, demonstrate significant performance boost over the state-of-the-art person re-identification approaches.
Esmaili, N, Piccardi, M, Kruger, B & Girosi, F 2019, 'Analysis of healthcare service utilization after transport-related injuries by a mixture of hidden Markov models (vol 13, e0206274, 2018)', PLOS ONE, vol. 14, no. 4.
View/Download from: Publisher's site
Feng, X, Hormuth, DA & Yankeelov, TE 2019, 'An adjoint-based method for a linear mechanically-coupled tumor model: Application to estimate the spatial variation of murine glioma growth based on diffusion weighted magnetic resonance imaging.', Comput Mech, vol. 63, no. 2, pp. 159-180.
View/Download from: Publisher's site
View description>>
We present an efficient numerical method to quantify the spatial variation of glioma growth based on subject-specific medical images using a mechanically-coupled tumor model. The method is illustrated in a murine model of glioma in which we consider the tumor as a growing elastic mass that continuously deforms the surrounding healthy-appearing brain tissue. As an inverse parameter identification problem, we quantify the volumetric growth of glioma and the growth component of deformation by fitting the model predicted cell density to the cell density estimated using the diffusion-weighted magnetic resonance imaging (DW-MRI) data. Numerically, we developed an adjoint-based approach to solve the optimization problem. Results on a set of experimentally measured, in vivo rat glioma data indicate good agreement between the fitted and measured tumor area and suggest a wide variation of in-plane glioma growth with the growth-induced Jacobian ranging from 1.0 to 6.0.
Gao, X, Du, J, Zhang, T & Guo, YJ 2019, 'High-T c Superconducting Fourth-Harmonic Mixer Using a Dual-Band Terahertz On-Chip Antenna of High Coupling Efficiency', IEEE Transactions on Terahertz Science and Technology, vol. 9, no. 1, pp. 55-62.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a dual-band on-chip antenna-coupled high-T c superconducting (HTS) Josephson-junction subterahertz (THz) fourth-harmonic mixer. The antenna utilizes a couple of different structured twin slots to enable the resonant radiations at two frequencies, and integrates a well-designed coplanar waveguide network for achieving good radiation coupling and signal isolation characteristics. The electromagnetic simulations show that coupling efficiencies as high as -4 and -3.5 dB are achieved for the 160- and 640-GHz operating frequency bands, respectively. Based on this dual-band antenna, a 640-GHz HTS fourth-harmonic mixer is developed and characterized in a range of operating temperatures. The mixer exhibits a measured conversion gain of around -18 dB at 20 K and -22 dB at 40 K, respectively. The achieved intermediate frequency bandwidth is larger than 23 GHz. These are the best results reported for HTS harmonic mixers at comparable sub-THz frequency bands to date.
Geng, J, Ziolkowski, RW, Wang, K, Zhao, X, Zhou, H, Chenhu, G, Liang, X & Jin, R 2019, 'Dual CP Polarization Diversity and Space Diversity Antennas Enabled by a Compact T-Shaped Feed Structure', IEEE Access, vol. 7, pp. 96284-96296.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. A compact T-shaped feed structure (TFS) is reported that enables the realization of two types of diversity antennas: a polarization diversity antenna (PDA) and a spatial diversity antenna (SDA). Both systems have a high potential for mobile wireless communication applications. The TFS includes four ports and two independent coaxial channels with effective isolation between them all. The PDA is a dual CP omnidirectional antenna. Its optimized prototype achieves measured impedance bandwidths of 16.4% and 15.28% in its LHCP and RHCP states, respectively, and realized gains in both between 4.8 and 6.46 dBic. The inner thin coaxial cable (ITCC) of the TFS directly drives its LHCP subsystem, facilitating its improved omnidirectional performance. This ITCC is also used to directly feed the SDA's low-profile directional planar equiangular spiral antenna and its side port drives its omnidirectional RHCP antenna. Good hemispherical coverage is realized with a measured common impedance bandwidth larger than 14.35% with more than 40-dB isolation between its two ports. The corresponding measured realized gain of the SDA is between 4 and 7.8 dBic. The measured results for both optimized prototypes confirm their simulated performance characteristics.
Gill, AQ & Chew, E 2019, 'Configuration information system architecture: Insights from applied action design research', Information and Management, vol. 56, no. 4, pp. 507-525.
View/Download from: Publisher's site
View description>>
© 2018 Elsevier B.V. One of the critical information systems that enables service resilience is the service configuration information system (CiS). The fundamental challenge for organisations is the effective designing and implementation of the CiS architecture. This paper addresses this important research problem and reports insights from a completed applied action design research (ADR) project in an Australian financial services organisation. This paper aims to provide guidance to researchers and practitioners contemplating ADR, rooted in the organisational context, for practice-oriented academia-industry collaborative research. This research also contributes in terms of the CiS reference architecture design knowledge and demonstrates the applicability of the ADR method.
Gong, S, Gao, L, Xu, J, Guo, Y, Hoang, DT & Niyato, D 2019, 'Exploiting Backscatter-Aided Relay Communications with Hybrid Access Model in Device-to-Device Networks', IEEE Transactions on Cognitive Communications and Networking, vol. 5, no. 4, pp. 835-848.
View/Download from: Publisher's site
View description>>
© 2015 IEEE. The backscatter and active RF radios can complement each other and bring potential performance gain. In this paper, we envision a dual-mode radio structure that allows each device to make smart decisions on mode switch between backscatter communications (i.e., the passive mode) or RF communications (i.e., the active mode), according to the channel and energy conditions. The flexibility in mode switching also makes it more complicated for transmission control and network optimization. To exploit the radio diversity gain, we consider a wireless powered device-to-device network of hybrid radios and propose a sum throughput maximization by jointly optimizing energy beamforming and transmission scheduling in two radio modes. We further exploit the user cooperation gain by allowing the passive radios to relay for the active radios. As such, the sum throughput maximization is reformulated into a non-convex. We first present a sub-optimal algorithm based on successive convex approximation, which optimizes the relays' reflection coefficients by iteratively solving semi-definite programs. We also devise a set of heuristic algorithms with reduced computational complexity, which are shown to significantly improve the sum throughput and amenable for practical implementation.
Gong, S, Hoang, DT, Niyato, D, El Shafie, A, De Domenico, A, Strinati, EC & Hoydis, J 2019, 'Introduction to the special section on deep reinforcement learning for future wireless communication networks', IEEE Transactions on Cognitive Communications and Networking, vol. 5, no. 4, pp. 1019-1023.
View/Download from: Publisher's site
Guo, L, Zhu, H & Abbosh, A 2019, 'Phase reconfigurable microwave power divider', IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 66, no. 1, pp. 21-25.
View/Download from: Publisher's site
View description>>
© 2004-2012 IEEE. A reconfigurable power divider (PD) that can operate either in an in-phase mode or out-of-phase mode is presented. To that end, the novel concept of using tunable phase shifter, which is a reflection-type loaded coupled lines, to emulate an effective variable length transmission line (TL) is utilized. The proposed PD uses a quarter-wavelength TL, a 100 Ω isolation resistor and two tunable phase shifters. The presented theoretical analysis shows that by properly selecting the parameters of the phase shifter, it can be used to approximate the performance of variable length TL. To validate the design, a prototype of dimensions 50 mm × 25 mm is built, using Rogers RO3010 substrate, and tested. The results indicate that the device can operate as an in-phase and out-of-phase by using suitable biasing voltages. Across the band 0.9-1.1 GHz, the device has more than 12 dB return loss at all the ports and more than 15 dB isolation between the two output ports with less than 5° phase deviation for both of the in-phase and out-of-phase states.
Haider, N, Ali, A, Suarez-Rodriguez, C & Dutkiewicz, E 2019, 'Optimal Mode Selection for Full-Duplex Enabled D2D Cognitive Networks', IEEE ACCESS, vol. 7, pp. 57298-57311.
View/Download from: Publisher's site
Han, L, Zhou, M, Han, S, Jia, W, Sun, C & Fu, C 2019, 'Targeting malware discrimination based on reversed association task', Concurrency Computation, vol. 31.
View/Download from: Publisher's site
View description>>
©2018 John Wiley & Sons, Ltd. Regarding the current situation that the recognition rate of malware is decreasing, the article points out that the reason for this dilemma is that more and more targeting malware have emerged, which share little or no common feature with traditional malware. The premise of malware recognition judging whether a software is malicious or benign is actually a decision problem. We propose that malware discrimination should resort to the corresponding task or purpose. We first present a formal definition of a task and then provide further classifications of malicious tasks. Based on the decidable theory, we prove that task performed by any software is recursive and determinable. By establishing a mapping from software to task, we prove that software is many-to-one reducible to corresponding tasks. Thus, we demonstrate that software, including malware, is also recursive and can be determined by the corresponding tasks. Finally, we present the discrimination process of our method. Nine real malwares are presented, which were firstly discriminated by our method but at that time could not be identified by Kaspersky, McAfee, Symantec Norton, or Kingsoft Antivirus.
Han, L, Zhou, M, Jia, W, Dalil, Z & Xu, X 2019, 'Intrusion detection model of wireless sensor networks based on game theory and an autoregressive model', Information Sciences, vol. 476, pp. 491-504.
View/Download from: Publisher's site
View description>>
© 2018 Elsevier Inc. An effective security strategy for Wireless Sensor Networks (WSNs) is imperative to counteract security threats. Meanwhile, energy consumption directly affects the network lifetime of a wireless sensor. Thus, an attempt to exploit a low-consumption Intrusion Detection System (IDS) to detect malicious attacks makes a lot of sense. Existing Intrusion Detection Systems can only detect specific attacks and their network lifetime is short due to their high energy consumption. For the purpose of reducing energy consumption and ensuring high efficiency, this paper proposes an intrusion detection model based on game theory and an autoregressive model. The paper not only improves the autoregressive theory model into a non-cooperative, complete-information, static game model, but also predicts attack pattern reliably. The proposed approach improves on previous approaches in two main ways: (1) it takes energy consumption of the intrusion detection process into account, and (2) it obtains the optimal defense strategy that balances the system's detection efficiency and energy consumption by analyzing the model's mixed Nash equilibrium solution. In the simulation experiment, the running time of the process is regarded as the main indicator of energy consumption of the system. The simulation results show that our proposed IDS not only effectively predicts the attack time and the next targeted cluster based on the game theory, but also reduces energy consumption.
Hesamian, MH, Jia, W, He, X & Kennedy, P 2019, 'Deep Learning Techniques for Medical Image Segmentation: Achievements and Challenges.', Journal of Digital Imaging, vol. 32, no. 4, pp. 582-596.
View/Download from: Publisher's site
View description>>
Deep learning-based image segmentation is by now firmly established as a robust tool in image segmentation. It has been widely used to separate homogeneous areas as the first and critical component of diagnosis and treatment pipeline. In this article, we present a critical appraisal of popular methods that have employed deep-learning techniques for medical image segmentation. Moreover, we summarize the most common challenges incurred and suggest possible solutions.
Hu, S, Xu, M, Zhang, H, Xiao, C & Gui, C 2019, 'Affective Content-aware Adaptation Scheme on QoE Optimization of Adaptive Streaming over HTTP', ACM Transactions on Multimedia Computing Communications and Applications, vol. 15, no. 3s, pp. 100-118.
View/Download from: Publisher's site
View description>>
The article presents a novel affective content-aware adaptation scheme (ACAA) to optimize Quality of Experience (QoE) for dynamic adaptive video streaming over HTTP (DASH). Most of the existing DASH adaptation schemes conduct video bit-rate adaptation based on an estimation of available network resources, which ignore user preference on affective content (AC) embedded in video data streaming over the network. Since the personal demands to AC is very different among all viewers, to satisfy individual affective demand is critical to improve the QoE in commercial video services. However, the results of video affective analysis cannot be applied into a current adaptive streaming scheme directly. Correlating the AC distributions in user's viewing history to each being streamed segment, the affective relevancy can be inferred as an affective metric for the AC related segment. Further, we have proposed an ACAA scheme to optimize QoE for user desired affective content while taking into account both network status and affective relevancy. We have implemented the ACAA scheme over a realistic trace-based evaluation and compared its performance in terms of network performance, QoE with that of Probe and Adaptation (PANDA), buffer-based adaptation (BBA), and Model Predictive Control (MPC). Experimental results show that ACAA can preserve available buffer time for future being delivered affective content preferred by viewer's individual preference to achieve better QoE in affective contents than those normal contents while remain the overall QoE to be satisfactory.
Huang, L, Zhang, J, Zuo, Y & Wu, Q 2019, 'Pyramid-Structured Depth MAP Super-Resolution Based on Deep Dense-Residual Network', IEEE Signal Processing Letters, vol. 26, no. 12, pp. 1723-1727.
View/Download from: Publisher's site
View description>>
© 1994-2012 IEEE. Although deep convolutional neural networks (DCNN) show significant improvement for single depth map (SD) super-resolution (SR) over the traditional counterparts, most SDSR DCNNs do not reuse the hierarchical features for depth map SR resulting in blurred high-resolution (HR) depth maps. They always stack convolutional layers to make network deeper and wider. In addition, most SDSR networks generate HR depth maps at a single level, which is not suitable for large up-sampling factors. To solve these problems, we present pyramid-structured depth map super-resolution based on deep dense-residual network. Specially, our networks are made up of dense residual blocks that use densely connected layers and residual learning to model the mapping between high-frequency residuals and low-resolution (LR) depth map. Furthermore, based on the pyramid structure, our network can progressively generate depth maps of various levels by taking advantages of features from different levels. The proposed network adopts a deep supervision scheme to reduce the difficulty of model training and further improve the performance. The proposed method is evaluated on Middlebury datasets which shows improved performance compared with 6 state-of-the-art methods.
Huang, L, Zhe, T, Wu, J, Wu, Q, Pei, C & Chen, D 2019, 'Robust Inter-Vehicle Distance Estimation Method Based on Monocular Vision', IEEE Access, vol. 7, pp. 46059-46070.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Advanced driver assistance systems (ADAS) based on monocular vision are rapidly becoming a popular research subject. In ADAS, inter-vehicle distance estimation from an in-car camera based on monocular vision is critical. At present, related methods based on a monocular vision for measuring the absolute distance of vehicles ahead experience accuracy problems in terms of the ranging result, which is low, and the deviation of the ranging result between different types of vehicles, which is large and easily affected by a change in the attitude angle. To improve the robustness of a distance estimation system, an improved method for estimating the distance of a monocular vision vehicle based on the detection and segmentation of the target vehicle is proposed in this paper to address the vehicle attitude angle problem. The angle regression model (ARN) is used to obtain the attitude angle information of the target vehicle. The dimension estimation network determines the actual dimensions of the target vehicle. Then, a 2D base vector geometric model is designed in accordance with the image analytic geometric principle to accurately recover the back area of the target vehicle. Lastly, area-distance modeling based on the principle of camera projection is performed to estimate distance. The experimental results on the real-world computer vision benchmark, KITTI, indicate that our approach achieves superior performance compared with other existing published methods for different types of vehicles (including front and sideway vehicles).
Huang, X, An, P, Cao, F, Liu, D & Wu, Q 2019, 'Light-field compression using a pair of steps and depth estimation', Optics Express, vol. 27, no. 3, pp. 3557-3573.
View/Download from: Publisher's site
Huang, X, Zhang, JA, Liu, RP, Guo, YJ & Hanzo, L 2019, 'Airplane-Aided Integrated Networking for 6G Wireless: Will It Work?', IEEE Vehicular Technology Magazine, vol. 14, no. 3, pp. 84-91.
View/Download from: Publisher's site
View description>>
As demand for wireless connectivity increases, communication technology is moving toward integrating terrestrial networks with space networks. Creating this integrated space and terrestrial network (ISTN) is critically important for industries such as logistics, mining, agriculture, fisheries, and defense. However, a number of significant technological challenges must be overcome for ISTN through low-cost airborne platforms and high-data-rate backbone links.
Huang, Y, Xu, J, Wu, Q, Zheng, Z, Zhang, Z & Zhang, J 2019, 'Multi-pseudo Regularized Label for Generated Data in Person Re-Identification.', IEEE Transactions on Image Processing, vol. 28, no. 3, pp. 1391-1403.
View/Download from: Publisher's site
View description>>
Sufficient training data normally is required to train deeply learned models. However, due to the expensive manual process for labelling large number of images (i.e., annotation), the amount of available training data (i.e., real data) is always limited. To produce more data for training a deep network, Generative Adversarial Network (GAN) can be used to generate artificial sample data (i.e., generated data). However, the generated data usually does not have annotation labels. To solve this problem, in this paper, we propose a virtual label called Multi-pseudo Regularized Label (MpRL) and assign it to the generated data. With MpRL, the generated data will be used as the supplementary of real training data to train a deep neural network in a semi-supervised learning fashion. To build the corresponding relationship between the real data and generated data, MpRL assigns each generated data a proper virtual label which reflects the likelihood of the affiliation of the generated data to predefined training classes in the real data domain. Unlike the traditional label which usually is a single integral number, the virtual label proposed in this work is a set of weight-based values each individual of which is a number in (0,1] called multi-pseudo label and reflects the degree of relation between each generated data to every pre-defined class of real data. A comprehensive evaluation is carried out by adopting two state-of-the-art convolutional neural networks (CNNs) in our experiments to verify the effectiveness of MpRL. Experiments demonstrate that by assigning MpRL to generated data, we can further improve the person re-ID performance on five re-ID datasets, i.e., Market-1501, DukeMTMC-reID, CUHK03, VIPeR, and CUHK01. The proposed method obtains +6.29%, +6.30%, +5.58%, +5.84%, and +3.48% improvements in rank-1 accuracy over a strong CNN baseline on the five datasets respectively, and outperforms state-of-the-art methods.
Huang, Y, Zhong, Y, Wu, Q, Dutkiewicz, E & Jiang, T 2019, 'Cost-Effective Foliage Penetration Human Detection under Severe Weather Conditions based on Auto-Encoder/Decoder Neural Network', IEEE Internet of Things Journal, vol. 6, no. 4, pp. 6190-6200.
View/Download from: Publisher's site
View description>>
IEEE Military surveillance events and rescue activities are vital missions for the Internet-of-things. To this end, foliage penetration for human detection plays an important role. However, although the feasibility of that mission has been validated, we observe that it still cannot performs promisingly under severe weather conditions such as rainy, foggy, and snowy days. Therefore, in this paper, experiments are conducted under severe weather conditions based on a proposed deep learning approach. We present an Auto-Encoder/Decoder (Auto-ED) deep neural network that can learn the deep representation and conduct classification task concurrently. Since the property of cost-effective, the device-free sensing (DFS) techniques are used to address human detection in our case. As we pursue the signal-based mission, two components are involved in the proposed Auto-ED approach. First, an encoder is utilized that encode signal-based inputs into higher dimensional tensors by fractionally-strided convolution operations. Then, a decoder is leveraged with convolution operations to extract deep representations and learn the classifier simultaneously. To verify the effectiveness of the proposed approach, we compare it with several machine learning approaches under different weather conditions. Also, a simulation experiment is conducted by adding Additive White Gaussian Noise (AWGN) to the original target signals with different Signal to Noise Ratios (SNRs). Experimental results demonstrate that the proposed approach can best tackle the challenge of human detection under severe weather conditions in the high-clutter foliage environment, which indicates its potential application values in the near future.
Jan, MA, Usman, M, He, X & Rehman, AU 2019, 'SAMS: A Seamless and Authorized Multimedia Streaming framework for WMSN-based IoMT', IEEE Internet of Things Journal, vol. 6, no. 2, pp. 1576-1583.
View/Download from: Publisher's site
View description>>
IEEE An Internet of Multimedia Things (IoMT) architecture aims to provide a support for real-time multimedia applications by using wireless multimedia sensor nodes that are deployed for long-term usage. These nodes are capable of capturing both multimedia and non-multimedia data, and form a network known as Wireless Multimedia Sensor Network (WMSN). In a WMSN, the underlying routing protocols need to provide an acceptable level of Quality-of-Service (QoS) support for the multimedia traffic. In this paper, we propose a seamless and authorized multimedia streaming framework (SAMS) for a cluster-based hierarchical WMSN. SAMS uses authentication at different levels to form secured clusters. The formation of these clusters allows only legitimate nodes to transmit captured data to their cluster heads. Each node senses the environment, stores captured data in its buffer, and waits for its turn to transmit to its cluster head. This waiting may result in an excessive packet loss and end-to-end delay for multimedia traffic. To address these issues, a channel allocation approach is proposed for inter-cluster communication. In case of buffer overflow, a member node in one cluster switches to a neighboring cluster head provided that the latter has an available channel for allocation. The experimental results show that SAMS provides an acceptable level of QoS and enhances the security of the underlying network.
Janeczko, C, Martelli, C, Canning, J & Dutra, G 2019, 'Assessment of Orchid Surfaces Using Top-Down Contact Angle Mapping', IEEE Access, vol. 7, pp. 31364-31375.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Top-down contact angle (CA) measurements are used to characterize the green leaves and purple flowers of both old and young the Cattleya warneri orchids. The top-down CA allows the characterization of large surfaces away from the leaf edge, avoiding traditional cutting required for side view CA measurement. This allows large area mapping without damaging leaves making the method amenable to fieldwork and useful in environmental diagnostics. Young leaves are found to be hydrophobic whilst old leaves become practically hydrophilic across their entirety, mostly as a result of continued exposure to changes in the environment over time. The flowers are hydrophobic because of their visual and tactile attractor function for pollinating animals and the self-cleaning of dirt and pathogens. Real-time measurement and mapping of CA of surfaces open a new tool to assess the long-term impact of plant aging, pollution, and more of organisms in the field. The method has clear applications elsewhere such as in industrial probing of surfaces and products.
Ji, L-Y, Qin, P-Y, Li, J-Y & Zhang, L-X 2019, '1-D Electronic Beam-Steering Partially Reflective Surface Antenna', IEEE Access, vol. 7, pp. 115959-115965.
View/Download from: Publisher's site
View description>>
A 1-D electronic beam-steering partially reflective surface (PRS) antenna using a new reconfigurable PRS unit cell is proposed in this paper. The proposed work addresses the challenge to achieve a large beam steering angle with small gain variation and a small number of active/lumped elements by using a reconfigurable PRS superstrate only. The PRS unit cell consists of two back-to-back T-shaped strips with one PIN diode inserted between them and a pair of trapezoid patches (a rectangular patch and a pair of triangle parasitic patches). Beam steering is achieved by controlling the different states of PIN diodes. Thanks to the trapezoid patches, the proposed unit cell can generate a larger phase difference between different states, thereby leading to a larger beam steering angle. Furthermore, due to the addition of more degrees of freedom in the proposed unit cell, the phase difference can be easily manipulated. Moreover, since the T-shaped strips in each unit cell is connected with adjacent ones, the biasing network is very simple without needing a large number of lumped elements and dc biasing lines. The beam steering property is analyzed by using phased array theory. An antenna prototype with a main beam direction towards 0°, -18° and 18° operating at 5.5 GHz in the H-plane is fabricated and measured. Good agreement between the predicted simulation and measurement results for the input reflection coefficients and radiation patterns is achieved, which validates the feasibility of the design. The measured realized gains are over 11 dBi for all states with a 0.8 dBi gain variation.
Jiang, S, Li, K & Xu, YDR 2019, 'Relative Pairwise Relationship Constrained Non-negative Matrix Factorisation', IEEE Transactions on Knowledge and Data Engineering, vol. 31, no. 8, pp. 1595-1609.
View/Download from: Publisher's site
View description>>
IEEE Non-negative Matrix Factorisation (NMF) has been extensively used in machine learning and data analytics applications. Most existing variations of NMF only consider how each row/column vector of factorised matrices should be shaped, and ignore the relationship among pairwise rows or columns. In many cases, such pairwise relationship enables better factorisation, for example, image clustering and recommender systems. In this paper, we propose an algorithm named, Relative Pairwise Relationship constrained Non-negative Matrix Factorisation (RPR-NMF), which places constraints over relative pairwise distances amongst features by imposing penalties in a triplet form. Two distance measures, squared Euclidean distance and Symmetric divergence, are used, and exponential and hinge loss penalties are adopted for the two measures respectively. It is well known that the so-called "multiplicative update rules" result in a much faster convergence than gradient descend for matrix factorisation. However, applying such update rules to RPR-NMF and also proving its convergence is not straightforward. Thus, we use reasonable approximations to relax the complexity brought by the penalties, which are practically verified. Experiments on both synthetic datasets and real datasets demonstrate that our algorithms have advantages on gaining close approximation, satisfying a high proportion of expected constraints, and achieving superior performance compared with other algorithms.
Ju, M, Ding, C, Guo, YJ & Zhang, D 2019, 'Remote Sensing Image Haze Removal Using Gamma-Correction-Based Dehazing Model', IEEE Access, vol. 7, pp. 5250-5261.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Haze is evident in most remote sensing (RS) images, particularly for the RS scenes captured in inclement weather, which severely hinders image interpretation. In this paper, two simple yet effective visibility restoration formulas are proposed for RGB-channel RS (RRS) images and multi-spectral RS (MSRS) images, respectively. More specifically, a robust gamma-correction-based dehazing model (RGDM) is first defined, which can better address the non-uniform illumination problem in hazy images. Then, the scene albedo restoration formula (SARF) used for the RRS images is obtained by imposing the existing prior knowledge on this RGDM, which enables us to simultaneously eliminate the interferences of haze and non-uniform illumination. In subsequence, according to Rayleigh's law, an expanded restoration formula (E-SARF) is further developed for MSRS data. Using the proposed E-SARF, the spatially varying haze in each band can be thoroughly removed without using any extra information. The experiments are conducted on the challenging RRS and MSRS images, including images with non-uniform illumination, non-uniform haze distribution, and heavy haze. The results reveal that the SARF and the E-SARF are superior to most other state-of-the-art techniques in terms of both the recover quality and the implementation efficiency.
Ju, M, Ding, C, Zhang, D & Guo, YJ 2019, 'BDPK: Bayesian Dehazing Using Prior Knowledge', IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, no. 8, pp. 2349-2362.
View/Download from: Publisher's site
View description>>
IEEE Atmospheric scattering model (ASM) has been widely used in hazy image restoration. However, the recovered albedo might deviate from the real scene once the input hazy image cannot fully satisfy the model’s assumptions such as the homogeneous atmosphere and even illumination. In this paper, we break these limitations and redefine a more reliable atmospheric scattering model (RASM) that is extremely adaptable for various practical scenarios. Benefiting from RASM, a simple yet effective Bayesian dehazing algorithm (BDPK) is further proposed based on the prior knowledge. Our strategy is to convert the single image dehazing problem into a maximum a-posteriori probability (MAP) one that can be approximated as an optimization function using the existing priori constraints. To efficiently solve this optimization function, the alternating minimizing technique (AMT) is introduced, which enables us to directly restore the scene albedo. Experiments on a number of challenging images reveal the power of BDPK on removing haze and verify its superiority over several state-of-the-art techniques in terms of quality and efficiency.
Karmokar, DK, Chen, SL, Bird, TS & Guo, YJ 2019, 'Single-Layer Multi-Via Loaded CRLH Leaky-Wave Antennas for Wide-Angle Beam Scanning With Consistent Gain', IEEE Antennas and Wireless Propagation Letters, vol. 18, no. 2, pp. 313-317.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. Achieving continuous backward-to-forward wide-angle beam scanning with consistent gain by employing composite right/left-handed (CRLH) leaky-wave antennas (LWAs) is reported. For structural and design simplicity, a single-layer one-dimensional structure is considered in which each unit cell consists of a patch shorted centrally to the ground plane. It was found that, when using only one via at the center of a unit cell, a continuous beam scan requires a large-diameter via when all other parameters remain unchanged. To eliminate this limitation while maintaining a single-layer structure, novel unit cells are proposed using multiple vias in each unit cell. An LWA design for continuous beam scan with three vias in each unit cell is investigated, and the results show good performance and design flexibility. A prototype antenna has been realized, and the measured results show that the antenna can scan the radiation beam continuously in a wide range, from -60° to +66° with a consistent gain. The measured gain variation within the scan range is only 2.9 dB, and the 3 dB gain bandwidth is 58.6%.
Khan, AA, Abolhasan, M, Ni, W, Lipman, J & Jamalipour, A 2019, 'A Hybrid-Fuzzy Logic Guided Genetic Algorithm (H-FLGA) Approach for Resource Optimization in 5G VANETs', IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, vol. 68, no. 7, pp. 6964-6974.
View/Download from: Publisher's site
Le, AT, Tran, LC, Huang, X & Guo, YJ 2019, 'Analog Least Mean Square Loop with I/Q Imbalance for Self-Interference Cancellation in Full-Duplex Radios', IEEE Transactions on Vehicular Technology, vol. 68, no. 10, pp. 9848-9860.
View/Download from: Publisher's site
View description>>
© 1967-2012 IEEE. Analog least mean square (ALMS) loop is a promising structure for self-interference (SI) mitigation in full-duplex radios due to its simplicity and adaptive capability. However, being constructed from in-phase/quadrature (I/Q) demodulators and modulators to process complex signals, the ALMS loop may face I/Q imbalance problems. Thus, in this paper, the effects of frequency-independent I/Q imbalance in the ALMS loop are investigated. It is revealed that I/Q imbalance affects the loop gain and the level of SI cancellation. The loop gain can be easily compensated by adjusting the gain at other stages of the ALMS loop. Meanwhile, the degradation on cancellation performance is proved to be insignificant even under severe conditions of I/Q imbalance. In addition, an upper bound of the degradation factor is derived to provide an essential reference for the system design. Simulations are conducted to confirm the theoretical analyses.
Le, AT, Tran, LC, Huang, X, Jay Guo, Y & Vardaxoglou, JYC 2019, 'Frequency-domain characterization and performance bounds of ALMS loop for RF self-interference cancellation', IEEE Transactions on Communications, vol. 67, no. 1, pp. 682-692.
View/Download from: Publisher's site
View description>>
© 1972-2012 IEEE. Analog least mean square (ALMS) loop is a promising method to cancel self-interference (SI) in in-band full-duplex (IBFD) systems. In this paper, the steady state analyses of the residual SI powers in both analog and digital domains are firstly derived. The eigenvalue decomposition is then utilized to investigate the frequency domain characteristics of the ALMS loop. Our frequency domain analyses prove that the ALMS loop has an effect of amplifying the frequency components of the residual SI at the edges of the signal spectrum in the analog domain. However, the matched filter in the receiver chain will reduce this effect, resulting in a significant improvement of the interference suppression ratio (ISR). It means that the SI will be significantly suppressed in the digital domain before information data detection. This paper also derives the lower bounds of ISRs given by the ALMS loop in both analog and digital domains. These lower bounds are joint effects of the loop gain, tap delay, number of taps, and transmitted signal properties. The discovered relationship among these parameters allows the flexibility in choosing appropriate parameters when designing the IBFD systems under given constraints.
Li, C, Xie, H, Fan, X, Xu, RYD, Huffel, SV, Sisson, SA & Mengersen, K 2019, 'Image denoising based on nonlocal Bayesian singular value thresholding and Stein’s unbiased risk estimator', IEEE Transactions on Image Processing, vol. 28, no. 10, pp. 4899-4911.
View/Download from: Publisher's site
Li, H, Wang, TQ & Huang, X 2019, 'Joint Adaptive AoA and Polarization Estimation Using Hybrid Dual-Polarized Antenna Arrays', IEEE ACCESS, vol. 7, pp. 76353-76366.
View/Download from: Publisher's site
Li, H, Wang, TQ, Huang, X & Guo, YJ 2019, 'Adaptive AoA and Polarization Estimation for Receiving Polarized mmWave Signals', IEEE Wireless Communications Letters.
View/Download from: Publisher's site
View description>>
IEEE This paper proposes a novel hybrid dual-polarized antenna array which exploits two orthogonally-collocated dipoles to capture the full power of a polarized millimeter wave signal. To maximize the received signal-to-noise ratio (SNR), we study the adaptive angle-of-arrival and polarization state estimation, and develop a differential beam tracking algorithm and a cross-correlation-to-power ratio polarization tracking algorithm for interleaved hybrid dual-polarized arrays. Simulation results verify the superior performance of the proposed algorithms, and confirm the significant improvement of SNR obtained by using the proposed array and algorithms.
Li, H, Wang, TQ, Huang, X, Zhang, JA & Guo, YJ 2019, 'Low-Complexity Multiuser Receiver for Massive Hybrid Array mmWave Communications', IEEE TRANSACTIONS ON COMMUNICATIONS, vol. 67, no. 5, pp. 3512-3524.
View/Download from: Publisher's site
Li, K, Ni, W, Abolhasan, M & Tovar, E 2019, 'Reinforcement Learning for Scheduling Wireless Powered Sensor Communications', IEEE Transactions on Green Communications and Networking, vol. 3, no. 2, pp. 264-274.
View/Download from: Publisher's site
View description>>
© 2017 IEEE. In a wireless powered sensor network, a base station transfers power to sensors by using wireless power transfer (WPT). Inadequately scheduling WPT and data transmission causes fast battery drainage and data queue overflow of some sensors who could have potentially gained high data reception. In this paper, scheduling WPT and data transmission is formulated as a Markov decision process (MDP) by jointly considering sensors' energy consumption and data queue. In practical scenarios, the prior knowledge about battery level and data queue length in MDP is not available at the base station. We study reinforcement learning at the sensors to find a transmission scheduling strategy, minimizing data packet loss. An optimal scheduling strategy with full-state information is also investigated, assuming that the complete battery level and data queue information are well known by the base station. This presents the lower bound of the data packet loss in wireless powered sensor networks. Numerical results demonstrate that the proposed reinforcement learning scheduling algorithm significantly reduces network packet loss rate by 60%, and increases network goodput by 67%, compared to existing non-MDP greedy approaches. Moreover, comparing the optimal solutions, the performance loss due to the lack of sensors' full-state information is less than 4.6%.
Li, W, Liu, BM, Liu, D, Liu, RP, Wang, P, Luo, S & Ni, W 2019, 'Unified Fine-grained Access Control for Personal Health Records in Cloud Computing.', IEEE journal of biomedical and health informatics, vol. 23, no. 3, pp. 1278-1289.
View/Download from: Publisher's site
View description>>
Attribute-based encryption has been a promising encryption technology to secure personal health records (PHRs) sharing in cloud computing. PHRs consist of the patient data often collected from various sources including hospitals and general practice centres. Different patients' access policies have a common access sub-policy. In this paper, we propose a novel attribute-based encryption scheme for fine-grained and flexible access control to PHRs data in cloud computing. The scheme generates shared information by the common access sub-policy which is based on different patients' access policies. Then the scheme combines the encryption of PHRs from different patients. Therefore, both time consumption of encryption and decryption can be reduced. Medical staff require varying levels of access to PHRs. The proposed scheme can also support multi-privilege access control so that medical staff can access the required level of information while maximizing patient privacy. Through implementation and simulation, we demonstrate that the proposed scheme is efficient in terms of time. Moreover, We prove the security of the proposed scheme based on security of the ciphertext-policy attribute-based encryption scheme.
Li, W, Tang, X & Yang, Y 2019, 'Design and implementation of SIW cavity-backed dual-polarization antenna array with dual high-order modes', IEEE Transactions on Antennas and Propagation, vol. 67, no. 7, pp. 4889-4894.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this communication, a novel approach for high-order modes excitation using substrate-integrated waveguide (SIW) is proposed for the implementation of a dual-polarization cavity-backed slot antenna. The antenna consists of a resonant SIW cavity with eight radiating slots and two separated SIW feeding networks. The vertically linear polarization (VLP) and horizontally linear polarization (HLP) are realized by adopting the TE430 and TE340 modes with different signal schemes. The field distributions and the surface currents of the TE430 and TE340 modes are used to analyze and illuminate the radiation mechanism. To further validate the proposed design, a 2 × 2 polarization-diverse SIW cavity-backed antenna array is fabricated and tested. The measured results show that the impedance bandwidths (S11 <-10 dB) for the two linear-polarization (LP) states are 10.73-10.9 GHz and 10.75-10.83 GHz, respectively, while the 3 dB axial ratio bandwidth for the right-hand circular polarization (RHCP) is 10.75-10.83 GHz. The measured peak gains of the two LP modes and CP mode are 13.4, 12.92, and 12.2 dBi, respectively. The proposed approach demonstrates an effective way of high-order modes (TE430 and TE340 modes) generation in a single SIW cavity, which has significant meaning in polarization-diversity applications.
Li, W, Tang, XH & Yang, Y 2019, 'A Ka-band circularly polarized substrate integrated cavity-backed antenna array', IEEE Antennas and Wireless Propagation Letters, vol. 18, no. 9, pp. 1882-1886.
View/Download from: Publisher's site
View description>>
© 2002-2011 IEEE. A high-gain 4 × 4 substrate integrated waveguide (SIW) circularly polarized (CP) antenna array at a Ka-band with a quadruple-ridge waveguide polarizer is proposed in this letter. The antenna array consists of 16 cavity-backed slot antenna elements, quadruple ridge waveguide polarizers, and SIW T-type power dividers. Dominant resonant mode TE110 is excited in the cavity-backed slot antenna element. The measured impedance bandwidth (S11 < -10 dB) is from 35.3 to 35.55 GHz, and 3 dB axial-ratio bandwidth is from 35.24 to 35.57 GHz. In addition, the maximum measured gain of 18.14 dBi at the boresight is experimentally obtained at 35.42 GHz. The antenna prototype was fabricated by a multilayer printed circuit board technology. To the best of our knowledge, the quadruple-ridge waveguide is used as polarizer to produce CP signals for the first time. Compared with an aperture antenna with a conventional rectangle or a circular waveguide polarizer, this work has high aperture radiation efficiency as well as compact size. This design idea may open a new way for development of millimeter-wave high-efficiency arrays.
Li, Z, Guo, YJ, Chen, SL & Wang, J 2019, 'A Period-Reconfigurable Leaky-Wave Antenna with Fixed-Frequency and Wide-Angle Beam Scanning', IEEE Transactions on Antennas and Propagation, vol. 67, no. 6, pp. 3720-3732.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A novel fixed-frequency beam-scanning leaky-wave antenna (LWA) based on a period-reconfigurable structure is presented. Operating at 5 GHz, the antenna consists of a slotted substrate integrated waveguide and 54 electrically small patches. Each patch element is etched with two dumbbell-shaped slots, and its operating state can be flexibly controlled by the biasing of the p-i-n diode on a parasitic strip. An ideal array model employing isotropic point sources is used for the analysis on the scanning mechanism, based on which a new method for suppressing the higher order space harmonics is developed. Using this method, the monoharmonic radiation range can be dramatically extended, and a wide-angle beam scanning can be achieved by manipulating the period length of the LWA. An FPGA controlling platform is designed for the electronic control of the antenna. The measured results validate that the proposed antenna achieves good performance of wide-angle scanning (125°) with a peak gain of 11.8 dBi at a fixed frequency.
Li, Z, Zhang, S, Wang, J, Li, Y, Chen, M, Zhang, Z & Guo, YJ 2019, 'A Method of Generating Radiation Null for Periodic Leaky-Wave Antennas', IEEE Transactions on Antennas and Propagation, vol. 67, no. 6, pp. 4241-4246.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A systematic method for generating a radiation null region in the radiation pattern of periodic leaky-wave antennas (LWAs) is proposed using the theory of effective radiation sections (ERSs). In this method, the aperture field is expanded into spatial harmonics using the mode expansion method, and the ERSs of the harmonics are studied. Based on this, the radiation null region is introduced by suppressing the radiation of the ERSs corresponding to the radiating mode. The proposed method is applied to a periodic-strip LWA. The validity of the proposed method is verified by both simulated and experimental results, showing an obvious radiation null in the prescribed angular range. This method has the advantages of easy calculation and implementation, and has little influence on the gain and beamwidth. It is illustrated that only simple modification to the antenna structure is needed to achieve nulling.
Liberal, I, Ederra, I & Ziolkowski, RW 2019, 'Control of a quantum emitter's bandwidth by managing its reactive power', PHYSICAL REVIEW A, vol. 100, no. 2.
View/Download from: Publisher's site
Liberal, I, Ederra, I & Ziolkowski, RW 2019, 'Grating Lobes in Higher-Order Correlation Functions of Arrays of Quantum Emitters: Directional Photon Bunching Versus Correlated Directions', PHOTONICS, vol. 6, no. 1.
View/Download from: Publisher's site
Lin, JY, Wong, SW, Wu, YM, Yang, Y, Zhu, L & He, Y 2019, 'Three-way multiple-mode cavity filtering crossover for narrowband and broadband applications', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 3, pp. 896-905.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. In this paper, the design of a cavity crossover with three intersecting channels is presented. The three fundamental modes of a cavity resonator, namely, TE011, TE101, and TM110 modes, are adopted to resonate at each of three channels, respectively. Due to the modal orthogonality of these fundamental modes, high isolation among three channels can be achieved. Two kinds of crossovers, for narrowband and broadband applications, are presented. For the narrowband case, the proposed crossover resonates at 2.91 GHz with the fractional bandwidth of 1.4%. For the broadband case, the proposed crossover resonates at 3 GHz with the fractional bandwidth of 24%. The isolations of both designs reach more than 50 dB. For a proof of concept, the broadband example of the cavity crossover structures is fabricated and measured. A good agreement between the simulated and the measured results verifies the accuracy of the proposed design methodology.
Lin, JY, Yang, Y, Wong, SW, Chen, RS, Li, Y, Zhang, L, He, Y & Zhu, L 2019, 'Cavity Filtering Magic-T and Its Integrations Into Balanced-to-Unbalanced Power Divider and Duplexing Power Divider', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 12, pp. 4995-5004.
View/Download from: Publisher's site
View description>>
IEEE In this article, a cavity filtering magic-T based on three fundamental modes, namely, TE₀₁₁, TE₁₀₁, and TM₁₁₀, in a single triple-mode resonator (TMR) is presented. Taking advantage of the magic-T concept, two types of cavity-based filtering power dividers (PDs) integrated with balanced and duplexing functions are investigated. For the first design of a balanced-to-unbalanced (B2U) PD, balanced functions are integrated at input ports. Three fundamental modes provide the odd- and even-symmetric field distributions so that in-phase and out-of-phase responses at output ports can be achieved. Meanwhile, the common-mode suppression can be achieved at the balanced ports, and high isolation is achieved at the single-end ports, respectively. For the second design of a duplexing PD, instead of using resistors for output ports isolation, isolated ports are applied for magic-Ts to achieve all ports impedance matched and high isolation between channels of the proposed duplexing PD. To verify the concept, a B2U PD and a duplexing PD are fabricated and tested. Good matching between simulated and measured results shows the accuracy of the proposed design methodologies.
Lin, L, Xu, L, Huang, Y, Xiang, Y & He, X 2019, 'On exploiting priority relation graph for reliable multi-path communication in mobile social networks', Information Sciences, vol. 477, pp. 490-507.
View/Download from: Publisher's site
View description>>
© 2018 Elsevier Inc. A mobile social network (MSN) consists of certain amount of mobile users with social characteristics, and it provides data delivery concerning social relationships between mobile users. In MSN, ordinary people contact each other more frequently if they have more social features in common. In this paper, we apply a new topology structure–priority relation graph (PRG) to evaluate the data delivery routing in MSN on the system-level. By using the natural order of nodes’ representation, the diameter, the regular degree and the multi-path technology, we determine the priority relation graph-based social feature routing (PRG-SFR) algorithm to find disjointed multi-paths in MSN. Here, the multi-path technology can be exploited for ensuring that, between each pair of sender and receiver, the important information can be delivered through a highly reliable path. Then we calculate the tolerant ability of ‘faults’ and estimate the availability of MSN on the theoretical level. Finally, we analyze the efficiency of PRG-SFR algorithm from the numerical standpoint in terms of fault tolerance, forwarding number, transmission time and delivery rate. Moreover, we make comparisons between PRG-SFR algorithm and certain state-of-the-art technologies.
Lin, W & Ziolkowski, RW 2019, 'Compact, high directivity, omnidirectional circularly polarized antenna array', IEEE Transactions on Antennas and Propagation, vol. 67, no. 7, pp. 4537-4547.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A compact omnidirectional circularly polarized (OCP) antenna system is reported, which achieves high directivity and operates with the 2.4 GHz WLAN band. It is formed by cascading several stages of electric (metallic strips) and magnetic (loops) radiators into a highly compact array. Two OCP antenna array designs are developed to demonstrate the approach and their resulting high directivities. First, a four-stage OCP antenna array is presented. It consists of three electric radiators (E-radiators) and four magnetic radiators (M-radiators) arranged in collinear formation. A prototype was realized by mechanically fabricating the folded copper loops. Measurements confirm that this compact cost-effective design generates OCP fields that have a 5.1 dBic peak LHCP realized gain in its horizontal plane. The overlapped impedance and AR bandwidth cover 130 MHz from 2.34 to 2.47 GHz. Second, a six-stage OCP antenna array with helical loops is implemented to further increase the directivity. Its prototype was realized with all-metal 3-D printing technology. Six stages of bar and helical loop radiators form five E-radiators and six M-radiators. This highly compact OCP array achieves a measured maximum realized gain of 7.1 dBic with a 110 MHz operational bandwidth that covers 2.37-2.48 GHz.
Lin, W & Ziolkowski, RW 2019, 'Electrically Small Huygens Antenna-Based Fully-Integrated Wireless Power Transfer and Communication System', IEEE Access, vol. 7, pp. 39762-39769.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. This paper introduces the first reported electrically small Huygens dual-functional wireless power transfer (WPT) and communication system operating in the 915-MHz ISM band. It is realized by the seamless combination of a Huygens linearly polarized (HLP) antenna and a highly efficient HLP rectenna. The configuration consists of two orthogonally oriented HLP subsystems. Each one intrinsically combines two pairs of metamaterial-inspired near-field resonant parasitic elements, i.e., an Egyptian axe dipole (EAD) and a capacitively loaded loop (CLL). Through the development of a very tightly coupled feed subsystem that includes the WPT mode's rectifier circuit and the communications mode's feedline while preserving their isolation, the independent operation of both functions is facilitated in an electrically small volume ( ka < 0.77 ). The measured results of its fabricated prototype agree well with their simulated values. The communications mode antenna resonates at 910 MHz and radiates a cardioid-shaped Huygens pattern with the peak gain of 2.7 dBi. The Huygens-based WPT rectenna achieves an 87.2% peak ac-to-dc conversion efficiency at 907 MHz. The dual-functional system is an ideal candidate for many emerging Internet-of-Things (IoT) wireless applications that require simultaneous wireless information and power transfer (SWIPT) and wirelessly powered communications (WPC).
Lin, W & Ziolkowski, RW 2019, 'Wirelessly Powered Light and Temperature Sensors Facilitated by Electrically Small Omnidirectional and Huygens Dipole Antennas.', Sensors, vol. 19, no. 9.
View/Download from: Publisher's site
View description>>
Wirelessly powered, very compact sensors are highly attractive for many emerging Internet-of-things (IoT) applications; they eliminate the need for on-board short-life and bulky batteries. In this study, two electrically small rectenna-based wirelessly powered light and temperature sensors were developed that operate at 915 MHz in the 902-928-MHz industrial, scientific, and medical (ISM) bands. First, a metamaterial-inspired near-field resonant parasitic (NFRP) Egyptian axe dipole (EAD) antenna was seamlessly integrated with a highly efficient sensor-augmented rectifier without any matching network. It was electrically small and very thin, and its omnidirectional property was ideal for capturing incident AC wireless power from any azimuthal direction and converting it into DC power. Both a photocell as the light sensor and a thermistor as the temperature sensor were demonstrated. The resistive properties of the photocell and thermistor changed the rectifier's output voltage level; an acoustic alarm was activated once a threshold value was attained. Second, an electrically small, low-profile NFRP Huygens antenna was similarly integrated with the same light- and temperature-sensor-augmented rectifiers. Their unidirectional nature was very suitable for surface-mounted wireless power transfer (WPT) applications (i.e., on-body and on-wall sensors). Measurements of the prototypes of both the light- and temperature-sensor-augmented omni- and unidirectional rectenna systems confirmed their predicted performance characteristics.
Lin, W, Ziolkowski, RW & Huang, J 2019, 'Electrically Small, Low-Profile, Highly Efficient, Huygens Dipole Rectennas for Wirelessly Powering Internet-of-Things Devices', IEEE Transactions on Antennas and Propagation, vol. 67, no. 6, pp. 3670-3679.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. Wireless power transfer (WPT) technologies are a major trend in emerging internet-of-things (IoT) applications. Because they negate the need for heavy, bulky batteries and can power multiple elements simultaneously, WPT systems enable very compact ubiquitous IoT wireless devices. However, the realization of high-performance, ultracompact (electrically small) rectennas, i.e., the rectifying antennas that enable midrange and far-field WPT, is challenging. We present the first electrically small ( \textit {ka} < 0.77 ) and low-profile ( 0.04~\lambda -{0} ) linearly (LP) and circularly (CP) polarized WPT rectennas at 915 MHz in the IMS band. They are facilitated by the seamless integration of highly efficient rectifiers, i.e., RF signal to dc power conversion circuits, with electrically small Huygens dipole LP and CP antennas. Their optimized prototypes have cardioid, broadside radiation patterns, and effective capture areas larger than their physical size. Experimental results validate that they achieve an 89% peak ac-to-dc conversion efficiency, effectively confirming that they are ideal candidates for many of the emerging IoT applications.
Liu, F, Liu, Y, Xu, KD, Ban, Y-L, Liu, QH & Guo, YJ 2019, 'Synthesizing Uniform Amplitude Sparse Dipole Arrays With Shaped Patterns by Joint Optimization of Element Positions, Rotations and Phases', IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, vol. 67, no. 9, pp. 6017-6028.
View/Download from: Publisher's site
Liu, H, Zhu, X, Lu, M, Sun, Y & Yeo, KS 2019, 'Design of Reconfigurable dB-Linear Variable-Gain Amplifier and Switchable-Order gm-C Filter in 65-nm CMOS Technology', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 12, pp. 5148-5158.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A system approach for a power-scalable analog baseband (ABB) design is presented in this article. Using this approach, the energy efficiency of an ABB can be maximized without compromising any other important specifications. To fulfill the feasibility study, a switchable-order gm-C lowpass filter (LPF) along with a voltage-controlled programmable-gain amplifier (VC-PGA) is designed. The selectivity of the LPF can be linearly scaled with power consumption. In addition, the power consumption of VC-PGA has a binary-weighted manner. In contrast to conventional PGAs, the gain step of the designed PGA can be continuously tuned by a control voltage. To prove the concept, the ABB is implemented in 65-nm CMOS technology. The measurements show that the frequency responses of the ABB can be configured as either fifth or seventh order with 16 gain steps. The bandwidth is approximately 50 MHz for all cases, and the gain step can be continuously tuned between 0 and 3 dB. At the high-gain mode, the output third-order intercept point and the input-referred noise of the LPF and PGA are approximate to be 8 dBm and 5 nV/sqrt Hz, respectively. The maximum power consumption of the ABB, excluding the output buffer, is approximately 19.8 mW with a 1.2-V supply voltage. The die area, excluding the pads, is only 0.18 mm2
Liu, L, Amirgholipour, S, Jiang, J, Jia, W, Zeibots, M, He, X & Amirgholipour Kasmani, S 2019, 'Performance-enhancing network pruning for crowd counting', Neurocomputing, vol. 360, pp. 246-253.
View/Download from: Publisher's site
View description>>
© 2019 The Counting Convolutional Neural Network (CCNN) has been widely used for crowd counting. However, they typically end up with a complicated network model resulting in a challenge for real-time processing. Existing solutions aim to reduce the size of the network model, but unavoidably sacrifice the network accuracy. Different from existing pruning solutions, in this paper, a new pruning strategy is proposed by considering the contributions of various filters to the final result. The filters in the original CCNN model are grouped into positive, negative and irrelevant types. We prune the irrelevant filters of which feature maps contain little information, and the negative filters determined by a mask learned from the training dataset. Our solution improves the results of the counting model without fine-tuning or retraining the pruned model. We demonstrate the advantages of our proposed approach on the problem of crowd counting. Our experimental results on benchmark datasets show that the network model pruned using our approach not only reduces the network size but also improves the counting accuracies by 4% to 17% less MAE than the state of the arts.
Liu, M, He, Y, Xue, Z, He, X & Chen, J 2019, 'MultiScan: A Private Online Virus Detection System', IEEE Consumer Electronics Magazine, vol. 8, no. 6, pp. 53-55.
View/Download from: Publisher's site
View description>>
© 2012 IEEE. Computer viruses can affect any consumer electronics that need an operating system, such as tablet, and smartphone. To address the issue of computer viruses, some detection systems that contain multiple antivirus engines are being used. The systems may save and share user-uploaded files to the community, which is not acceptable for users with high privacy requirement. This paper presents a private multiengine online virus detection system called MultiScan that can perform the isolated detection and update to guarantee that the uploaded confidential samples are not exposed to the Internet.
Liu, M, Xue, Z, He, X & Chen, J 2019, 'Cyberthreat-Intelligence Information Sharing: Enhancing Collaborative Security', IEEE Consumer Electronics Magazine, vol. 8, no. 3, pp. 17-22.
View/Download from: Publisher's site
View description>>
© 2012 IEEE. The term cyberthreat intelligence (CTI) is widespread and gaining much attention in the area of the Internet of Things (IoT), as CTI has shown capabilities regarding the defense of advanced persistent threats (APTs). The key component of CTI is the sharing of threat information, which conventionally was a time-consuming manual process.
Liu, Y, Yang, X, Jia, Y & Guo, YJ 2019, 'A Low Correlation and Mutual Coupling MIMO Antenna', IEEE Access, vol. 7, pp. 127384-127392.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. A new two-element multiple-input multiple-output (MIMO) antenna with low correlation and high port isolation is presented. First, two hybrid electromagnetic band gap (EBG) structures with the ability to support and stop surface wave propagation, respectively, are utilized simultaneously for achieving an extremely low envelope correlation coefficient (ECC). Then, based on studying of the ground current of the MIMO antenna with EBG structure, a new defected ground structure (DGS) is used to reduce the mutual coupling by controlling the polarization of the coupling field. The two antenna elements have an edge to edge spacing of 0.13\lambda where \lambda is the free space wavelength at the resonant frequency. Finally, the rectangular slots are introduced to the patch to improve the cross polarization. Experimental results show that the ECC of the MIMO antenna is lower than 0.002. Furthermore, the maximum mutual coupling (MC) reduction of 22dB can be achieved within the working bandwidth. All of above make the MIMO antenna a potential candidate for mobile terminal-based MIMO antenna systems.
Lu, Y, Fang, J, Guo, Z & Zhang, J 2019, 'Distributed Transmit Beamforming for UAV to Base Communications', China Communications, vol. 16, no. 1, pp. 15-25.
Lu, Y, Fang, J, Guo, Z & Zhang, JA 2019, 'Performance characterization and receiver design for random temporal multiple access in non-coordinated networks', China Communications, vol. 16, no. 6, pp. 173-184.
View/Download from: Publisher's site
View description>>
© 2013 China Institute of Communications. Random access is a well-known multiple access method for uncoordinated communication nodes. Existing work mainly focuses on optimizing iterative access protocols, assuming that packets are corrupted once they are collided, or that feedback is available and can be exploited. In practice, a packet may still be able to be recovered successfully even when collided with other packets. System design and performance analysis under such a situation, particularly when the details of collision are taken into consideration, are less known. In this paper, we provide a framework for analytically evaluating the actual detection performance in a random temporal multiple access system where nodes can only transmit. Explicit expressions are provided for collision probability and signal to interference and noise ratio (SINR) when different numbers of packets are collided. We then discuss and compare two receiver options for the AP, and provide detailed receiver design for the premium one. In particular, we propose a synchronization scheme which can largely reduce the preamble length. We also demonstrate that system performance could be a convex function of preamble length both analytically and via simulation, as well as the forward error correction (FEC) coding rate.
Luo, X, Shi, C, Zeng, HQ, Ewurum, HC, Wan, Y, Guo, Y, Pagnha, S, Zhang, XB, Du, YP & He, X 2019, 'Evolutionarily Optimized Electromagnetic Sensor Measurements for Robust Surgical Navigation', IEEE Sensors Journal, vol. 19, no. 22, pp. 10859-10868.
View/Download from: Publisher's site
View description>>
© 2001-2012 IEEE. Miniaturized electromagnetic sensors are increasingly introduced to navigate surgical instruments to anatomical targets during minimally invasive procedures, such as endoscopic surgery. These sensors are usually attached at the distal tips of surgical instruments to track their three-dimensional motion represented by the position and orientation in six degrees of freedom. Unfortunately, these sensors suffer from inaccurate measurements and jitter errors due to the patient movement (e.g., respiratory motion) and magnetic field distortion. This paper proposes an evolutionary computing strategy to optimize the sensor measurements and improve the tracking accuracy of surgical navigation. We modified two evolutionary computation algorithms and proposed adaptive particle swarm optimization (APSO) and observation-boosted differential evolution (OBDE) to enhance the navigation accuracy. The experimental results demonstrate that our modified algorithms to evolutionarily optimize electromagnetic sensor measurements can critically reduce the tracking error from 4.8 to 2.9 mm. In particular, OBDE outperforms APSO for electromagnetic endoscopic navigation.
Luo, Y, Zhang, JA, Huang, X, Ni, W & Pan, J 2019, 'Optimization and Quantization of Multibeam Beamforming Vector for Joint Communication and Radio Sensing', IEEE TRANSACTIONS ON COMMUNICATIONS, vol. 67, no. 9, pp. 6468-6482.
View/Download from: Publisher's site
Luong, NC, Hoang, DT, Gong, S, Niyato, D, Wang, P, Liang, YC, Kim, DI & Dinh, H 2019, 'Applications of Deep Reinforcement Learning in Communications and Networking: A Survey', IEEE Communications Surveys and Tutorials, vol. 21, no. 4, pp. 3133-3174.
View/Download from: Publisher's site
View description>>
© 1998-2012 IEEE. This paper presents a comprehensive literature review on applications of deep reinforcement learning (DRL) in communications and networking. Modern networks, e.g., Internet of Things (IoT) and unmanned aerial vehicle (UAV) networks, become more decentralized and autonomous. In such networks, network entities need to make decisions locally to maximize the network performance under uncertainty of network environment. Reinforcement learning has been efficiently used to enable the network entities to obtain the optimal policy including, e.g., decisions or actions, given their states when the state and action spaces are small. However, in complex and large-scale networks, the state and action spaces are usually large, and the reinforcement learning may not be able to find the optimal policy in reasonable time. Therefore, DRL, a combination of reinforcement learning with deep learning, has been developed to overcome the shortcomings. In this survey, we first give a tutorial of DRL from fundamental concepts to advanced models. Then, we review DRL approaches proposed to address emerging issues in communications and networking. The issues include dynamic network access, data rate control, wireless caching, data offloading, network security, and connectivity preservation which are all important to next generation networks, such as 5G and beyond. Furthermore, we present applications of DRL for traffic routing, resource sharing, and data collection. Finally, we highlight important challenges, open issues, and future research directions of applying DRL.
Lyu, B, Yang, Z & Dinh, H 2019, 'User Cooperation in Wireless-Powered Backscatter Communication Networks', IEEE Wireless Communications Letters, vol. 8, no. 2, pp. 632-635.
View/Download from: Publisher's site
View description>>
© 2012 IEEE. In this letter, we introduce new user-cooperation schemes for wireless devices in a wireless-powered backscatter communication network with the aim to improve communication and energy efficiency for the whole network. In particular, we consider two types of wireless devices which can support different communication modes, i.e., backscatter and harvest-then-transmit, and they can cooperate to deliver the information to the access point. To improve energy transmission efficiency for the devices, energy beamforming is deployed at the power beacon. We then formulate the weighted sum-rate maximization problem by jointly optimizing time schedule, power allocation, and energy beamforming. Due to the non-convex issue of the optimization problem, we employ the variable substitutions and semidefinite relaxation techniques to obtain the optimal solution. Simulation results show that the proposed cooperation framework can significantly improve the communication efficiency compared with non-cooperation approach.
Lyu, X, Ren, C, Ni, W, Tian, H, Liu, RP & Dutkiewicz, E 2019, 'Optimal Online Data Partitioning for Geo-Distributed Machine Learning in Edge of Wireless Networks', IEEE Journal on Selected Areas in Communications, vol. 37, no. 10, pp. 2393-2406.
View/Download from: Publisher's site
View description>>
© 1983-2012 IEEE. To enable machine learning at the edge of wireless networks (such as edge cloud), close to mobile users, is critical for future wireless networks, but challenging since the lower layers in edge cloud are substantially different from existing machine learning configurations in the cloud. In such geo-distributed computing environment, streaming data need to be evenly and cost-efficiently partitioned for different workers to produce an unbiased learning model with reduced parameter synchronization frequency. This paper presents a new online approach to optimally partitioning streaming data under time-varying network conditions. A new measure is proposed to quantify the evenness of data partitioning and restrain the optimization of data admission, partitioning, and processing. Stochastic gradient descent is applied to learn the optimal decisions online and asymptotically maximize the time-average utility of data partitioning. A new protocol is designed to further reduce the measurements of link costs, while preserving the asymptotic optimality, data evenness, and stability of the platform. Simulation results show that the proposed approach is superior to the state of the art in terms of throughput and cost efficiency, while only 24% of the links need to be measured to achieve the asymptotic optimality.
Makhdoom, I, Abolhasan, M, Abbas, H & Ni, W 2019, 'Blockchain's adoption in IoT: The challenges, and a way forward', Journal of Network and Computer Applications, vol. 125, pp. 251-279.
View/Download from: Publisher's site
View description>>
The underlying technology of Bitcoin is blockchain, which was initially designed for financial value transfer only. Nonetheless, due to its decentralized architecture, fault tolerance and cryptographic security benefits such as pseudonymous identities, data integrity and authentication, researchers and security analysts around the world are focusing on the blockchain to resolve security and privacy issues of IoT. However, presently, not much work has been done to assess blockchain's viability for IoT and the associated challenges. Hence, to arrive at intelligible conclusions, this paper carries out a systematic study of the peculiarities of the IoT environment including its security and performance requirements and progression in blockchain technologies. We have identified the gaps by mapping the security and performance benefits inferred by the blockchain technologies and some of the blockchain-based IoT applications against the IoT requirements. We also discovered some practical issues involved in the integration of IoT devices with the blockchain. In the end, we propose a way forward to resolve some of the significant challenges to the blockchain's adoption in IoT.
Makhdoom, I, Abolhasan, M, Lipman, J, Liu, RP & Ni, W 2019, 'Anatomy of Threats to The Internet of Things', Communications Surveys and Tutorials, IEEE Communications Society, vol. 21, no. 2, pp. 1636-1675.
View/Download from: Publisher's site
View description>>
The world is resorting to the Internet of Things (IoT) for ease of control and monitoring of smart devices. The ubiquitous use of IoT ranges from industrial control systems (ICS) to e-Health, e-Commerce, smart cities, supply chain management, smart cars, cyber physical systems (CPS), and a lot more. Such reliance on IoT is resulting in a significant amount of data to be generated, collected, processed, and analyzed. The big data analytics is no doubt beneficial for business development. However, at the same time, numerous threats to the availability and privacy of the user data, message, and device integrity, the vulnerability of IoT devices to malware attacks and the risk of physical compromise of devices pose a significant danger to the sustenance of IoT. This paper thus endeavors to highlight most of the known threats at various layers of the IoT architecture with a focus on the anatomy of malware attacks. We present a detailed attack methodology adopted by some of the most successful malware attacks on IoT, including ICS and CPS. We also deduce an attack strategy of a distributed denial of service attack through IoT botnet followed by requisite security measures. In the end, we propose a composite guideline for the development of an IoT security framework based on industry best practices and also highlight lessons learned, pitfalls and some open research challenges.
Nanda, A, Nanda, P, He, X, Jamdagni, A & Puthal, D 2019, 'A hybrid encryption technique for Secure-GLOR: The adaptive secure routing protocol for dynamic wireless mesh networks', Future Generation Computer Systems: the international journal of grid computing: theory, methods and applications.
View/Download from: Publisher's site
View description>>
As we progress in into a digital era where most aspects of our life depend upon a network of computers,it is essential to focus on digital security. Each component of a network, be it a physical network, virtualnetwork or social network requires security when transmitting data. Hence the dynamic wireless meshnetwork must also deploy high levels of security as found in current legacy networks. This paper presentsa secure Geo-Location Oriented Routing (Secure-GLOR) protocol for wireless mesh networks, whichincorporates a hybrid encryption scheme for its multilevel security framework. The hybrid encryptiontechnique improves the network’s overall performance compared to the basic encryption by using acombination of symmetric key as well as asymmetric key encryption. Using the combination of the twoencryption schemes, the performance of the network can be improved by reducing the transmitted datasize, reduced computational overhead and faster encryption–decryption cycles. In this paper discussedmultiple encryption schemes for both symmetric and asymmetric encryption, compare their performancein various experimental scenarios. Proposed security scheme achieves better performance based on theresults obtained with most viable options for our network model.
Nguyen, CT, Dinh, H, Nguyen, DN, Niyato, D, Nguyen, HT & Dutkiewicz, E 2019, 'Proof-of-Stake Consensus Mechanisms for Future Blockchain Networks: Fundamentals, Applications and Opportunities', IEEE Access, vol. 7, pp. 85727-85745.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. The rapid development of blockchain technology and their numerous emerging applications has received huge attention in recent years. The distributed consensus mechanism is the backbone of a blockchain network. It plays a key role in ensuring the network's security, integrity, and performance. Most current blockchain networks have been deploying the proof-of-work consensus mechanisms, in which the consensus is reached through intensive mining processes. However, this mechanism has several limitations, e.g., energy inefficiency, delay, and vulnerable to security threats. To overcome these problems, a new consensus mechanism has been developed recently, namely proof of stake, which enables to achieve the consensus via proving the stake ownership. This mechanism is expected to become a cutting-edge technology for future blockchain networks. This paper is dedicated to investigating proof-of-stake mechanisms, from fundamental knowledge to advanced proof-of-stake-based protocols along with performance analysis, e.g., energy consumption, delay, and security, as well as their promising applications, particularly in the field of Internet of Vehicles. The formation of stake pools and their effects on the network stake distribution are also analyzed and simulated. The results show that the ratio between the block reward and the total network stake has a significant impact on the decentralization of the network. Technical challenges and potential solutions are also discussed.
Nguyen, H, Nguyen, D, Dinh, H & Dutkiewicz, E 2019, 'Jam Me If You Can: Defeating Jammer with Deep Dueling Neural Network Architecture and Ambient Backscattering Augmented Communications', IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, vol. 37, no. 11, pp. 2603-2620.
View/Download from: Publisher's site
Nguyen, H, Nguyen, D, Dinh, H, Dutkiewicz, E & Mueck, M 2019, 'Ambient Backscatter: A Novel Method to Defend Jamming Attacks for Wireless Networks', IEEE Wireless Communications Letters.
Nguyen, HV, Dinh, H, Nguyen, DN & Dutkiewicz, E 2019, 'Optimal and Fast Real-Time Resource Slicing with Deep Dueling Neural Networks', IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, vol. 37, no. 6, pp. 1455-1470.
View/Download from: Publisher's site
View description>>
© 1983-2012 IEEE. Effective network slicing requires an infrastructure/network provider to deal with the uncertain demands and real-time dynamics of the network resource requests. Another challenge is the combinatorial optimization of numerous resources, e.g., radio, computing, and storage. This paper develops an optimal and fast real-time resource slicing framework that maximizes the long-term return of the network provider while taking into account the uncertainty of resource demands from tenants. Specifically, we first propose a novel system model that enables the network provider to effectively slice various types of resources to different classes of users under separate virtual slices. We then capture the real-time arrival of slice requests by a semi-Markov decision process. To obtain the optimal resource allocation policy under the dynamics of slicing requests, e.g., uncertain service time and resource demands, a Q-learning algorithm is often adopted in the literature. However, such an algorithm is notorious for its slow convergence, especially for problems with large state/action spaces. This makes Q-learning practically inapplicable to our case, in which multiple resources are simultaneously optimized. To tackle it, we propose a novel network slicing approach with an advanced deep learning architecture, called deep dueling, that attains the optimal average reward much faster than the conventional Q-learning algorithm. This property is especially desirable to cope with the real-time resource requests and the dynamic demands of the users. Extensive simulations show that the proposed framework yields up to 40% higher long-term average return while being few thousand times faster, compared with the state-of-the-art network slicing approaches.
Nguyen, HV, Dinh, H, Nguyen, DN, Dutkiewicz, E, Niyato, D & Wang, P 2019, 'Optimal and Low-Complexity Dynamic Spectrum Access for RF-Powered Ambient Backscatter System with Online Reinforcement Learning', IEEE Transactions on Communications, vol. 67, no. 8, pp. 5736-5752.
View/Download from: Publisher's site
Nguyen, HV, Nguyen, V-D, Dobre, OA, Nguyen, D, Dutkiewicz, E & Shin, O 2019, 'Joint Power Control and User Association for NOMA-Based Full-Duplex Systems', IEEE Transactions on Communications, vol. 67, no. 11, pp. 8037-8055.
View/Download from: Publisher's site
Nguyen, TV, Vu, HD, Nguyen, D & Nguyenb, HT 2019, 'Performance Analysis of Protograph LDPC Codes Over Large-Scale MIMO Channels With Low-Resolution ADCs', IEEE Access, vol. 7, pp. 145145-145160.
View/Download from: Publisher's site
Nie, X, Wang, L, Ding, H & Xu, M 2019, 'Strawberry Verticillium Wilt Detection Network Based on Multi-Task Learning and Attention', IEEE Access, vol. 7, pp. 170003-170011.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Plant disease detection has an inestimable effect on plant cultivation. Accurate detection of plant disease can control the spread of disease early and prevent unnecessary loss. Strawberry verticillium wilt is a soil-borne, multi-symptomatic disease. To detect strawberry verticillium wilt accurately, we first propose a disease detection network based on Faster R-CNN and multi-task learning to detect strawberry verticillium wilt. Then, the strawberry verticillium wilt detection network (SVWDN), which uses attention mechanisms in the feature extraction of the disease detection network, is proposed. SVWDN detects verticillium wilt according to the symptoms of detected plant components (i.e.,young leaves and petioles). Compared with other existing methods for detecting disease from the whole plant appearance, the SVWDN automatically classifies the petioles and young leaves while determining whether the strawberry has verticillium wilt. To provide a dataset for evaluating and testing our method, we construct a large dataset that contains 3, 531 images with 4 categories (Healthy-leaf, Healthy-petiole, Verticillium-leaf and Verticillium-petiole). Each image also has a label to indicate whether the strawberry is suffering from verticillium wilt. With the proposed strawberry verticillium wilt detection network, we achieved a mAP of 77.54% on object detection of 4 categories and 99.95% accuracy for strawberry verticillium wilt detection.
Qin, C, Ni, W, Tian, H, Lu, L & Liu, RP 2019, 'Radio over Cloud (RoC): Cloud-Assisted Distributed Beamforming for Multi-class Traffic', IEEE Transactions on Mobile Computing, vol. 18, no. 6, pp. 1368-1379.
View/Download from: Publisher's site
View description>>
IEEE Cloud has yet to be applied to computationally intensive radio signal processing, due to closely coupled computing tasks resulting from interference. This paper presents a new cloud-assisted joint beamforming architecture, where computations are decoupled for individual wireless users and pipelined for cloud execution, using Difference of Convex (DC), l1-norm approximations, and dual decompositions. User-specific tasks are constructed and aligned with the cloud to leverage computation reuses and minimize overhead. The time-complexity is dramatically improved to support networks with tens to hundreds of base stations and users, without compromising the sum rate and quality-of-service. Further, the superiority of DC to the state-of-the-art Weighted Minimum Mean Square Error (WMMSE) in terms of convex relaxation is observed and discussed. Corroborated by simulations, the reason is revealed as WMMSE aggressively increases the data rate at interim stages, hence adversely interacting with l1-norm approximation and reducing the feasible solution regions at later stages.
Qin, P-Y, Song, L-Z & Guo, YJ 2019, 'Beam Steering Conformal Transmitarray Employing Ultra-Thin Triple-Layer Slot Elements', IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, vol. 67, no. 8, pp. 5390-5398.
View/Download from: Publisher's site
Rahman, AU, Geng, J, Ziolkowski, RW, Hang, T, Hayat, Q, Liang, X, Rehman, SU & Jin, R 2019, 'Photoluminescence Revealed Higher Order Plasmonic Resonance Modes and Their Unexpected Frequency Blue Shifts in Silver-Coated Silica Nanoparticle Antennas', APPLIED SCIENCES-BASEL, vol. 9, no. 15.
View/Download from: Publisher's site
Rao, T, Li, X, Zhang, H & Xu, M 2019, 'Multi-level region-based Convolutional Neural Network for image emotion classification', Neurocomputing, vol. 333, pp. 429-439.
View/Download from: Publisher's site
View description>>
© 2018 Analyzing emotional information of visual content has attracted growing attention for the tendency of internet users to share their feelings via images and videos online. In this paper, we investigate the problem of affective image analysis, which is very challenging due to its complexity and subjectivity. Previous research reveals that image emotion is related to low-level to high-level visual features from both global and local view, while most of the current approaches only focus on improving emotion recognition performance based on single-level visual features from a global view. Aiming to utilize different levels of visual features from both global and local view, we propose a multi-level region-based Convolutional Neural Network (CNN) framework to discover the sentimental response of local regions. We first employ Feature Pyramid Network (FPN) to extract multi-level deep representations. Then, an emotional region proposal method is used to generate proper local regions and remove excessive non-emotional regions for image emotion classification. Finally, to deal with the subjectivity in emotional labels, we propose a multi-task loss function to take the probabilities of images belonging to different emotion classes into consideration. Extensive experiments show that our method outperforms the state-of-the-art approaches on various commonly used benchmark datasets.
Ren, C, Lyu, X, Ni, W, Tian, H & Liu, RP 2019, 'Distributed Online Learning of Fog Computing under Nonuniform Device Cardinality', IEEE Internet of Things Journal, vol. 6, no. 1, pp. 1147-1159.
View/Download from: Publisher's site
View description>>
© 2014 IEEE. Processing data around the point of capture, fog computing can support computationally demanding Internet-of-Things (IoT) services. Distributed online optimization is important given the size of IoT, but challenging due to time variations of random traffic and nonuniform connectivity (or cardinality) of edge servers and IoT devices. This paper presents a distributed online learning approach to asymptotically minimizing the time-average cost of fog computing in the absence of the a-priori knowledge on traffic randomness, for light-weight, and delay-tolerant application scenarios. Stochastic gradient descent is exploited to decouple the optimizations between time slots. A graph matching problem is then formulated for every time slot by decoupling and unifying the nonuniform cardinalities, and solved in a distributed manner by developing a new linear (1/2)-approximation method. We prove that the optimality loss resulting from the distributed approximate graph matching method can be compensated and diminish by increasing the learning time. Corroborated by simulations, the proposed distributed online learning is asymptotically optimal and superior to the state of the art in terms of throughput and energy efficiency.
Ren, C, Lyu, X, Ni, W, Tian, H & Liu, RP 2019, 'Profitable cooperative region for distributed online edge caching', IEEE Transactions on Communications, vol. 67, no. 7, pp. 4696-4708.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Cooperative caching can unify network storage to improve efficiency, but the effective placement and search of contents are challenging especially in distributed edge clouds with neither a-priori knowledge on content requests nor instantaneous global view. This paper establishes a new profitable cooperative region for every content request admitted at an edge server, within which the content, if cached, can be retrieved with guaranteed profit against a direct retrieval from the network backbone. This narrows down the search for the content. The caching density of the content can also be significantly reduced, e.g., to a cached copy per region. The regions are based on a novel distributed framework which allows individual servers to spontaneously admit/dispatch requests and deliver/forward contents, while asymptotically maximizing the time-average profit of caching. The cooperative region for content is erected at individual servers by comparing the upper and lower bounds for the backlogs of unsatisfied requests of the content. Simulations show the substantially improved profit of the proposed approach over existing solutions. The regions can help automate the placement of contents with reduced density and improved efficiency.
Roselin, AG, Nanda, P, Nepal, S, He, X & Wright, J 2019, 'Exploiting the Remote Server Access Support of CoAP Protocol', IEEE Internet of Things Journal, vol. 6, no. 6, pp. 9338-9349.
View/Download from: Publisher's site
Rouzbehi, K, Miranian, A, Escaño, JM, Rakhshani, E, Shariati, N & Pouresmaeil, E 2019, 'A Data-Driven Based Voltage Control Strategy for DC-DC Converters: Application to DC Microgrid', Electronics, vol. 8, no. 5, pp. 493-493.
View/Download from: Publisher's site
View description>>
This paper develops a data-driven strategy for identification and voltage control for DC-DC power converters. The proposed strategy does not require a pre-defined standard model of the power converters and only relies on power converter measurement data, including sampled output voltage and the duty ratio to identify a valid dynamic model for them over their operating regime. To derive the power converter model from the measurements, a local model network (LMN) is used, which is able to describe converter dynamics through some locally active linear sub-models, individually responsible for representing a particular operating regime of the power converters. Later, a local linear controller is established considering the identified LMN to generate the control signal (i.e., duty ratio) for the power converters. Simulation results for a stand-alone boost converter as well as a bidirectional converter in a test DC microgrid demonstrate merit and satisfactory performance of the proposed data-driven identification and control strategy. Moreover, comparisons to a conventional proportional-integral (PI) controllers demonstrate the merits of the proposed approach.
Sang, L, Xu, M, Qian, SS & Wu, X 2019, 'Multi-modal multi-view Bayesian semantic embedding for community question answering', Neurocomputing, vol. 334, pp. 44-58.
View/Download from: Publisher's site
View description>>
© 2018 Semantic embedding has demonstrated its value in latent representation learning of data, and can be effectively adopted for many applications. However, it is difficult to propose a joint learning framework for semantic embedding in Community Question Answer (CQA), because CQA data have multi-view and sparse properties. In this paper, we propose a generic Multi-modal Multi-view Semantic Embedding (MMSE) framework via a Bayesian model for question answering. Compared with existing semantic learning methods, the proposed model mainly has two advantages: (1) To deal with the multi-view property, we utilize the Gaussian topic model to learn semantic embedding from both local view and global view. (2) To deal with the sparse property of question answer pairs in CQA, social structure information is incorporated to enhance the quality of general text content semantic embedding from other answers by using the shared topic distribution to model the relationship between these two modalities (user relationship and text content). We evaluate our model for question answering and expert finding task, and the experimental results on two real-world datasets show the effectiveness of our MMSE model for semantic embedding learning.
Saputra, YM, Dinh, HT, Nguyen, D & Dutkiewicz, E 2019, 'A Novel Mobile Edge Network Architecture with Joint Caching-Delivering and Horizontal Cooperation', IEEE Transactions on Mobile Computing, pp. 1-1.
View/Download from: Publisher's site
Saputra, YM, Hoang, DT, Nguyen, DN, Dutkiewicz, E, Niyato, D & Kim, DI 2019, 'Distributed Deep Learning at the Edge: A Novel Proactive and Cooperative Caching Framework for Mobile Edge Networks', IEEE Wireless Communications Letters, vol. 8, no. 4, pp. 1220-1223.
View/Download from: Publisher's site
View description>>
We propose two novel proactive cooperative caching approaches using deep learning (DL) to predict users' content demand in a mobile edge caching network. In the first approach, a content server (CS) takes responsibilities to collect information from all mobile edge nodes (MENs) in the network and then performs the proposed DL algorithm to predict the content demand for the whole network. However, such a centralized approach may disclose the private information because MENs have to share their local users' data with the CS. Thus, in the second approach, we propose a novel distributed deep learning (DDL)-based framework. The DDL allows MENs in the network to collaborate and exchange information to reduce the error of content demand prediction without revealing the private information of mobile users. Through simulation results, we show that our proposed approaches can enhance the accuracy by reducing the root mean squared error (RMSE) up to 33.7% and reduce the service delay by 47.4% compared with other machine learning algorithms.
Seifollahi, S, Bagirov, A, Zare Borzeshi, E & Piccardi, M 2019, 'A simulated annealing‐based maximum‐margin clustering algorithm', Computational Intelligence, vol. 35, pp. 23-41.
View/Download from: Publisher's site
View description>>
Maximum‐margin clustering is an extension of the support vector machine (SVM) to clustering. It partitions a set of unlabeled data into multiple groups by finding hyperplanes with the largest margins. Although existing algorithms have shown promising results, there is no guarantee of convergence of these algorithms to global solutions due to the nonconvexity of the optimization problem. In this paper, we propose a simulated annealing‐based algorithm that is able to mitigate the issue of local minima in the maximum‐margin clustering problem. The novelty of our algorithm is twofold, ie, (i) it comprises a comprehensive cluster modification scheme based on simulated annealing, and (ii) it introduces a new approach based on the combination of k‐means++ and SVM at each step of the annealing process. More precisely, k‐means++ is initially applied to extract subsets of the data points. Then, an unsupervised SVM is applied to improve the clustering results. Experimental results on various benchmark data sets (of up to over a million points) give evidence that the proposed algorithm is more effective at solving the clustering problem than a number of popular clustering algorithms.
Shashidharan, S, Zhu, F & Yang, Y 2019, 'Coupling of supermodes in dual-core mPOF and its application in temperature and strain sensing', Optik, vol. 195.
View/Download from: Publisher's site
View description>>
© 2019 Elsevier GmbH This paper shows a novel dual core microstructured polymer fiber made of poly-methyl-methacrylate (PMMA) with an external diameter of 128 μm. The two solid fiber cores of diameter 4 μm were made of polycarbonate separated by a single air hole of 1 μm diameter at the center of the structure. The cutoff wavelength of each core was designed to be 600 nm. A very minute change in the shape/position of cores or air hole diameter 0.5 microns will result a change in the coupling length between the cores. Mode coupling in dual-core mPOF for y- and x- polarization were examined using effective index of propagating modes. The effect of temperature and strain to the mPOF causes the transmittance nulls to either red shift or blue shift. The numerical calculation from the shifts in transmission nulls shows a sensitivity up to 1.3 nm N/ m2 in the wavelength range of 600-850nm
Shashidharan, S, Zhu, F & Yang, Y 2019, 'Microstructured Multicore Polymer Optical Fiber Temperature-insensitive Stress Sensor', Optik, vol. 186, pp. 458-463.
View/Download from: Publisher's site
View description>>
© 2018 We describe a novel microstructured multicore polymer optical fiber (m-MPOF) made of PMMA with air-holes running along the entire length of the fiber. In polymer, refractive index decreases with increasing temperature, a negative thermo-optic coefficient which results in the increase of coupling with increasing temperature thus reducing the beat length and causing a blue shift in transmission nulls, but with increase in temperature the spacing between cores will also increase which results in decreases of coupling and an increases the beat length, so a red shift in transmission nulls –a positive thermal expansion coefficient. In most part of the wavelength the thermo-optic effect dominates the thermal expansion effect, but at a particular wavelength both the effects cancel each other and have a zero change in the Neff with a change in temperature. So, as a result, both these effects get nullified at certain temperature making it a temperature insensitive fiber. Numerical calculations show the fiber is temperature insensitive for a range of 40℃ to 60℃ at a wavelength of 700 nm. The fiber is shown to be sensitive to stress at 700 nm with a sensitivity of 1.6 nm/Pa, which makes the said fiber temperature insensitive stress sensor.
Shen, W, Wu, Y, Yuan, J, Duan, L, Zhang, J & Jia, Y 2019, 'Robust Distracter-Resistive Tracker via Learning a Multi-Component Discriminative Dictionary', IEEE Transactions on Circuits and Systems for Video Technology, vol. 29, no. 7, pp. 2012-2028.
View/Download from: Publisher's site
View description>>
IEEE Discriminative dictionary learning (DDL) provides an appealing paradigm for appearance modeling in visual tracking. However, most existing DDL based trackers cannot handle drastic appearance changes, especially for scenarios with background cluster and/or similar object interference. One reason is that they often suffer from the loss of subtle visual information which is critical to distinguish an object from distracters. In this paper, we explore the use of deep features extracted from the Convolutional Neural Networks (CNNs) to improve the object representation and propose a robust distracter-resistive tracker via learning a multi-component discriminative dictionary. The proposed method exploits both the intra-class and the interclass visual information to learn shared atoms and the classspecific atoms. By imposing several constraints into the objective function, the learned dictionary is reconstructive, compressive and discriminative, thus can better distinguish an object from the background. In addition, our convolutional features (deep features extracted from CNNs) have structural information for object localization and balance the discriminative power and semantic information of the object. Tracking is carried out within a Bayesian inference framework where a joint decision measure is used to construct the observation model. To alleviate the drift problem, the reliable tracking results obtained online are accumulated to update the dictionary. Both the qualitative and quantitative results on the CVPR2013 benchmark, the VOT2015 dataset and the SPOT dataset demonstrate that our tracker achieves better performance over the state-of-the-art approaches.
Shi, J, Chu, L & Braun, R 2019, 'A Kriging Surrogate Model for Uncertainty Analysis of Graphene Based on a Finite Element Method', International Journal of Molecular Sciences, vol. 20, no. 9, pp. 1-16.
View/Download from: Publisher's site
View description>>
Due to the inevitable presence of random defects, unpredictable grain boundaries in macroscopic samples, stress concentration at clamping points, and unknown load distribution in the investigation of graphene sheets, uncertainties are crucial and challenging issues that require more exploration. The application of the Kriging surrogate model in vibration analysis of graphene sheets is proposed in this study. The Latin hypercube sampling method effectively propagates the uncertainties in geometrical and material properties of the finite element model. The accuracy and convergence of the Kriging surrogate model are confirmed by a comparison with the reported references. The uncertainty analysis for both Zigzag and Armchair graphene sheets are compared and discussed.
Shi, T, Tang, MC, Wu, Z, Xu, HX & Ziolkowski, RW 2019, 'Improved signal-to-noise ratio, bandwidth-enhanced electrically small antenna augmented with internal non-foster elements', IEEE Transactions on Antennas and Propagation, vol. 67, no. 4, pp. 2763-2768.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. Non-Foster technology facilitates the ability to surpass the Chu bandwidth limit associated with electrically small antennas (ESAs). Nonetheless, in addition to challenging stability issues, the enhanced performance can come at the cost of increased noise and resistance losses generated by the active circuit. Consequently, low total efficiency and degraded signal-to-noise ratio (SNR) values can arise. Stability and SNR have dominated most reports to date; little has been discussed with regard to the underlying innovative physics of non-Foster augmented radiators. In this communication, we propose a broad bandwidth non-Foster ESA, emphasizing those aspects. By embedding a non-Foster element into the near-field resonant parasitic element of a metamaterial-inspired antenna, its electrically small size is maintained. On the other hand, a 5-times enhancement of its -10 dB fractional bandwidth (15 times its -3 dB bandwidth) is measured, significantly surpassing its passive Chu limit. Under a good matching, the measurements demonstrate that this non-Foster ESA achieves a 1.05 dBi peak gain and realizes average 5.0 dB SNR and 17 dB gain improvements over its passive counterpart.
Siyari, P, Krunz, MM & Nguyen, D 2019, 'Distributed Power Control in Single-stream MIMO Wiretap Interference Networks with Full-duplex Jamming Receivers', IEEE Transactions on Signal Processing, vol. 67, no. 3, pp. 594-608.
Song, X, Fan, X, Xiang, C, Ye, Q, Liu, L, Wang, Z, He, X, Yang, N & Fang, G 2019, 'A Novel Convolutional Neural Network Based Indoor Localization Framework With WiFi Fingerprinting', IEEE Access, vol. 7, pp. 110698-110709.
View/Download from: Publisher's site
View description>>
With the ubiquitous deployment of wireless systems and pervasive availability of smart devices, indoor localization is empowering numerous location-based services. With the established radio maps, WiFi fingerprinting has become one of the most practical approaches to localize mobile users. However, most fingerprint-based localization algorithms are computation-intensive, with heavy dependence on both offline training phase and online localization phase. In this paper, we propose CNNLoc, a Convolutional Neural Network (CNN) based indoor localization system with WiFi fingerprints for multi-building and multi-floor localization. Specifically, we devise a novel classification model and a novel positioning model by combining a Stacked Auto-Encoder (SAE) with a one-dimensional CNN. The SAE is utilized to precisely extract key features from sparse Received Signal Strength (RSS) data while the CNN is trained to effectively achieve high accuracy in the positioning phase. We evaluate the proposed system on the UJIIndoorLoc dataset and Tampere dataset and compare the performance with several state-of-the-art methods. Moreover, we further propose a newly collected WiFi fingerprinting dataset UTSIndoorLoc and test the positioning model of CNNLoc on it. The results show CNNLoc outperforms the existing solutions with 100% and 95% success rates on building-level localization and floor-level localization, respectively.
Suarez-Rodriguez, C, He, Y & Dutkiewicz, E 2019, 'Theoretical analysis of REM-based handover algorithm for heterogeneous networks', IEEE Access, vol. 7, pp. 96719-96731.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Handover has been a widely studied topic since the beginning of the mobile communications era, but with the advent of another generation, it is worth seeing it with fresh eyes. Data traffic is expected to keep growing as new use cases will coexist under the same umbrella, e.g., vehicle-to-vehicle or massive-machine-type communications. Heterogeneous networks will give way to multi-tiered networks, and mobility management will become challenging once again. Under the current approach, based uniquely on measurements, the number of handovers will soar, so will the signaling. We propose a handover algorithm that employs multidimensional radio-cognitive databases, namely radio environment maps, to predict the best network connection according to the user's trajectory. Radio environment maps have been extensively used in spectrum-sharing scenarios, and recently, some advances in other areas have been supported by them, such as coverage deployment or interference management. We also present a geometric model that translates the 3GPP specifications into geometry and introduce a new framework that can give useful insights into our proposed technique's performance. We validate our framework through Monte Carlo simulations, and the results show that a drastic reduction of at least 10% in the ping-pong handovers can be achieved, thus reducing the signaling needed.
Sun, F, Zhu, H, Zhu, X, Yang, Y & Gomez-Garcia, R 2019, 'Design of On-Chip Millimeter-Wave Bandpass Filters Using Multilayer Patterned-Ground Element in 0.13-μ m (Bi)-CMOS Technology', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 12, pp. 5159-5170.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A novel design methodology for transmission-zero (TZ) generation in on-chip millimeter-wave (mm-wave) bandpass filters (BPFs) based on the original concept of multilayer patterned-ground (MPG) element is presented in this article. Unlike most of prior-art techniques available in the technical literature, this method has two distinct features. First, it is inherently suitable for miniaturized BPF design since the MPG element can be implemented through the layers below the top-metal layer and, thus, without occupying any additional die/chip area. Second, it provides a simple but effective way to produce a TZ at the upper stopband without adversely affecting other BPF performance metrics. To fully understand the operational insight of the engineered approach, a simplified LC-equivalent behavioral circuit model for the MPG element is developed. Using this model, three second-order BPFs based on different circuit configurations are codesigned to further demonstrate the experimental feasibility of the technique. All the filter prototypes are fabricated in a standard 0.13-μ m bipolar complementary metal-oxide-semiconductor [(Bi)-CMOS] technology. The obtained on-wafer measurements show that all fabricated BPF chips have the capability to suppress the second-order harmonic by more than 30 dB, which indicates the effectiveness of the proposed integrated BPF design approach with the MPG element.
Sun, F, Zhu, H, Zhu, X, Yang, Y, Sun, Y & Zhang, X 2019, 'Design of millimeter-wave bandpass filters with broad bandwidth in Si-based technology', IEEE Transactions on Electron Devices, vol. 66, no. 3, pp. 1174-1181.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this paper, a novel design approach is proposed for on-chip bandpass filter (BPF) design with improved passband flatness and stopband suppression. The proposed approach simply uses a combination of meander-line structures with metal-insulator-metal (MIM) capacitors. To demonstrate the insight of this approach, a simplified equivalent LC-circuit model is used for theoretical analysis. Using the analyzed results as a guideline along with a full-wave electromagnetic (EM) simulator, two BPFs are designed and implemented in a standard 0.13-μm (Bi)-CMOS technology. The measured results show that good agreements between EM simulated and measured results are achieved. For the first BPF, the return loss is better than 10 dB from 13.5 to 32 GHz, which indicates a fractional bandwidth (FBW) of more than 78%. In addition, the minimum insertion loss of 2.3 dB is achieved within the frequency range from 17 to 27 GHz and the in-band magnitude ripple is less than 0.1 dB. The chip size of this design, excluding the pads, is 0.148 mm 2 . To demonstrate a miniaturized design, a second design example is given. The return loss is better than 10 dB from 17.3 to 35.9 GHz, which indicates an FBW of more than 70%. In addition, the minimum insertion loss of 2.6 dB is achieved within the frequency range from 21.4 to 27.7 GHz and the in-band magnitude ripple is less than 0.1 dB. The chip size of the second design, excluding the pads, is only 0.066 mm 2 .
Sun, HH, Ding, C, Zhu, H, Jones, B & Guo, YJ 2019, 'Suppression of Cross-Band Scattering in Multiband Antenna Arrays', IEEE Transactions on Antennas and Propagation, vol. 67, no. 4, pp. 2379-2389.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. This paper presents a novel method of suppressing cross-band scattering in dual-band dual-polarized antenna arrays. The method involves introducing chokes into low-band (LB) elements to suppress high-band (HB) scattering currents. The experimental results show that by inserting LB-pass HB-stop chokes into LB radiators, suppression of induced HB currents on the LB elements is achieved. This greatly reduces the pattern distortion of the HB array caused by the presence of LB elements. The array considered is configured as two columns of HB antennas operating from 1.71 to 2.28 GHz interleaved with a single column of LB antennas operating from 0.82 to 1.0 GHz. The realized array with choked LB element has stable and symmetrical radiation in both HB and LB.
Sun, X, Gui, G, Li, Y, Liu, RP & An, Y 2019, 'ResInNet: A Novel Deep Neural Network with Feature Reuse for Internet of Things', IEEE Internet of Things Journal, vol. 6, no. 1, pp. 679-691.
View/Download from: Publisher's site
View description>>
© 2014 IEEE. Deep neural networks (DNNs) have widely used in various Internet-of-Things (IoT) applications. Pursuing superior performance is always a hot spot in the field of DNN modeling. Recently, feature reuse provides an effective means of achieving favorable nonlinear approximation performance in deep learning. Existing implementations utilizes a multilayer perception (MLP) to act as a functional unit for feature reuse. However, determining connection weight and bias of MLP is a rather intractable problem, since the conventional back-propagation learning approach encounters the limitations of slow convergence and local optimum. To address this issue, this paper develops a novel DNN considering a well-behaved alternative called reservoir computing, i.e., reservoir in network (ResInNet). In this structure, the built-in reservoir has two notable functions. First, it behaves as a bridge between any two restricted Boltzmann machines in the feature learning part of ResInNet, performing a feature abstraction once again. Such reservoir-based feature translation provides excellent starting points for the following nonlinear regression. Second, it serves as a nonlinear approximation, trained by a simple linear regression using the most representative (learned) features. Experimental results over various benchmark datasets show that ResInNet can achieve the superior nonlinear approximation performance in comparison to the baseline models, and produce the excellent dynamic characteristics and memory capacity. Meanwhile, the merits of our approach is further demonstrated in the network traffic prediction related to real-world IoT application.
Sutton, GJ, Zeng, J, Liu, RP, Ni, W, Nguyen, DN, Jayawickrama, BA, Huang, X, Abolhasan, M, Zhang, Z, Dutkiewicz, E & Lv, T 2019, 'Enabling Technologies for Ultra-Reliable and Low Latency Communications: From PHY and MAC Layer Perspectives', IEEE Communications Surveys & Tutorials, vol. 21, no. 3, pp. 2488-2524.
View/Download from: Publisher's site
View description>>
Future 5th generation networks are expected to enable three key services-enhanced mobile broadband, massive machine type communications and ultra-reliable and low latency communications (URLLC). As per the 3rd generation partnership project URLLC requirements, it is expected that the reliability of one transmission of a 32 byte packet will be at least 99.999% and the latency will be at most 1 ms. This unprecedented level of reliability and latency will yield various new applications, such as smart grids, industrial automation and intelligent transport systems. In this survey we present potential future URLLC applications, and summarize the corresponding reliability and latency requirements. We provide a comprehensive discussion on physical (PHY) and medium access control (MAC) layer techniques that enable URLLC, addressing both licensed and unlicensed bands. This paper evaluates the relevant PHY and MAC techniques for their ability to improve the reliability and reduce the latency. We identify that enabling long-term evolution to coexist in the unlicensed spectrum is also a potential enabler of URLLC in the unlicensed band, and provide numerical evaluations. Lastly, this paper discusses the potential future research directions and challenges in achieving the URLLC requirements.
Tang, MC, Duan, Y, Wu, Z, Chen, X, Li, M & Ziolkowski, RW 2019, 'Pattern reconfigurable, vertically polarized, low-profile, compact, near-field resonant parasitic antenna', IEEE Transactions on Antennas and Propagation, vol. 67, no. 3, pp. 1467-1475.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A vertically polarized, low-profile, compact, near-field resonant parasitic antenna with pattern reconfigurability is demonstrated. The antenna has three dynamic end-fire states facilitated with only three p-i-n diodes. The radiation pattern in each state covers more than 120° in its azimuth plane and, hence, it achieves beam scanning that covers the entire azimuth plane. The antenna height and transverse size are, respectively, only 0.048λ 0 and 0.1λ 02 . Measured results, in good agreement with their simulated values, demonstrate that the antenna exhibits a 11% fractional impedance bandwidth, and a 6.6 dBi peak realized gain in all three of its pattern-reconfigurable states. Stable and high peak realized gain values are realized over its entire operational band surrounding 2.22 GHz.
Tang, M-C, Li, D, Chen, X, Wang, Y, Hu, K & Ziolkowski, RW 2019, 'Compact, Wideband, Planar Filtenna With Reconfigurable Tri-Polarization Diversity', IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, vol. 67, no. 8, pp. 5689-5694.
View/Download from: Publisher's site
Tang, MC, Wu, Z, Shi, T & Ziolkowski, RW 2019, 'Dual-Band, Linearly Polarized, Electrically Small Huygens Dipole Antennas', IEEE Transactions on Antennas and Propagation, vol. 67, no. 1, pp. 37-47.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. Two electrically small, dual-band Huygens dipole antennas are reported. In both designs, two pairs of magnetic and electric near-field resonant parasitic (NFRP) elements are combined organically within an electrically small, low profile package. The NFRP elements are excited effectively using only one coaxial-fed-driven element. One dual-band Huygens system produces parallel, linearly polarized (LP) fields at its two operating frequencies. The other dual-band system produces two orthogonal LP fields. Additional parasitic elements are introduced to mitigate the mutual coupling effects between the pairs of NFRP elements. The measured values for prototypes of both antennas in the L-band demonstrate their electrically small size (ka< 1) and low profile (∼ 0.03λ0). They also confirm their broadside radiation and polarization performance characteristics, as well as the isolation between each operating frequency. Their fractional bandwidths, peak realized gains, front-to-back ratios, and radiation efficiencies are, respectively, 0.6%, > 2 dBi, > 10 dB, and > 60% at both frequencies. These dual-band systems would provide multifunctional performance in a variety of portable, compact wireless devices.
Tian, J, Li, R, Yoo, SH, Poumellec, B, Garcia-Caurel, E, Ossikovski, R, Stchakovsky, M, Eypert, C, Canning, J & Lancry, M 2019, 'Spectral dependence of femtosecond laser induced circular optical properties in silica', OSA Continuum, vol. 2, no. 4, pp. 1233-1242.
View/Download from: Publisher's site
View description>>
© 2019 Optical Society of America. Transmission Mueller-matrix spectroscopic ellipsometry is applied to study femtosecond laser induced nanogratings in silica glass in a wide spectral range 250-1800 nm. By using differential decomposition of the Mueller matrix, the circular birefringence and dichroism of femtosecond laser irradiated SiO2 are quantified for the first time in the UV and near-IR range. A maximum value of the effective specific rotation of α ~ -860°/mm at 290 nm is found. In the near-IR range, we found a linear and circular dichroism band peaking around 1240 nm, which might be attributed to the formation of anisotropic species like the formation of oriented OH species and Si-O-Si bond.
Tofigh, F, Amiri, M, Shariati, N, Lipman, J & Abolhasan, M 2019, 'Low-Frequency Metamaterial Absorber Using Space-Filling Curve', Journal of Electronic Materials, vol. 48, pp. 6451-6459.
View/Download from: Publisher's site
View description>>
The extensive use of metamaterials and metamaterial absorbers increases the demand for compact structures in various frequencies. Designing electrically small absorbers for lower frequencies, especially sub-gigahertz applications, is one of the open issues in this field. In this paper, a space filling curve is used to design an absorber operating on low frequencies. The unit cell design is based on a Sierpinski curve with the size of 25×25×1.6mm3 and air-gap of 10 mm. The structure shows 99.9% absorption at 900 MHz on the third step. The system also shows multiple resonances due to its structure. The proposed structure is fabricated and tested and shows a good agreement with simulation results.
Unanue, IJ, Borzeshi, EZ, Esmaili, N & Piccardi, M 2019, 'ReWE: Regressing word embeddings for regularization of neural machine translation systems', NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference, vol. 1, pp. 430-436.
View description>>
© 2019 Association for Computational Linguistics Regularization of neural machine translation is still a significant problem, especially in low-resource settings. To mollify this problem, we propose regressing word embeddings (ReWE) as a new regularization technique in a system that is jointly trained to predict the next word in the translation (categorical value) and its word embedding (continuous value). Such a joint training allows the proposed system to learn the distributional properties represented by the word embeddings, empirically improving the generalization to unseen sentences. Experiments over three translation datasets have showed a consistent improvement over a strong baseline, ranging between 0.91 and 2.54 BLEU points, and also a marked improvement over a state-of-the-art system.
Usman, M, He, X, Lam, KKM, Xu, M, Chen, J, Bokhari, SMM & Jan, MA 2019, 'Error Concealment for Cloud-based and Scalable Video Coding of HD Videos', IEEE Transactions on Cloud Computing, vol. 7, no. 4, pp. 975-987.
View/Download from: Publisher's site
View description>>
IEEE The encoding of HD videos faces two challenges: requirements for a strong processing power and a large storage space. One time-efficient solution addressing these challenges is to use a cloud platform and to use a scalable video coding technique to generate multiple video streams with varying bit-rates. Packet-loss is very common during the transmission of these video streams over the Internet and becomes another challenge. One solution to address this challenge is to retransmit lost video packets, but this will create end-to-end delay. Therefore, it would be good if the problem of packet-loss can be dealt with at the user's side. In this paper, we present a novel system that encodes and stores the videos using the Amazon cloud computing platform, and recover lost video frames on user side using a new Error Concealment (EC) technique. To efficiently utilize the computation power of a user's mobile device, the EC is performed based on a multiple-thread and parallel process. The simulation results clearly show that, on average, our proposed EC technique outperforms the traditional Block Matching Algorithm (BMA) and the Frame Copy (FC) techniques.
Usman, M, Jan, MA, He, X & Chen, J 2019, 'A survey on big multimedia data processing and management in smart cities', ACM Computing Surveys, vol. 52, no. 3.
View/Download from: Publisher's site
View description>>
© 2019 Association for Computing Machinery. All rights reserved. Integration of embedded multimedia devices with powerful computing platforms, e.g., machine learning platforms, helps to build smart cities and transforms the concept of Internet of Things into Internet of Multimedia Things (IoMT). To provide different services to the residents of smart cities, the IoMT technology generates big multimedia data. The management of big multimedia data is a challenging task for IoMT technology. Without proper management, it is hard to maintain consistency, reusability, and reconcilability of generated big multimedia data in smart cities. Various machine learning techniques can be used for automatic classification of raw multimedia data and to allow machines to learn features and perform specific tasks. In this survey, we focus on various machine learning platforms that can be used to process and manage big multimedia data generated by different applications in smart cities. We also highlight various limitations and research challenges that need to be considered when processing big multimedia data in real-time.
Usman, M, Jan, MA, He, X & Chen, J 2019, 'A survey on representation learning efforts in cybersecurity domain', ACM Computing Surveys, vol. 52, no. 6.
View/Download from: Publisher's site
View description>>
© 2019 Association for Computing Machinery. In this technology-based era, network-based systems are facing new cyber-attacks on daily bases. Traditional cybersecurity approaches are based on old threat-knowledge databases and need to be updated on a daily basis to stand against new generation of cyber-threats and protect underlying network-based systems. Along with updating threat-knowledge databases, there is a need for proper management and processing of data generated by sensitive real-time applications. In recent years, various computing platforms based on representation learning algorithms have emerged as a useful resource to manage and exploit the generated data to extract meaningful information. If these platforms are properly utilized, then strong cybersecurity systems can be developed to protect the underlying network-based systems and support sensitive real-time applications. In this survey, we highlight various cyber-threats, real-life examples, and initiatives taken by various international organizations. We discuss various computing platforms based on representation learning algorithms to process and analyze the generated data. We highlight various popular datasets introduced by well-known global organizations that can be used to train the representation learning algorithms to predict and detect threats. We also provide an in-depth analysis of research efforts based on representation learning algorithms made in recent years to protect the underlying network-based systems against current cyber-threats. Finally, we highlight various limitations and challenges in these efforts and available datasets that need to be considered when using them to build cybersecurity systems.
Usman, M, Jan, MA, He, X & Chen, J 2019, 'P2DCA: A Privacy-Preserving-Based Data Collection and Analysis Framework for IoMT Applications', IEEE Journal on Selected Areas in Communications, vol. 37, no. 6, pp. 1222-1230.
View/Download from: Publisher's site
View description>>
© 1983-2012 IEEE. The concept of Internet of Multimedia Things (IoMT) is becoming popular nowadays and can be used in various smart city applications, e.g., traffic management, healthcare, and surveillance. In the IoMT, the devices, e.g., Multimedia Sensor Nodes (MSNs), are capable of generating both multimedia and non-multimedia data. The generated data are forwarded to a cloud server via a Base Station (BS). However, it is possible that the Internet connection between the BS and the cloud server may be temporarily down. The limited computational resources restrict the MSNs from holding the captured data for a longer time. In this situation, mobile sinks can be utilized to collect data from MSNs and upload to the cloud server. However, this data collection may create privacy issues, such as revealing identities and location information of MSNs. Therefore, there is a need to preserve the privacy of MSNs during mobile data collection. In this paper, we propose an efficient privacy-preserving-based data collection and analysis (P2DCA) framework for IoMT applications. The proposed framework partitions an underlying wireless multimedia sensor network into multiple clusters. Each cluster is represented by a Cluster Head (CH). The CHs are responsible to protect the privacy of member MSNs through data and location coordinates aggregation. Later, the aggregated multimedia data are analyzed on the cloud server using a counter-propagation artificial neural network to extract meaningful information through segmentation. Experimental results show that the proposed framework outperforms the existing privacy-preserving schemes, and can be used to collect multimedia data in various IoMT applications.
Wang, J, Yeh, WC, Xiong, NN, Wang, J, He, X & Huang, CL 2019, 'Building an Improved Internet of Things Smart Sensor Network Based on a Three-Phase Methodology', IEEE Access, vol. 7, pp. 141728-141737.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. In recent years, the Internet of Things (IoT) has allowed the easy, intelligent, and efficient connection of many devices used in daily life by means of numerous smart sensors which communicate with each other using wireless signals. The rapid development of the IoT has been a result of recent advances in sensing technology. This paper proposes a three-phase methodology to improve the quality of experience for IoT system technologies. The proposed method employs the concepts of simple routing and two well-known multi-criteria decision-making method (MCDM) techniques: The Analytic Hierarchy Process (AHP) and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). First, all simple routings are obtained using the proposed depth-first search technology (DFS). AHP is applied to analyze the structure of the problem and to obtain weights for various selected criteria in the second phase. In the third phase, TOPSIS is utilized to rank the simple routings, which are simple paths. A case study example is provided to demonstrate the proposed three-phase methodology. The results from the numerical experiments show that the proposed methodology can successfully achieve the aim of this paper.
Wang, N, Gao, C, Ding, C, Jia, H-Z, Sui, G-R & Gao, X-M 2019, 'A Thermal Management System to Reuse Thermal Waste Released by High-Power Light-Emitting Diodes', IEEE TRANSACTIONS ON ELECTRON DEVICES, vol. 66, no. 11, pp. 4790-4797.
View/Download from: Publisher's site
Wang, S, Mao, G & Zhang, JA 2019, 'Joint Time-of-Arrival Estimation for Coherent UWB Ranging in Multipath Environment with Multi-User Interference', IEEE Transactions on Signal Processing, vol. 67, no. 14, pp. 3743-3755.
View/Download from: Publisher's site
View description>>
© 1991-2012 IEEE. Time-of-Arrival (ToA) estimation becomes extremely challenging for ultra-wideband ranging systems in dense multipath environment with multiuser interference. In this paper, we propose a high accuracy joint ToA estimation (JToAE) algorithm, which can provide dominantly better performance than existing techniques. Based on the requirement of time synchronization among base stations, our proposed JToAE algorithm jointly exploits spatial information of each base station and the ToA of each multipath component of each received signal in ToA estimation. Our scheme is insensitive to the selection of the threshold, and does not require any additional information such as channel, noise power, preamble, or synchronization between transmitters and receivers. We also propose how to effectively distinguish the ToA of the first path of the desired user from the interfering signals in multi-user case without generating error propagation. The proposed JToAE is verified by extensive Monto Carlo simulation that is based on IEEE 802.15.4a channel models, and the simulation results indicate that even in low signal-to-noise ratio and multi-user case, our proposed technique can achieve significantly higher ranging accuracy compared to those in the literature in recent decades.
Wang, TQ, Li, H & Huang, X 2019, 'Analysis and Mitigation of Clipping Noise in Layered ACO-OFDM Based Visible Light Communication Systems', IEEE Transactions on Communications, vol. 67, no. 1, pp. 564-577.
View/Download from: Publisher's site
View description>>
IEEE Due to the limited dynamic range of the off-theshelf electrical and optical components, deliberate digital clipping (DDC) is widely applied to optical orthogonal frequency division multiplexing (OFDM) based visible light communication systems. In this paper, we present a theoretical characterization of the layered asymmetrically clipped optical OFDM (ACO-OFDM) signals subject to peak clipping. We decouple a clipped L-layer ACO-OFDM symbol to L single-layer ACO-OFDM symbols, each corresponding to a layer, and show that these symbols are subject to symmetrical peak clippings at random levels. Using the Bussgang’s theorem, the resulting attenuation factors and variances of the additive noise associated with each layer are derived. It is shown that the clipping noise caused by the DDC mainly falls onto the first layer, and its impact is gradually reduced in the subsequent layers. In order to combat the clipping noise, a novel receiver based on decision aided reconstruction is proposed. Simulation results show that the proposed receiver can effectively mitigate the clipping noise, leading to significant improvement of bit error rates over the conventional receiver.
Wang, W, Hoang, DT, Hu, P, Xiong, Z, Niyato, D, Wang, P, Wen, Y & Kim, DI 2019, 'A Survey on Consensus Mechanisms and Mining Strategy Management in Blockchain Networks', IEEE Access, vol. 7, pp. 22328-22370.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. The past decade has witnessed the rapid evolution in blockchain technologies, which has attracted tremendous interests from both the research communities and industries. The blockchain network was originated from the Internet financial sector as a decentralized, immutable ledger system for transactional data ordering. Nowadays, it is envisioned as a powerful backbone/framework for decentralized data processing and data-driven self-organization in flat, open-access networks. In particular, the plausible characteristics of decentralization, immutability, and self-organization are primarily owing to the unique decentralized consensus mechanisms introduced by blockchain networks. This survey is motivated by the lack of a comprehensive literature review on the development of decentralized consensus mechanisms in blockchain networks. In this paper, we provide a systematic vision of the organization of blockchain networks. By emphasizing the unique characteristics of decentralized consensus in blockchain networks, our in-depth review of the state-of-the-art consensus protocols is focused on both the perspective of distributed consensus system design and the perspective of incentive mechanism design. From a game-theoretic point of view, we also provide a thorough review of the strategy adopted for self-organization by the individual nodes in the blockchain backbone networks. Consequently, we provide a comprehensive survey of the emerging applications of blockchain networks in a broad area of telecommunication. We highlight our special interest in how the consensus mechanisms impact these applications. Finally, we discuss several open issues in the protocol design for blockchain consensus and the related potential research directions.
Wang, X, Song, B, Ni, W, Liu, RP, Guo, YJ, Niu, X & Zheng, K 2019, 'Group-based susceptible-infectious-susceptible model in large-scale directed networks', Security and Communication Networks, vol. 2019.
View/Download from: Publisher's site
View description>>
© 2019 Xu Wang et al. Epidemic models trade the modeling accuracy for complexity reduction. This paper proposes to group vertices in directed graphs based on connectivity and carries out epidemic spread analysis on the group basis, thereby substantially reducing the modeling complexity while preserving the modeling accuracy. A group-based continuous-time Markov SIS model is developed. The adjacency matrix of the network is also collapsed according to the grouping, to evaluate the Jacobian matrix of the group-based continuous-time Markov model. By adopting the mean-field approximation on the groups of nodes and links, the model complexity is significantly reduced as compared with previous topological epidemic models. An epidemic threshold is deduced based on the spectral radius of the collapsed adjacency matrix. The epidemic threshold is proved to be dependent on network structure and interdependent of the network scale. Simulation results validate the analytical epidemic threshold and confirm the asymptotical accuracy of the proposed epidemic model.
Wang, X, Yu, G, Zha, X, Ni, W, Liu, RP, Guo, YJ, Zheng, K & Niu, X 2019, 'Capacity of blockchain based Internet-of-Things: Testbed and analysis', Internet of Things, vol. 8.
View/Download from: Publisher's site
Wang, X, Zha, X, Ni, W, Liu, RP, Guo, YJ, Niu, X & Zheng, K 2019, 'Survey on blockchain for Internet of Things', Computer Communications, vol. 136, pp. 10-29.
View/Download from: Publisher's site
View description>>
© 2019 Elsevier B.V. The Internet of Things (IoT) is poised to transform human life and unleash enormous economic benefit. However, inadequate data security and trust of current IoT are seriously limiting its adoption. Blockchain, a distributed and tamper-resistant ledger, maintains consistent records of data at different locations, and has the potential to address the data security concern in IoT networks. While providing data security to the IoT, Blockchain also encounters a number of critical challenges inherent in the IoT, such as a huge number of IoT devices, non-homogeneous network structure, limited computing power, low communication bandwidth, and error-prone radio links. This paper presents a comprehensive survey on existing Blockchain technologies with an emphasis on the IoT applications. The Blockchain technologies which can potentially address the critical challenges arising from the IoT and hence suit the IoT applications are identified with potential adaptations and enhancements elaborated on the Blockchain consensus protocols and data structures. Future research directions are collated for effective integration of Blockchain into the IoT networks.
Wang, Y, Shuai, Y, Zhu, Y, Zhang, J & An, P 2019, 'Jointly learning perceptually heterogeneous features for blind 3D video quality assessment', Neurocomputing, vol. 332, pp. 298-304.
View/Download from: Publisher's site
View description>>
© 2018 Elsevier B.V. 3D videos quality assessment (3D-VQA) is essential to various 3D video processing applications. However, it has not been well investigated on how to make use of perceptual multi-channel video information to improve 3D-VQA under different distortion categories and degrees, especially under asymmetrical distortions. In the paper, we propose a new blind 3D-VQA metric by jointly learning perceptually heterogeneous features. Firstly, a binocular spatio-temporal internal generative mechanism (BST-IGM) is proposed to decompose the views of 3D video into multi-channel videos. Then, we extract perceptually heterogeneous features by proposed multi-channel natural video statistics (MNVS) model, which are characterized 3D video information. Furthermore, a robust AdaBoosting Radial Basis Function (RBF) neural network is utilized to map the features to the overall quality of 3D video. On two benchmark databases, the extensive evaluations demonstrate that the proposed algorithm significantly outperforms several state-of-the-art quality metrics in term of prediction accuracy and robustness.
Wang, Z, Xu, M, Ye, N, Wang, R & Huang, H 2019, 'RF-Focus: Computer Vision-assisted Region-of-interest RFID Tag Recognition and Localization in Multipath-prevalent Environments', Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 3, no. 1, pp. 1-30.
View/Download from: Publisher's site
Wei, F, Yang, ZJ, Qin, PY, Guo, YJ, Li, B & Shi, XW 2019, 'A Balanced-to-Balanced In-Phase Filtering Power Divider with High Selectivity and Isolation', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 2, pp. 683-694.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. In this paper, a balanced-to-balanced in-phase filtering power divider (FPD) is proposed, which can realize a two-way equal power division with high selectivity and isolation. The design of the proposed FPD is primarily based on microstrip/slotline transition structures and slotline T-junction. A U-type microstrip feed line integrated with a stepped-impedance slotline resonator is adopted at the input and output ports, which makes the differential-mode (DM) responses independent of the common-mode (CM) ones. Meanwhile, superior DM transmission and CM suppression are achieved intrinsically, thereby simplifying the design procedure significantly. By employing slotline resonators loaded with resistors, the isolation between the two output ports can be improved greatly. In addition, a DM passband with a sharp filtering performance is realized by introducing the microstrip stub-loaded resonators (SLRs). By changing the electrical length of the open stub of the SLR, the fractional bandwidth is controllable. In order to verify the feasibility of the proposed design method, two prototype circuits of the proposed FPDs with different bandwidths are fabricated and measured. Good agreement between the simulation and measurement results is observed.
Wen, Z, Tang, MC & Ziolkowski, RW 2019, 'Band- And frequency-reconfigurable circularly polarised filtenna for cognitive radio applications', IET Microwaves, Antennas and Propagation, vol. 13, no. 7, pp. 1003-1008.
View/Download from: Publisher's site
View description>>
© The Institution of Engineering and Technology 2019. A band- and frequency-reconfigurable circularly polarised (CP) filtenna that exhibits good selectivity and out-of-band rejection in both its wideband and continuously tunable narrowband states is reported for cognitive radio (CR) applications. The filtenna provides two switchable operational states: A wideband state from 2.52 to 3.53 GHz for sensing and a tunable state of narrowband states from 2.44 to 3.19 GHz for communications. The corresponding 3-dB axial ratio bandwidths completely overlap these impedance bandwidths. A single PIN diode is utilised to switch between these two operational states to achieve the band-reconfigurable performance. Two varactor diodes are employed to realise the frequency-reconfigurable performance by facilitating a continuous shift of the narrowband states over the operational communication frequencies.
Wu, K, Ni, W, Su, T, Liu, RP & Guo, Y 2019, 'Exploiting Spatial-Wideband Effect for Fast AoA Estimation at Lens Antenna Array', IEEE Journal on Selected Topics in Signal Processing, vol. 13, no. 5, pp. 902-917.
View/Download from: Publisher's site
View description>>
© 2007-2012 IEEE. Energy-efficient, highly integrated lens antenna arrays (LAAs) have found widespread applications in wideband millimeter wave or terahertz communications, localization and tracking, and wireless power transfer. Accurate estimation of angle-of-arrival (AoA) is key to those applications, but has been hindered by a spatial-wideband effect in wideband systems. This paper proposes to exploit (rather than circumventing) the spatial-wideband effect to develop a fast and accurate AoA estimation approach for LAAs. Specifically, we unveil new spatial-frequency patterns based on the spatial-wideband effect, and establish one-to-one mappings between the patterns and the strongest discrete Fourier transform (DFT) beam containing the AoA. With the strongest DFT beam identified, we propose a method to estimate the AoA uniquely and accurately, using only a few training symbols. This is achieved by deriving a new one-to-one mapping between the AoA and the set of DFT beams judiciously selected based on the strongest. In the case that an impinging path is uniformly distributed in [0,2π], simulations show that the proposed algorithm is able to reduce the mean squared error of the AoA estimation by as much as 82.1% while reducing the number of required symbols by 93.2$, as compared to existing techniques. The algorithm can also increase the spectral efficiency by 89% when the average SNR is 20 dB at each antenna of the receiver.
Wu, K, Ni, W, Su, T, Liu, RP & Guo, YJ 2019, 'Efficient angle-of-arrival estimation of lens antenna arrays for wireless information and power transfer', IEEE Journal on Selected Areas in Communications, vol. 37, no. 1, pp. 116-130.
View/Download from: Publisher's site
View description>>
© 1983-2012 IEEE. Antenna design and angle-of-arrival (AoA) estimation are critical to the efficiency of wireless information and power transfer. The AoA estimation is challenging for energy-efficient lens antenna arrays (LAAs), due to discrete sets of fixed discrete Fourier transform (DFT) beams. This paper presents a novel fast and accurate approach for the AoA estimation of LAAs. The key idea is that we prove the two differential outputs of three adjacent lens beams, referred to as 'DFT beam differences (DBDs),' that are the strongest at the two sides of an AoA. They are easy to identify and robust to noises, and their powers are proved to provide an accurate estimate of the AoA. Another important aspect is a new beam synthesis technique which produces different beam widths based on DFT beams and practical 1-bit phase shifts in real time. As a result, the angular region containing the AoA can exponentially narrow down, and the two strongest DBDs can be quickly identified. The proposed approach can operate in coupling with successive interference cancellation to estimate the AoAs of multiple paths. Simulations show that the proposed approach is able to outperform the state of the art by orders of magnitude in terms of accuracy. The power transfer efficiency can be dramatically improved.
Wu, K, Ni, W, Su, T, Liu, RP & Guo, YJ 2019, 'Expeditious estimation of angle-of-arrival for hybrid butler matrix arrays', IEEE Transactions on Wireless Communications, vol. 18, no. 4, pp. 2170-2185.
View/Download from: Publisher's site
View description>>
© 2002-2012 IEEE. Arrays of Butler matrices provide a promising front-end design for massive MIMO transceivers with low cost and low complexity. However, this advanced design does not necessarily translate to effective applications, unless the angle-of-arrival (AoA) of signals avails to the Butler matrices. This paper presents an efficient approach to the unprecedented AoA estimation for the arrays of Butler matrices. Specifically, we design a new beam synthesis method to recursively narrow down and increasingly focus on the angular region of interest, and hence achieving robust estimation of the phase offset between Butler matrices. With the phase offset canceled in the received signals, we are able to identify the set of critical Butler beams with the dominating effect on the AoA estimation, and estimate the AoA accordingly with minimum signaling. The mean squared error of the proposed estimation is analyzed in the presence of non-negligible noises, with closed-form lower bounds derived. Validated by simulations, the proposed algorithm is able to indistinguishably approach the lower bounds, and significantly outperforms the state-of-the-art developed for discrete antenna arrays by orders of magnitude in terms of accuracy, especially in low signal-to-noise regimes.
Wu, K, Ni, W, Su, T, Liu, RP & Guo, YJ 2019, 'Recent Breakthroughs on Angle-of-Arrival Estimation for Millimeter-Wave High-Speed Railway Communication', IEEE Communications Magazine, vol. 57, no. 9, pp. 57-63.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. With significantly improved efficiency, largescale hybrid antenna arrays with tens to hundreds of antennas have great potential to support millimeter-wave (mmWave) communication for high-speed railway (HSR) applications. The significant beamforming gains rely on fast and accurate estimation of the angle-of-arrival (AoA), but this can be impeded by the high train speed, the cost/energy oriented design of arrays, and the severe attenuation of mmWave signals. This article reviews these challenges, and discusses the limitations of existing AoA estimation techniques under hybrid antenna array settings. The article further reveals a few recent theoretical breakthroughs that can potentially enable fast and reliable estimation, even based on severely attenuated signals. Under a speed setting of 500 km/h, a performance study is carried out to confirm the significant improvements of estimation accuracy and subsequent beamforming gains as the results of the breakthroughs.
Wu, YM, Wong, SW, Lin, JY, Yang, Y, Zhang, L, Choi, WW, Zhu, L & He, Y 2019, 'Design of triple-band and triplex slot antenna using triple-mode cavity resonator', IET Microwaves, Antennas and Propagation, vol. 13, no. 13, pp. 2303-2309.
View/Download from: Publisher's site
View description>>
© 2019 The Institution of Engineering and Technology. A class of triple-band and triplex cavity-backed slot antenna is proposed by using three fundamental modes in a single metal cavity: TE011, TE101, and TM110 modes, simultaneously. These three resonant modes can be excited by changing the position of the feeding slot without any extra components configured inside the cavity. By opening a single radiation slot, a single band triple-mode cavity-backed slot antenna is achieved. By opening three slots at three side walls of the cavity, a triple-band cavity-backed slot antenna is realised with three different radiation directions at three different operation frequencies. Moreover, a triplex triple-band antenna can be formed by combining a single slot antenna and three input ports in the proposed rectangular cavity. Finally, a triple-band multi-directional cavity-backed slot antenna prototype and a triplex triple-band cavity-backed single slot antenna prototype are fabricated and tested. The tested results are in good agreement with the simulated results, which indicates the feasibility of the proposed design methodology.
Xiao, C, Zeng, J, Ni, W, Liu, RP, Su, X & Wang, J 2019, 'Delay Guarantee and Effective Capacity of Downlink NOMA Fading Channels', IEEE Journal on Selected Topics in Signal Processing, vol. 13, no. 3, pp. 508-523.
View/Download from: Publisher's site
View description>>
© 2007-2012 IEEE. Nonorthogonal multiple access (NOMA) is promising for increasing connectivity and capacity. But there has been little consideration on the quality of service of NOMA; let alone that in generic fading channels. This paper establishes closed-form upper bounds for the delay violation probability of downlink Nakagami-m and Rician NOMA channels, by exploiting stochastic network calculus (SNC). The key challenge addressed is to derive the Mellin transforms of the service processes in the NOMA fading channels. The transforms are proved to be stable, and incorporated into the SNC to provide the closed-form upper bounds of the delay violation probability. The paper also applies the Mellin transforms to develop the closed-form expressions for the effective capacity of the NOMA fading channels, which measures the channel capacity under statistical delay guarantees. By further applying the min-max and max-min rules, two new power allocation algorithms are proposed to optimize the closed-form expressions, which can provide the NOMA users fairness in terms of delay violation probability and effective capacity. Simulation results substantiate the derived upper bounds of the delay violation probabilities, and the effective capacity. The proposed power allocation algorithms are also numerically validated.
Xiao, C, Zeng, J, Ni, W, Su, X, Liu, RP, Lv, T & Wang, J 2019, 'Downlink MIMO-NOMA for Ultra-Reliable Low-Latency Communications', IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, vol. 37, no. 4, pp. 780-794.
View/Download from: Publisher's site
Xu, JX, Li, HY, Zhang, XY, Yang, Y, Xue, Q & Dutkiewicz, E 2019, 'Compact Dual-Channel Balanced Filter and Balun Filter Based on Quad-Mode Dielectric Resonator', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 2, pp. 494-504.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. In this paper, we propose a method for designing a dual-channel balanced filter and a dual-channel balun filter based on the quad-mode dielectric resonator (DR) for the first time. By sharing one common quad-mode DR, two balanced filters or two balun filters are integrated as one single-cavity configuration, featuring compact size and high integration. A cylindrical DR with two short ends is investigated to construct the quad-mode DR. By properly arranging input and output feeding probes, two modes of the DR are only excited by the feeding probes of one channel and the other two modes are excited by that of the other channel. Accordingly, signals cannot be transmitted between the two channels, resulting in high isolation. Moreover, the required out-of-phase characteristics of the balanced and balun filters can be obtained by the inherent electromagnetic field properties of the DR without adding additional circuits, featuring a simple structure. For demonstration, a dual-channel balanced filter and a dual-channel balun filter are designed and fabricated, showing excellent balanced or balun filter performance of each channel as well as high isolation between the two channels. As compared to the other reported DR balanced and balun filters, the proposed designs exhibit a significant size reduction, which are attractive in wireless systems.
Xu, J-X, Yang, L, Yang, Y & Zhang, XY 2019, 'High-< inline-formula > < tex-math notation="LaTeX">$Q$ -Factor Tunable Bandpass Filter With Constant Absolute Bandwidth and Wide Tuning Range Based on Coaxial Resonators', IEEE TRANSACTIONS ON MICROWAVE THEORY AND TECHNIQUES, vol. 67, no. 10, pp. 4186-4195.
View/Download from: Publisher's site
Yang, D, Zou, YX, Zhang, J & Li, G 2019, 'C-RPNs: Promoting object detection in real world via a cascade structure of Region Proposal Networks', Neurocomputing, vol. 367, pp. 20-30.
View/Download from: Publisher's site
View description>>
© 2019 Elsevier B.V. Recently, significant progresses have been made in object detection on common benchmarks (i.e., Pascal VOC). However, object detection in real world is still challenging due to the serious data imbalance. Images in real world are dominated by easy samples like the wide range of background and some easily recognizable objects, for example. Although two-stage detectors like Faster R-CNN achieved big successes in object detection due to the strategy of extracting region proposals by Region Proposal Network, they show their poor adaption in real-world object detection as a result of without considering mining hard samples during extracting region proposals. To address this issue, we propose a Cascade framework of Region Proposal Networks, referred to as C-RPNs, which adopts multiple stages to mine hard samples while extracting region proposals and learn stronger classifiers. Meanwhile, a feature chain and a score chain are proposed to help learning more discriminative representations for proposals. Moreover, a loss function of cascade stages is designed to train cascade classifiers through backpropagation. Our proposed method has been evaluated on Pascal VOC and several challenging datasets like BSBDV 2017, CityPersons, etc. Our method achieves competitive results compared with the current state-of-the-arts and attains all-sided improvements in error analysis, validating its efficacy for detection in real world.
Yang, T, Ding, C & Guo, Y 2019, 'A Highly Birefringent and Nonlinear AsSe2-As2S5 Photonic Crystal Fiber with Two Zero-Dispersion Wavelengths', IEEE Photonics Journal, vol. 11, no. 1.
View/Download from: Publisher's site
View description>>
OAPA A hybrid AsSe$_{2}$-As$_{2}$S$_{5}$ photonic crystal fiber (PCF) with a solid elliptical core is proposed and studied theoretically by full-vector finite element method (FEM). The core and cladding of the PCF are made of AsSe$_{2}$ and As$_{2}$S$_{5}$ glasses, respectively. Simulation results demonstrates that, at the operating wavelength of 1.55 μm, the proposed PCF not only exhibits a very high birefringence of 0.091 but also has large nonlinear coefficients of 147.8 and 78.2 W$^{-1}$m$^{-1}$ for the X- and Y-polarized (X- and Y-pol) modes, respectively. Moreover, it is able to achieve two zero-dispersion wavelengths (ZDWs) for both the X-pol (1.52 and 2.19 μm) and Y-pol (1.43 and 2.12 μm) modes. The proposed hybrid PCF exhibits excellent polarization maintaining and nonlinearity performance, thus suitable to be used in supercontinuum spectrum generation and polarization maintaining nonlinear signal processing.
Yang, T, Ding, C, Ziolkowski, RW & Guo, YJ 2019, 'A Terahertz (THz) Single-Polarization-Single-Mode (SPSM) Photonic Crystal Fiber (PCF).', Materials, vol. 12, no. 15.
View/Download from: Publisher's site
View description>>
This paper presents a novel approach to attain a single-polarization-single-mode (SPSM) photonic crystal fiber (PCF) in the terahertz (THz) regime. An initial circular hole PCF design is modified by introducing asymmetry in the first ring of six air holes in the cladding, i.e., epsilon-near-zero (ENZ) material is introduced into only four of those air holes and the other two remain air-filled but have different diameters. The resulting fundamental X-polarized (XP) and Y-polarized (YP) modes have distinctly different electric field distributions. The asymmetry is arranged so that the YP mode has a much larger amount of the field distributed in the ENZ material than the XP mode. Since the ENZ material is very lossy, the YP mode suffers a much higher loss than the XP mode. Consequently, after a short propagation distance, the loss difference (LD) between the XP and YP modes will be large enough that only the XP mode still realistically exists in the PCF. To further enhance the outcome, gain material is introduced into the core area to increase the LDs between the wanted XP mode and any unwanted higher order (HO) modes, as well as to compensate for the XP mode loss without affecting the LD between the XP and YP modes. The optimized PCF exhibits LDs between the desired XP mode and all other modes greater than 8.0 dB/cm across a wide frequency range of 0.312 THz. Consequently, the reported PCF only needs a length of 2.5 cm to attain an SPSM result, with the unwanted modes being more than 20 dB smaller than the wanted mode over the entire operational band.
Yang, Y, Liu, Y, Ma, X, Li, M, Xu, KD & Guo, YJ 2019, 'Synthesizing Unequally Spaced Pattern-Reconfigurable Linear Arrays with Minimum Interspacing Control', IEEE Access, vol. 7, pp. 58893-58900.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. Previously, the alternating convex optimization (ACO) was used to reduce the number of elements in the single-pattern linear array. This work extends the ACO method to synthesize the unequally spaced sparse linear arrays with reconfigurable multiple patterns. In this extended ACO, the minimum interspacing constraint can be easily incorporated in the sparse array synthesis by performing a set of constrained alternating convex optimizations. Three examples for synthesizing sparse linear array with different multiple-pattern requirements are conducted to validate the effectiveness, robustness, and advantages of the proposed method. The synthesis results show that the proposed method can effectively reduce the number of elements in the reconfigurable multiple-pattern linear arrays with good control of the sidelobe levels and minimum interspacing. The comparisons with other methods are also given in the examples.
Yao, X, Wu, Q, Zhang, P & Bao, F 2019, 'Adaptive rational fractal interpolation function for image super-resolution via local fractal analysis', Image and Vision Computing, vol. 82, pp. 39-49.
View/Download from: Publisher's site
View description>>
© 2019 Elsevier B.V. Image super-resolution aims to generate high-resolution image based on the given low-resolution image and to recover the details of the image. The common approaches include reconstruction-based methods and interpolation-based methods. However, these existing methods show difficulty in processing the regions of an image with complicated texture. To tackle such problems, fractal geometry is applied on image super-resolution, which demonstrates its advantages when describing the complicated details in an image. The common fractal-based method regards the whole image as a single fractal set. That is, it does not distinguish the complexity difference of texture across all regions of an image regardless of smooth regions or texture rich regions. Due to such strong presumption, it causes artificial errors while recovering smooth area and texture blurring at the regions with rich texture. In this paper, the proposed method produces rational fractal interpolation model with various setting at different regions to adapt to the local texture complexity. In order to facilitate such mechanism, the proposed method is able to segment the image region according to its complexity which is determined by its local fractal dimension. Thus, the image super-resolution process is cast to an optimization problem where local fractal dimension in each region is further optimized until the optimization convergence is reached. During the optimization (i.e. super-resolution), the overall image complexity (determined by local fractal dimension) is maintained. Compared with state-of-the-art method, the proposed method shows promising performance according to qualitative evaluation and quantitative evaluation.
Yao, Y, Shen, F, Zhang, J, Liu, L, Tang, Z & Shao, L 2019, 'Extracting Multiple Visual Senses for Web Learning', IEEE Transactions on Multimedia, vol. 21, no. 1, pp. 184-196.
View/Download from: Publisher's site
View description>>
© 1999-2012 IEEE. Labeled image datasets have played a critical role in high-level image understanding. However, the process of manual labeling is both time consuming and labor intensive. To reduce the dependence on manually labeled data, there have been increasing research efforts on learning visual classifiers by directly exploiting web images. One issue that limits their performance is the problem of polysemy. Existing unsupervised approaches attempt to reduce the influence of visual polysemy by filtering out irrelevant images, but do not directly address polysemy. To this end, in this paper, we present a multimodal framework that solves the problem of polysemy by allowing sense-specific diversity in search results. Specifically, we first discover a list of possible semantic senses from untagged corpora to retrieve sense-specific images. Then, we merge visual similar semantic senses and prune noise by using the retrieved images. Finally, we train one visual classifier for each selected semantic sense and use the learned sense-specific classifiers to distinguish multiple visual senses. Extensive experiments on classifying images into sense-specific categories and reranking search results demonstrate the superiority of our proposed approach.
Yao, Y, Shen, F, Zhang, J, Liu, L, Tang, Z & Shao, L 2019, 'Extracting Privileged Information for Enhancing Classifier Learning', IEEE Transactions on Image Processing, vol. 28, no. 1, pp. 436-450.
View/Download from: Publisher's site
View description>>
© 1992-2012 IEEE. The accuracy of data-driven learning approaches is often unsatisfactory when the training data is inadequate either in quantity or quality. Manually labeled privileged information (PI), e.g., attributes, tags or properties, is usually incorporated to improve classifier learning. However, the process of manually labeling is time-consuming and labor-intensive. Moreover, due to the limitations of personal knowledge, manually labeled PI may not be rich enough. To address these issues, we propose to enhance classifier learning by exploring PI from untagged corpora, which can effectively eliminate the dependency on manually labeled data and obtain much richer PI. In detail, we treat each selected PI as a subcategory and learn one classifier for each subcategory independently. The classifiers for all subcategories are integrated together to form a more powerful category classifier. Particularly, we propose a novel instance-level multi-instance learning model to simultaneously select a subset of training images from each subcategory and learn the optimal SVM classifiers based on the selected images. Extensive experiments on four benchmark data sets demonstrate the superiority of our proposed approach.
Ye, J, Yang, X, Xu, M, Chan, PK-S & Ma, C 2019, 'Novel N-Substituted oseltamivir derivatives as potent influenza neuraminidase inhibitors: Design, synthesis, biological evaluation, ADME prediction and molecular docking studies', EUROPEAN JOURNAL OF MEDICINAL CHEMISTRY, vol. 182.
View/Download from: Publisher's site
Yuan, C, Tao, X, Li, N, Ni, W, Liu, RP & Zhang, P 2019, 'Analysis on Secrecy Capacity of Cooperative Non-Orthogonal Multiple Access With Proactive Jamming', IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, vol. 68, no. 3, pp. 2682-2696.
View/Download from: Publisher's site
Yuan, W, Wu, N, Guo, Q, Huang, X, Li, Y & Hanzo, L 2019, 'TOA-based passive localization constructed over factor graphs: A unified framework', IEEE Transactions on Communications, vol. 67, no. 10, pp. 6952-6962.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Passive localization based on time of arrival (TOA) measurements is investigated, where the transmitted signal is reflected by a passive target and then received at several distributed receivers. After collecting all measurements at receivers, we can determine the target location. The aim of this paper is to provide a unified factor graph-based framework for passive localization in wireless sensor networks based on TOA measurements. Relying on the linearization of range measurements, we construct a Forney-style factor graph model and conceive the corresponding Gaussian message passing algorithm to obtain the target location. It is shown that the factor graph can be readily modified for handling challenging scenarios such as uncertain receiver positions and link failures. Moreover, a distributed localization method based on consensus-aided operation is proposed for a large-scale resource constrained network operating without a fusion center. Furthermore, we derive the Cramér-Rao bound (CRB) to evaluate the performance of the proposed algorithm. Our simulation results verify the efficiency of the proposed unified approach and of its distributed implementation.
Yuan, X, Feng, Z, Ni, W, Wei, Z, Liu, RP & Zhang, JA 2019, 'Secrecy Rate Analysis against Aerial Eavesdropper', IEEE Transactions on Communications, vol. 67, no. 10, pp. 7027-7042.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper studies the threat that an aerial eavesdropper can pose to terrestrial wireless communications, from an information-theoretic point of view. The achievable ergodic and the average ϵ-outage secrecy rates with no channel state information at the transmitter (i.e., with no CSIT) are analyzed for a transmitter-receiver pair on the ground, in the presence of an aerial eavesdropper which flies a random trajectory following a smooth turn (ST) mobility model in a three-dimensional (3D) space. The ST mobility model induces a uniform distribution (of the eavesdropper's waypoints) within the considered 3D volume. Closed-form asymptotic approximations of the achievable secrecy rates are derived based on the almost sure convergence and non-trivial mathematical manipulations. Validated by simulations, our analysis is tight and reveals that the ground transmission is particularly vulnerable to aerial eavesdropping which can be carried out in a distance without being noticed. 3D spherical regions are identified, within which the secrecy rates vanish. This sheds useful insights to protect terrestrial wireless networks from aerial eavesdropping.
Zeng, J, Lv, T, Liu, RP, Su, X, Beaulieu, NC & Guo, YJ 2019, 'Linear Minimum Error Probability Detection for Massive MU-MIMO with Imperfect CSI in URLLC', IEEE Transactions on Vehicular Technology, vol. 68, no. 11, pp. 11384-11388.
View/Download from: Publisher's site
View description>>
© 1967-2012 IEEE. It is challenging to realize ultra-reliable and low latency communications (URLLC) under severe shadow fading and imperfect channel state information (CSI). However, reliability can be increased by exploiting space diversity from multiple receive antennas rather than retransmission with limited latency. Massive multi-user multiple-input-multiple-output (MU-MIMO) is studied to enable URLLC with imperfect CSI from least-square channel estimation. The linear minimum error probability (MEP) detector with a given length of pilots (LoP) is derived. Further, the LoP is optimized to minimize the error probability of the uplink with a limited number of channel uses, using the finite blocklength information theory and one-dimensional search methods. Numerical results verify that the proposed linear MEP detection incorporated in massive MU-MIMO improves reliability with limited latency and imperfect CSI.
Zeng, J, Lv, T, Ni, W, Liu, RP, Beaulieu, N & Guo, YJ 2019, 'Ensuring Max-Min Fairness of UL SIMO-NOMA: A Rate Splitting Approach', IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, vol. 68, no. 11, pp. 11080-11093.
View/Download from: Publisher's site
Zhan, J, Ge, XJ, Huang, S, Zhao, L, Wong, JKW & He, SXJ 2019, 'Improvement of the inspection-repair process with building information modelling and image classification', Facilities, vol. 37, no. 7-8, pp. 395-414.
View/Download from: Publisher's site
View description>>
© 2019, Emerald Publishing Limited. Purpose: Automated technologies have been applied to facility management (FM) practices to address labour demands of, and time consumed by, inputting and processing manual data. Less attention has been focussed on automation of visual information, such as images, when improving timely maintenance decisions. This study aims to develop image classification algorithms to improve information flow in the inspection-repair process through building information modelling (BIM). Design/methodology/approach: To improve and automate the inspection-repair process, image classification algorithms were used to connect images with a corresponding image database in a BIM knowledge repository. Quick response (QR) code decoding and Bag of Words were chosen to classify images in the system. Graphical user interfaces (GUIs) were developed to facilitate activity collaboration and communication. A pilot case study in an inspection-repair process was applied to demonstrate the applications of this system. Findings: The system developed in this study associates the inspection-repair process with a digital three-dimensional (3D) model, GUIs, a BIM knowledge repository and image classification algorithms. By implementing the proposed application in a case study, the authors found that improvement of the inspection-repair process and automated image classification with a BIM knowledge repository (such as the one developed in this study) can enhance FM practices by increasing productivity and reducing time and costs associated with ecision-making. Originality/value: This study introduces an innovative approach that applies image classification and leverages a BIM knowledge repository to enhance the inspection-repair process in FM practice. The system designed provides automated image-classifying data from a smart phone, eliminates time required to input image data manually and improves communication and collaboration between FM personnel for maintenance i...
Zhang, A, Huang, X, Guo, YJ, Yuan, J & Heath, RW 2019, 'Multibeam for Joint Communication and Sensing Using Steerable Analog Antenna Arrays', IEEE Transactions on Vehicular Technology, vol. 68, no. 1, pp. 671-685.
View/Download from: Publisher's site
View description>>
IEEE Beamforming has great potential for joint communication and sensing (JCAS), which is becoming a demanding feature on many emerging platforms such as unmanned aerial vehicles and smart cars. Although beamforming has been extensively studied for communication and radar sensing respectively, its application in the joint system is not straightforward due to different beamforming requirements by communication and sensing. In this paper, we propose a novel multibeam framework using steerable analog antenna arrays, which allows seamless integration of communication and sensing. Different to conventional JCAS schemes that support JCAS using a single beam, our framework is based on the key innovation of multibeam technology: providing fixed subbeam for communication and packet-varying scanning subbeam for sensing, simultaneously from a single transmitting array. We provide a system architecture and protocols for the proposed framework, complying well with modern packet communication systems with multicarrier modulation. We also propose low-complexity and effective multibeam design and generation methods, which offer great flexibility in meeting different communication and sensing requirements. We further develop sensing parameter estimation algorithms using conventional digital Fourier transform and 1D compressive sensing techniques, matching well with the multibeam framework. Simulation results are provided and validate the effectiveness of our proposed framework, beamforming design methods and the sensing algorithms.
Zhang, J 2019, 'V2X-Communication Assisted Interference Minimization for Automotive Radars', China Communications, vol. 16, no. 10, pp. 100-111.
View description>>
With the development of automated driving vehicles, more and more vehicles will be fitted with more than one automotive radars, and the radar mutual interference will become very significant. Vehicle to everything (V2X) communication is a potential way for coordinating automotive radars and reduce the mutual interference. In this paper, we analyze the positional relation of the two radars that interfere with each other, and evaluate the mutual interference for different types of automotive radars based on Poisson point process (PPP). We also propose a centralized framework and the corresponding algorithm, which relies on V2X communication systems to allocate the spectrum resources for automotive radars to minimize the interference. The minimum spectrum resources required for zero-interference are analyzed for different cases. Simulation results validate the analysis and show that the proposed framework can achieve near-zero-interference with the minimum spectrum resources.
Zhang, J, Wu, Q, Zhang, J, Shen, C, Lu, J & Wu, Q 2019, 'Heritage image annotation via collective knowledge', PATTERN RECOGNITION, vol. 93, pp. 204-214.
View/Download from: Publisher's site
Zhang, R, Chengpo, M, Xu, M, Lixin, X & Xiaofeng, X 2019, 'Facial Component-Landmark Detection with Weakly-supervised LR-CNN', IEEE Access, vol. 7, pp. 10263-10277.
View/Download from: Publisher's site
Zhang, R, Mu, C, Xu, M, Xu, L, Shi, Q & Wang, J 2019, 'Synthetic IR image refinement using adversarial learning with bidirectional mappings', IEEE Access, vol. 7, pp. 153734-153750.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Collecting a large dataset of real infrared (IR) images is expensive, time-consuming, and even unavailable in some specific scenarios. With recent progress in machine learning, it has become more feasible to replace real IR images with qualified synthetic IR images in learning-based IR systems. However, this alternative may fail to achieve the desired performance, due to the gap between real and synthetic IR images. Inspired by adversarial learning for image-to-image translation, we propose the Synthetic IR Refinement Generative Adversarial Network (SIR-GAN) to narrow this gap. By learning the bidirectional mappings between two unpaired domains, the realism of the simulated IR images generated from the IR Simulator are significantly improved, where the source domain contains a large number of simulated IR images, where the target domain contains a limited quantity of real IR images. Specifically, driven by the idea of transferring infrared characteristic and protect target semantic information simultaneously, we propose a SIR refinement loss to consider an infrared loss and a structure loss further to the adversarial loss and the consistency loss. To further reduce the gap, stabilize training, and avoid artefacts, we modify the proposed algorithm by developing a training strategy, adding the U-net in the generators, using the dilated convolution in the discriminators and invoking the N-Adam acts as the optimizer. Qualitative, quantitative, and ablation study experiments demonstrate the superiority of the proposed approach compared with the state-of-the-art techniques in terms of realism and fidelity. In addition, our refined IR images are evaluated in the context of a feasibility study, where the accuracy of the trained classifier is significantly improved by adding our refined data into a small real-data training set.
Zhang, T, Bao, J-F, Zeng, R-Z, Yang, Y, Bao, L-L, Bao, F-H, Zhang, Y & Qin, F 2019, 'Long lifecycle MEMS double-clamped beam based on low stress graphene compound film', SENSORS AND ACTUATORS A-PHYSICAL, vol. 288, pp. 39-46.
View/Download from: Publisher's site
Zhang, X, Lv, T, Ren, Y, Ni, W, Beaulieu, NC & Guo, YJ 2019, 'Economical Caching for Scalable Videos in Cache-Enabled Heterogeneous Networks', IEEE Journal on Selected Areas in Communications, vol. 37, no. 7, pp. 1608-1621.
View/Download from: Publisher's site
View description>>
© 1983-2012 IEEE. We develop the optimal economical caching schemes in cache-enabled heterogeneous networks, while delivering multimedia video services with personalized viewing qualities to mobile users. By applying scalable video coding (SVC), each video file to be requested is divided into one base layer (BL) and several enhancement layers (ELs). In order to assign different transmission tasks, the serving small-cell base stations (SBSs) are grouped into K clusters. The SBSs are able to cache and cooperatively transmit BL and EL contents to the user. We analytically derive expressions for successful transmission probability and ergodic service rate, and then the closed-form expression for EConomical Efficiency (ECE) is obtained. In order to enhance the ECE performance, we formulate the ECE optimization problems for two cases. In the first case, with equal cache size equipped at each SBS, the layer caching indicator is determined. Since this problem is NP-hard, after the l-{0} -norm approximation, the discrete optimization variables are relaxed to be continuous, and this relaxed problem is convex. Next, based on the optimal solution derived from the relaxed problem, we devise a greedy-strategy based heuristic algorithm to achieve the near-optimal layer caching indicators. In the second case, the cache size for each SBS, the layer size, and the layer caching indicator are jointly optimized. This problem is a mixed integer programming problem, which is more challenging. To effectively solve this problem, the original ECE maximization problem is divided into two subproblems. These two subproblems are iteratively solved until the original optimization problem is convergent. Numerical results verify the correctness of the theoretical derivations. Additionally, compared to the most popular layer placement strategy, the performance superiority of the proposed SVC-based caching schemes is testified.
Zhang, X, Zhang, S, Lin, J, Sun, F, Zhu, X, Yang, Y, Tong, X & Yang, H 2019, 'An Efficient Seismic Data Acquisition Based on Compressed Sensing Architecture With Generative Adversarial Networks', IEEE ACCESS, vol. 7, pp. 105948-105961.
View/Download from: Publisher's site
Zhang, Z, Wu, Q, Wang, Y & Chen, F 2019, 'High-Quality Image Captioning with Fine-Grained and Semantic-Guided Visual Attention', IEEE Transactions on Multimedia, vol. 21, no. 7, pp. 1681-1693.
View/Download from: Publisher's site
View description>>
IEEE The soft-attention mechanism is regarded as one of the representative methods for image captioning. Based on the end-to-end Convolutional Neural Network (CNN)-Long Short Term Memory (LSTM) framework, the soft-attention mechanism attempts to link the semantic representation in text (i.e., captioning) with relevant visual information in the image for the first time. Motivated by this approach, several state-of-the-art attention methods are proposed. However, due to the constraints of CNN architecture, the given image is only segmented to the fixed-resolution grid at a coarse level. The visual feature extracted from each grid indiscriminately fuses all inside objects and/or their portions. There is no semantic link between grid cells. In addition, the large area "stuff" (e.g., the sky or a beach) cannot be represented using the current methods. To address these problems, this paper proposes a new model based on the Fully Convolutional Network (FCN)-LSTM framework, which can generate an attention map at a fine-grained grid-wise resolution. Moreover, the visual feature of each grid cell is contributed only by the principal object. By adopting the grid-wise labels (i.e., semantic segmentation), the visual representations of different grid cells are correlated to each other. With the ability to attend to large area "stuff", our method can further summarize an additional semantic context from semantic labels. This method can provide comprehensive context information to the language LSTM decoder. In this way, a mechanism of fine-grained and semantic-guided visual attention is created, which can accurately link the relevant visual information with each semantic meaning inside the text. Demonstrated by three experiments including both qualitative and quantitative analyses, our model can generate captions of high quality, specifically high levels of accuracy, completeness, and diversity. Moreover, our model significantly outperforms all other methods that use VGG-based ...
Zheng, D, Zhang, H, Zhang, JA & Li, Y 2019, 'Consensus of the Second-order Multi-agent Systems under Asynchronous Switching with a Controller Fault', INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS, vol. 17, no. 1, pp. 136-144.
View/Download from: Publisher's site
Zheng, G, Yang, W, Valli, C, Shankaran, R, Abbas, H, Zhang, G, Fang, G, Chaudhry, J & Qiao, L 2019, 'Fingerprint Access Control for Wireless Insulin Pump Systems Using Cancelable Delaunay Triangulations', IEEE Access, vol. 7, pp. 75629-75641.
View/Download from: Publisher's site
View description>>
© 2013 IEEE. An insulin pump is a small wearable medical gadget which can mimic the way a healthy pancreas functions by providing continuous subcutaneous insulin infusion for the patient. However, in the current products, the access to the pump is not securely controlled, rendering the pump insecure to harmful or even lethal attacks, such as those that lead to the privacy breach of the patient and the delivery of abnormal dosage of insulin to the patient. In a conventional symmetric key-based security solution, how to distribute and manage the key is quite challenging, since the patient bearing an insulin pump may visit any hospital or clinic to receive treatment from any qualified doctor. In order to prevent malicious access to the pump, in this paper, we propose a Fingerprint-based Insulin Pump security (FIPsec) scheme which requires an entity to first pass the fingerprint authentication process before it is allowed to access the insulin pump. With this scheme, the request from an adversary to access the pump will be blocked thereby protecting the pump from the possibility of being subjected to serious attacks during the post-request period. In the scheme, a cancelable Delaunay triangle-based fingerprint matching algorithm is proposed for the insulin pump, which has capabilities to resist against nonlinear fingerprint image distortion and the influence of missing or spurious minutiae. In order to evaluate the performance of the proposed fingerprint matching algorithm, we perform comprehensive experiments on FVC2002 DB1 and DB2 fingerprint databases. The results show that the FIPsec scheme can achieve a reasonably low equal error rate and thus becomes a viable solution to thwarting unauthorized access to the insulin pump.
Zhou, W, Sutton, GJ, Zhang, JA, Liu, RP & Pan, S 2019, 'Delay-Guaranteed Admission Control for LAA Coexisting With WiFi', IEEE WIRELESS COMMUNICATIONS LETTERS, vol. 8, no. 4, pp. 1048-1051.
View/Download from: Publisher's site
Zhu, H & Guo, YJ 2019, 'Wideband Filtering Phase Shifter Using Transversal Signal-Interference Techniques', IEEE Microwave and Wireless Components Letters, vol. 29, no. 4, pp. 252-254.
View/Download from: Publisher's site
View description>>
© 2001-2012 IEEE. A wideband filtering phase shifter is presented in this letter using transversal signal-interference techniques. The proposed structure creates two signal-propagation paths by introducing extra coupling between two short-ended stubs. Cascaded coupled-line sections and parallel short-ended stubs can generate a constant passband and multiple transmission zeros (TZs) in the stopband and meanwhile provide the arbitrary value of phase shift within the filtering band range. The positions of TZs and phase shift range can be easily controlled. Explicit relations between the objective filtering band and the related parameters are given, and cases with different in-band phase shift values are studied. To validate the design, a prototype is built, simulated, and tested using microstrip lines. The experimental results agree with the predicted ones, demonstrating that the device can realize any given in-band phase shift with a wideband filtering response.
Zhu, H, Cheng, Z & Guo, YJ 2019, 'Design of wideband in-phase and out-of-phase power dividers using microstrip-to-slotline transitions and slotline resonators', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 4, pp. 1412-1424.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A new class of in-phase and out-of-phase power dividers with constant equal-ripple frequency response and wide operating bandwidth is presented in this paper. The proposed design is based on microstrip-to-slotline transitions and slotline resonators. A slotted T-junction is adopted to split the power into two parts and obtain wideband isolation between the two output signals at the same time. The characteristic impedance of the transitions and resonators determines the operating bandwidth and in-band magnitude response. By reversing the placement direction of the slotline-to-microstrip transition, the electrical field is reversed, thus resulting in out-of-phase responses between output ports. A thorough analysis of the relations between the structure and the characteristic functions is provided to guide the selection of parameters of the structure in order to meet the design objectives. In the structure, simulation and measurement are conducted to verify the design method. For both in-phase and out-of-phase cases, more than 110% bandwidth has been achieved with excellent matching at all ports and isolation of output signals. Constant in-band ripple is obtained within the operating band of the power dividers, indicating that the proposed design can realise minimal power deviations, which is extremely desired in wireless systems.
Zhu, H, Lin, JY & Guo, YJ 2019, 'Filtering Balanced-to-Single-Ended Power Dividers with Wide Range and High Level of Common-Mode Suppression', IEEE Transactions on Microwave Theory and Techniques, vol. 67, no. 12, pp. 5038-5048.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. A new design approach for developing balanced-to-single-ended (BTSE) power dividers (PDs) to achieve high-level and wide range of common-mode (CM) suppression and minimum mode-conversion level is presented. The suppression of CM signals and mode conversion is realized by a microstrip-to-slotline transition, which is due to the orthogonality between the electric field of the microstrip line and slotline. Filtering responses are included in the differential-mode transmission performance, and multiple transmission zeros are generated by loading shunted coupled-line stubs. A multimode slotline resonator is used to provide multiple resonances in the passband, and these resonances can be easily controlled so that the operating bandwidth can be varied in a large frequency range. Based on this design approach, several BTSE PDs with different bandwidths are simulated in the EM environment. Prototypes are fabricated and tested to verify the design. The experimental results reveal that 200% fractional bandwidth of CM suppression is obtained. The CM suppression and mode-conversion levels are below-35 dB at all frequencies, which is extremely desired in differential circuits and systems.
Zhu, H, Sun, HH, Jones, B, Ding, C & Guo, YJ 2019, 'Wideband Dual-Polarized Multiple Beam-Forming Antenna Arrays', IEEE Transactions on Antennas and Propagation, vol. 67, no. 3, pp. 1590-1604.
View/Download from: Publisher's site
View description>>
IEEE Wideband multi-beam antenna arrays based on three-beam Butler matrices are presented in this paper. The proposed beam-forming arrays are particularly suited to increasing the capacity of 4G long-term evolution (LTE) base stations. Although dual-polarized arrays are widely used in LTE base stations, analogue beam-forming arrays have not been realized before, due to the huge challenge of achieving wide operating bandwidth and stable array patterns. To tackle these problems, for the first time, we present a novel wideband multiple beam-forming antenna array based on Butler matrices. The described beam-forming networks produce three beams but the methods are applicable to larger networks. The essential part of the beam-forming array is a wideband three-beam Butler matrix, which comprises quadrature couplers and fixed wideband phase shifters. Wideband quadrature and phase shifters are developed using striplines, which provide the required power levels and phase differences at the outputs. To achieve the correct beamwidth and to obtain the required level of crossover between adjacent beams, beam-forming networks consisting of augmented three-beam Butler matrices using power dividers are presented to expand the number of output ports from three to five or six. Dual-polarized, three-beam antenna arrays with five and six elements covering LTE band are developed. Prototypes comprising beam-forming networks and arrays are tested according to LTE base station specification. The test results show close agreement with the simulation ones and compliance with LTE requirements. The designs presented are applicable to a wide range of wideband multi-beam arrays.
Zhu, J, Yang, Y, Li, S, Liao, S & Xue, Q 2019, 'Dual-Band Dual Circularly Polarized Antenna Array Using FSS-Integrated Polarization Rotation AMC Ground for Vehicle Satellite Communications', IEEE Transactions on Vehicular Technology, vol. 68, no. 11, pp. 10742-10751.
View/Download from: Publisher's site
View description>>
© 1967-2012 IEEE. This paper presents a new dual-band dual circularly polarized (CP) high gain patch antenna array for vehicle satellite communications. The array consists of 16 linearly polarized dual-band elements backed with a new frequency-selective-surface (FSS)-integrated polarization rotation ground. The polarization rotation ground is located underneath the array, which not only acts as a reflector to increase the array boresight gain but also enables the array's dual CP radiation. The electromagnetic (EM) waves reflected by the ground retard either 90° or 270° with respect to the EM waves of the array directly radiated to the upper sphere at two frequency bands, leading to the dual-band dual CP radiation. This is totally different from many of the former counterparts which are based on either the perturbation on the stacked patch or external feeding network. Measured results show the -10-dB impedance bandwidth is from 7.8 to more than 8.5-GHz for X-band and from 14 to 15.3-GHz for Ku-band. The 3-dB axial ratio bandwidth is from 8.15 to 8.35-GHz for the lower band (right-hand CP) and from 14.2 to 14.8-GHz for the higher band (left-hand CP), respectively.
Zhu, J, Yang, Y, Li, S, Liao, S & Xue, Q 2019, 'Single-ended-fed high-gain LTCC planar aperture antenna for 60 GHz antenna-in-package applications', IEEE Transactions on Antennas and Propagation, vol. 67, no. 8, pp. 5154-5162.
View/Download from: Publisher's site
View description>>
© 1963-2012 IEEE. This paper presents new single-ended-fed planar aperture antennas (SPAAs) using low-temperature co-fired ceramics (LTCC) process technology. The SPAA element is proposed first, which not only inherits the merits of the aperture antennas including high gain and wide bandwidth but also exhibits advantages of low profile and compact size. The aperture is excited by a cross-shaped patch, and a loop-shaped balun structure placed below the patch is introduced to convert the single-ended signal into differential one to drive the patch. In this way, the energy can propagate on the patch in a traveling waveform and illuminate the aperture with uniform E-field distributions. Therefore, the antenna achieves good electrical and radiation performances, which are comparable to its balanced-fed counterparts, while processing a simplified structure. Measured results demonstrate that the impedance bandwidth of the antenna covers the 60 GHz license-free band (57-64 GHz), and the maximum gain can reach 11.5 dBi with a cavity area of only about 27 mm2. Furthermore, the element is successfully extended to a 4 × 4 element array using a substrate-integrated-waveguide based feeding network to further increase the gain up to 20.4 dBi. The measured results show that the impedance bandwidth of the array is from 57.5 to 65.7 GHz and the radiation performances are very stable over the operating frequency band.
Ziolkowski, RW 2019, 'Custom-Designed Electrically Small Huygens Dipole Antennas Antennas Achieve Efficient, Directive Emissions Into Air When Mounted on a High Permittivity Block', IEEE ACCESS, vol. 7, pp. 163365-163383.
View/Download from: Publisher's site
Zuo, Y, Fang, Y, Yang, Y, Shang, X & Wu, Q 2019, 'Depth Map Enhancement by Revisiting Multi-scale Intensity Guidance within Coarse-to-fine Stages', IEEE Transactions on Circuits and Systems for Video Technology.
View/Download from: Publisher's site
View description>>
IEEE Being different from the most methods of guided depth map enhancement based on deep convolutional neural network which focus on increasing the depth of networks, this paper is to improve the effectiveness of intensity guidance when the network goes deep. Overall, the proposed network upsamples the low-resolution depth maps from coarse to fine. Within each refinement stage of certain-scale depth features, the current-scale and all coarse-scales of the guidance features are revisited by dense connection. Therefore, the multi-scale guidance is efficiently maintained as the propagation of features. Furthermore, the proposed network maintains the intensity features in the high-resolution domain from which the multi-scale guidance is directly extracted. This design further improves the quality of intensity guidance. In addition, the shallow depth features upsampled via transposed convolution layer are directly transferred to the final depth features for reconstruction, which is called global residual learning in feature domain. Similarly, the global residual learning in pixel domain learns the difference between the depth ground truth and the coarsely upsampled depth map. Also, the local residual learning is to maintain the low frequency within each refinement stage and progressively recover the high frequency. The proposed method is tested for noise-free and noisy cases which compares against 16 state-of-the-art methods. Our experimental results show the improved performances based on the qualitative and quantitative evaluations.
Abdollahi, M, Abolhasan, M, Shariati, N, Lipman, J, Jamalipour, A & Ni, W 2019, 'A Routing Protocol for SDN-based Multi-hop D2D Communications', 2019 16th IEEE Annual Consumer Communications and Networking Conference, CCNC 2019, EEE Annual Consumer Communications & Networking Conference, IEEE, USA, pp. 895-898.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a new Multi-hop Device-to-Device (MD2D) routing protocol, referred to as SMDRP (SDN-based Multi-hop D2D Routing Protocol), for SDN-based wireless networks. Our proposed protocol can be considered as a semi-distributed routing protocol, where an SDN controller manages and controls part of the overall MD2D routing functionality to increase scalability while enabling network operators to control and maintain the out-of-band packet forwarding network. This paper also extends prior work on the Hybrid SDN Architecture for Wireless Distributed Networks (HSAW) [1] and is adapted to the framework presented in this paper. In HSAW, since all link state information is flooded by the controller to the nodes, the network will experience scalability problem. In our approach, this problem is overcome by only passing the next hop for each active route to the mobile nodes. To investigate this, we performed a theoretical and simulation studies comparing HSAW with SMDRP. From our result, it can be seen that for larger density populated networks, SMDRP shows better scalability than HSAW. In addition, mobile nodes need less memory and energy for their communications.
Abeywickrama, HV, He, Y, Dutkiewicz, E & Jayawickrama, BA 2019, 'An Adaptive UAV Network for Increased User Coverage and Spectral Efficiency', IEEE Wireless Communications and Networking Conference, WCNC, IEEE Wireless Communications and Networking Conference, IEEE, Marrakesh, Morocco.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Unmanned Aerial Vehicles (UAVs) are fast becoming a popular choice in a variety of applications in wireless communication systems. UAV-mounted base stations (UAV-BSs) are an effective and cost-efficient solution for providing wireless connectivity where fixed infrastructure is not available or destroyed. We present a method of using UAV-BSs to provide coverage to mobile users in a fixed area. We propose an algorithm for predicting the user locations based on their mobility data and clustering the predicted locations, so that one UAV-BS would provide coverage to one user cluster. The proposed method, hence is similar to the UAV-BSs following the users to keep them under the coverage region. Simulation results show that the proposed method increases the user coverage by 47%-72% and increases the spectral efficiency by 43%-55% depending on the scenario and in addition, reduces the number of UAV-BSs required to provide coverage.
Abeywickrama, HV, Jayawickrama, BA, He, Y & Dutkiewicz, E 2018, 'Empirical Power Consumption Model for UAVs', 2018 IEEE 88th Vehicular Technology Conference (VTC-Fall), IEEE Vehicular Technology Conference, IEEE, Chicago, IL, USA, USA, pp. 1-5.
View/Download from: Publisher's site
View description>>
Unmanned Aerial Vehicles (UAV) are gaining popularity in a range of areas and are already being used for a wide variety of purposes. While UAVs have many desirable features, limited battery lifetime is identified as a key restriction in UAV applications. Typical UAVs being electric devices, powered by on-board batteries, this constrain has limited their capabilities to a considerable extent. Thus planning UAV missions in an energy efficient manner is of utmost importance. To achieve this, for prediction of power consumption, it is necessary to have a reliable power consumption model. In this paper, we present a consistent and complete power consumption model for UAVs based on empirical studies of battery usage for various UAV activities. The power consumption model presented in this paper can be readily used for energy efficient UAV mission planning.
Acut, RVP, Hora, JA, Gerasta, OJL, Zhu, X & Dutkiewicz, E 2019, 'PV-TEG- WiFi multiple sources design energy harvesting system for WSN application', 2019 4th IEEE International Circuits and Systems Symposium, ICSyS 2019.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Ambient energy harvesting is becoming an essential factor for wireless sensor nodes applications. Harvested energy from ambient sources, such as solar, indoor light, thermal energy, vibration, and radio frequencies (RF) provide a potential power capability that may supplement batteries to provide longevity to the sensor nodes. In an indoor setting, the intermittent availability of these sources will likely lessen the energy densities. Using only a single ambient-energy harvested source may not sustain the power requirement for the design of a battery-less wireless sensor node. Thus, this work investigates and design a power combiner circuit of the indoor light, thermal. The design utilized a cross-coupled charge pump operation to combined the same DC ambient sources, the PV cell, and TEG energy transducers. The system design is simulated using the 65 nm CMOS process technology. The simulation shows that the system is able to supply peak power of 1.69 mW at ±500Ω load. Peak power efficiency is 69% with 919 μA ILOAD.
Ajayan, AR, Al-Doghman, F & Chaczko, Z 2018, 'Visualizing Multimodal Big Data Anomaly Patterns in Higher-Order Feature Spaces', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, NSW, Australia, pp. 1-9.
View/Download from: Publisher's site
View description>>
The world today, as we know it, is profuse with information about humans and objects. Datasets generated by cyber-physical systems are orders of magnitude larger than their current information processing capabilities. Tapping into these big data flows to uncover much deeper perceptions into the functioning, operational logic and smartness levels attainable has been investigated for quite a while. Knowledge Discovery & Representation capabilities across mutiple modalities holds much scope in this direction, with regards to their information holding potential. This paper investigates the applicability of an arithmetic tool Tensor Decompositions and Factorizations in this scenario. Higher order datasets are decomposed for Anomaly Pattern capture which encases intelligence along multiple modes of data flow. Preliminary investigations based on data derived from Smart Grid Smart City Project are compliant with our hypothesis. The results proved that Abnormal patterns detected in decomposed Tensor factors encompass deep information energy content from Big Data as efficiently as other Pattern Extraction and Knowledge Discovery frameworks, while salvaging time and resources.
Al-Doghman, F, Chaczko, Z & Brookes, W 2018, 'Adaptive Consensus-based Aggregation for Edge Computing', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
The swift expansion in employing IoT and the tendency to apply its application have encompassed a wide range of fields in our life. The heterogeneity and the massive amount of data produced from IoT require adaptive collection and transmission processes that function closed to front-end to mitigate these issues. In this paper, We introduced a method
of aggregating IoT data in a consensus way using Bayesian analysis and Markov Chain techniques. The aim is to enhance the quality of data traveling within IoT framework.
Al-Doghman, F, Chaczko, Z, Brooke, W & Gordon, LC 2019, 'Social Consensus-inspired Aggregation Algorithms for Edge Computing', 2019 3rd Cyber Security in Networking Conference, CSNet 2019, pp. 138-141.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The current interest about the∗nternet of Things (IoT) evokes the establishment of infinite services giving huge, active, and varied information sets. Within it, an enormous mass of heterogeneous data are generated and interchanged by billions of device which can yield to an enormous information traffic jam and affects network efficiency. To get over this issue, there's a necessity for an effective, smart, distributed, and in-network technique that uses a cooperative effort to aggregate data along the pathway from the network edge to its sink. we tend to propose an information organization blueprint that systematizes data aggregation and transmission within the bounds of the Edge domain from the front-end until the Cloud. A social consensus technique obtained by applying statistical analysis is employed within the blueprint to get and update a policy concerning a way to aggregate and transmit data according to the order of information consumption inside the network. The Propose technique, consensus Aggregation, uses statistical Machine Learning to consolidate the approach and appraise its performance. inside the normal operation of the approach, data aggregation is performed with the utilization of data distribution. A notable information delivery efficiency was obtained with a nominal loss in precision as the blueprint was tested inside a particular environment as a case study. The conclusion of the strategy showed that the consensus approach overcome the individual ones in several directions.
Anwar, M & Gill, A 2019, 'A Review of the Seven Modelling Approaches for Digital Ecosystem Architecture', 2019 IEEE 21st Conference on Business Informatics (CBI), IEEE Conference on Business Informatics, IEEE, Moscow, Russia.
View/Download from: Publisher's site
View description>>
A dynamic digital ecosystem is an interrelated network of organisations, people and/or entities that interact and collaborate for value co-creation. The challenge is how to effectively model the digital ecosystems operating in a highly complex and dynamic environment. There are several modelling approaches to choose from. There is a need to evaluate the existing modelling approaches to support the modelling of digital ecosystems. This paper evaluates the scope and coverage of the selected seven modelling approaches (Adaptive Enterprise Architecture, ArchiMate, TOGAF, FAML, ISO/IEC/IEEE 42010, SABSA, and ITIL) for modelling the digital ecosystems. Adaptive enterprise architecture is taken as a reference architecture for this review due to its higher relevance to digital ecosystem layers. The results of this review indicate that every modelling methodology is different in scope and coverage and demands the integration and tailoring of a context specific modelling approaches to provide the type of support needed for digital ecosystems.
Asad, M, Yang, Z, Khan, Z, Yang, J & He, X 2019, 'Feature fusion based deep spatiotemporal model for violence detection in videos', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Neural Information Processing, Springer, Sydney, pp. 405-417.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. It is essential for public monitoring and security to detect violent behavior in surveillance videos. However, it requires constant human observation and attention, which is a challenging task. Autonomous detection of violent activities is essential for continuous, uninterrupted video surveillance systems. This paper proposed a novel method to detect violent activities in videos, using fused spatial feature maps, based on Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) units. The spatial features are extracted through CNN, and multi-level spatial features fusion method is used to combine the spatial features maps from two equally spaced sequential input video frames to incorporate motion characteristics. The additional residual layer blocks are used to further learn these fused spatial features to increase the classification accuracy of the network. The combined spatial features of input frames are then fed to LSTM units to learn the global temporal information. The output of this network classifies the violent or non-violent category present in the input video frame. Experimental results on three different standard benchmark datasets: Hockey Fight, Crowd Violence and BEHAVE show that the proposed algorithm provides better ability to recognize violent actions in different scenarios and results in improved performance compared to the state-of-the-art methods.
Ashtari, S, Tofigh, F, Abolhasan, M, Lipman, J & Ni, W 2019, 'Efficient Cellular Base Stations Sleep Mode Control Using Image Matching', 2019 IEEE 89TH VEHICULAR TECHNOLOGY CONFERENCE (VTC2019-SPRING), 89th IEEE Vehicular Technology Conference (VTC Spring), IEEE, Kuala Lumpur, MALAYSIA.
Baccini, D, Hinckley, S, Canning, J, Cook, K, Allwood, G, Wild, G, Davies, J & Banos, C 2019, 'Gamma irradiation response in photonic crystal and standard optical fiber Bragg grating sensors for radiation dosimetry', Proceedings of SPIE - The International Society for Optical Engineering, AOS Australian Conference on Optical Fibre Technology and Australian Conference on Optics, Lasers, and Spectroscopy, SPIE, Melbourne, Australia.
View/Download from: Publisher's site
View description>>
© COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only. In this paper, we report the response of Cobalt-60 gamma irradiation on Photonic Crystal Fibre Bragg gratings (PCFFBGs) and standard commercial FBGs (STD-FBGs). Optical measurements were performed to determine the shift of the Bragg wavelength as a function of accumulated dose and relaxation time. To simulate real time conditions of a radiation dosimeter, the FBGs are examined through three consecutive radiation stages followed by very limited recovery times. We were able to obtain a Bragg wavelength shift with both sets of FBGs. The PCF-FBGs response included strong recovery after each irradiation compared to the STD-FBGs. This makes the PCF-FBGs strong candidates as optical fibre FBG sensors in the area of radiation dosimetry.
Basnet, S, Jayawickrama, BA, He, Y & Dutkiewicz, E 2018, 'Fairness Aware Resource Allocation for Average Capacity Maximisation in General Authorized Access User', 2018 IEEE 88th Vehicular Technology Conference (VTC-Fall), IEEE Vehicular Technology Conference, IEEE, Chicago, IL, USA, USA, pp. 1-5.
View/Download from: Publisher's site
View description>>
Spectrum Access System (SAS) is a three-tier spectrum sharing framework proposed for 3.5 GHz by Federal Communication Commission (FCC) in the United States. General Authorized Access (GAA) users in SAS do not have an assigned channel and can opportunistically access the Priority Access Licensee (PAL) channel satisfying the interference constraint proposed by FCC. Coexistence among GAA users in SAS is a key problem to be solved to enhance the system capacity to meet the increasing traffic demand. In this work, we propose a method for fair and efficient spectrum utilisation for GAA users. To achieve the fairness among GAA users equal interference budget allocation scheme is proposed for each set of GAA users that can hear each other. Our proposed method decide the optimal channel switching schedule that maximises the average capacity of GAA users while satisfying the interference constraint at PAL protection area. This work jointly considers the fairness between GAA users and the average capacity maximisation of GAA network. Simulation result justifies the performance of our proposed method for average capacity maximisation of GAA users and fairness between GAA users by comparing with existing works.
Basnet, S, Jayawickrama, BA, He, Y & Dutkiewicz, E 2018, 'Transmit Power Allocation for General Authorized Access in Spectrum Access System Using Carrier Sensing Range', IEEE Vehicular Technology Conference, IEEE Vehicular Technology Conference, IEEE, Chicago, IL, USA, USA, pp. 1-5.
View/Download from: Publisher's site
View description>>
The optimal use of spectrum is a key focus for all regulatory bodies. Federal Communications Commission has introduced Spectrum Access System (SAS) to maximise the spectrum utilisation in the US 3.5 GHz band. SAS is a three-tier spectrum sharing framework where Citizen Broadband Radio Service (CBRS) devices can access the channel when it is not used by Incumbent Access users. CBRS consists of Priority Access Licensee (PAL) and General Authorized Access (GAA). In this paper, we consider the problem of optimum transmit power allocation for GAA users using a carrier sensing range i.e. maximum distance a user can be sensed while guaranteeing the interference to PAL from GAA users is below the threshold. We use carrier sensing range to find the sets of GAA users that cannot transmit at the same time and adjust the interference budget of transmitting GAA users. We present an algorithm for transmit power allocation for GAA users in the SAS. The proposed algorithm uses the transmission characteristics and location information provided by Citizen Broadband Radio Service Devices to SAS to maximise the peak capacity of GAA users ensuring the interference constraint to PAL. Simulation results show that the proposed algorithm significantly increases the peak capacity of GAA users by considering the carrier sensing range and adjusted interference budget.
Bautista, MG, Zhu, H, Zhu, X, Yang, Y, Sun, Y, Dutkiewicz, E & Zhang, F 2019, 'Millimeter-wave BPFs design using quasi-lumped elements in 0.13-µm (Bi)-CMOS technology', Proceedings - IEEE International Symposium on Circuits and Systems, IEEE International Symposium on Circuits and Systems, IEEE, Sapporo, Japan.
View/Download from: Publisher's site
View description>>
© 2019 IEEE A design methodology using quasi-lumped elements for compact millimeter-wave on-chip bandpass filter (BPF) is presented in this work. To implement BPF using this approach, a novel inductor cell is presented first and then using this cell along with metal-insulator-metal (MIM) capacitors, two BPFs are designed. For the purpose of proof-of-concept, all three designs are implemented and fabricated in a standard 0.13-µm (Bi)-CMOS technology. The measurements show that the inductor cell generates a notch at 47 GHz with a chip size of 0.096 × 0.294 mm2 without pads. Moreover, the 1st BPF has the center frequency at 27 GHz with an insertion loss of 2.5 dB and it has one transmission zero at 58 GHz with a peak attenuation of 23 dB. Unlike the 1st design, the 2nd design has two transmission zeros. The center frequency of this BPF is located at 29 GHz with a minimum insertion loss of 3.5 dB. Without the measurement pads, the chip sizes of the two BPFs are 0.076 × 0.296 mm2 and 0.096 × 0.296 mm2, respectively.
Bekhit, M, Abolhasan, M, Lipman, J, Liu, RP & Ni, W 2018, 'Multi objective resource optimisation for network function virtualisation requests', 26th International Conference on Systems Engineering, ICSEng 2018 - Proceedings, International Conference on Systems Engineering, IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. Network function vitalization (NFV) as a new research concept, for both academia and industry, faces many challenges to network operators before it can be accepted into mainstream. One challenge addressed in this paper is to find the optimal placement f or a set of incoming requests with VNF service chains to serve in suitable Virtual Machines (VMs) such that a set of conflicting objectives are met. Mainly, focus is placed on maximizing the total saving cost by increasing the total CPU utilization during the processing time and increasing the processing time for every service request in the cloud network. Moreover, we aim to maximize the admitted traffic simultaneously while considering the system constraints. We formulate the problem as a multi-objective optimization problem and use a Resource Utilization Multi-Objective Evolutionary Algorithm based on Decomposition (RU-MOEA/D) algorithm to solve the problem considering the two objectives simultaneously. Extensive simulations are carried out to evaluate the effects of the different network sizes, genetic parameters and the number of server resources on the acceptable ratio of the arrival chains to serve in the available VMs. The empirical results illustrate that the proposed algorithm can solve the problem efficiently and compute the optimal solution for two objectives together within a reasonable running time.
Bird, TS 2019, 'Role of the parkes radiotelescope in the first moon landing', 2019 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, APSURSI 2019 - Proceedings, pp. 1377-1378.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The Parkes radiotelescope and Honeysuckle Creek stations in Australia received the telemetry from the Lunar Module 'Eagle', including the first television pictures from the first moonwalk for distribution to millions of people on Earth. Parkes' TV picture quality was superior to those received at the other ground stations, therefore NASA used Parkes as the TV source for all but the first nine minutes of the over 5 hours of the TV broadcast. The Parkes radiotelescope is historically important for its contribution to the first Moon landing as well as its design and for the corrugated feed horn first used at Parkes.
Bożejko, W, Chaczko, Z, Nadybski, P & Wodecki, M 2018, 'Meta-heuristic task scheduling algorithm for computing cluster with 2D packing problem approach', Advances in Intelligent Systems and Computing, International Conference on Dependability and Complex Systems, Springer, Brunów, Poland, pp. 74-82.
View/Download from: Publisher's site
View description>>
© Springer International Publishing AG, part of Springer Nature 2019. In this paper we present a mathematical model and an algorithm for solving a task scheduling problem in computing cluster. The problem is considered as a 2D packing problem. Each multi-node task is treated as a set of separate subtasks with common constrains. For optimization the tabu search metaheuristic algorithm is applied.
Braun, R, Bone, D, Brookes, W, Trede, F & Hadgraft, R 2019, 'Studios in DE and EE at UTS: Structure and rationale', 2019 18th International Conference on Information Technology Based Higher Education and Training, ITHET 2019, International Conference on Information Technology Based Higher Education and Training, IEEE, Magdeburg, Germany.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. We describe the Studios we have introduced into our Data and Electronic Engineering programs. We explain the purpose of the Studios, and the structure of activities. We describe the rationale for the significant components. We comment on the success of the components, and lessons learned.
Braun, R, Brookes, W, Hadgraft, R & Chaczko, Z 2019, 'Assessment Design for Studio-Based Learning', Proceedings of the Twenty-First Australasian Computing Education Conference, Australasian Computing Education Conference, ACM, Sydney, Australia, pp. 106-111.
View/Download from: Publisher's site
View description>>
Studio-based learning is not new to computing education, however as the ecosystem of available Open Educational Resources (OERs) expands, the capacity and desire for student self-directed learning is growing. However increasing student autonomy in how and when learning takes place creates challenges around assessment. This paper introduces the design of assessment tasks to support studio-based learning at undergraduate level. It describes an example of using learning contracts and portfolio-based assessment for evaluating individual and team performance. The paper presents some initial observations of the approach taken, and its transferability to other areas of the curriculum.
Cai, Z, Tang, X, Li, Z, Zhang, T, Liu, Y & Yang, Y 2019, 'A Low Phase Noise Differential Oscillator Employing Stub-Loaded Nested Split-Ring Resonator Inspired Balanced Bandpass Filter', IEEE MTT-S International Microwave Symposium Digest, IEEE MTT-S International Microwave Symposium, IEEE, Boston, MA, USA, pp. 967-970.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a low phase noise differential oscillator by employing a balanced feedback bandpass filter (BPF) designed with the proposed stub-loaded nested split-ring resonator (SLNSRR). The proposed balanced BPF functions as a frequency stabilization element in its feedback loop. Taking advantage of the balanced structure, a high group delay is obtained by introducing a transition zero near the upper passband of feedback loop filter. The proposed differential oscillator can present a differential output with low phase noise performance. For proof of the concept, a 2 GHz differential oscillator has been designed, fabricated and measured. The measured results show that the 180o out-of-phase differential signals are obtained with almost the same peak-peak voltage among two channels. The output power is 9.18 dBm when oscillates at 2.004 GHz with the second harmonic suppression of 40.63 dB. The measured phase noise is -126.72 dBc/Hz at 100 kHz frequency offset. The figure-of-merit (FOM) at 100 kHz is -196.19 dBc/Hz. The phase noise performance of proposed differential oscillator is one of the best among open literatures.
Calam, RCM, Hora, JA, Gerasta, OJL, Zhu, X & Dutkiewicz, E 2019, 'A self-calibrating off-time controller for WSN/IoT synchronous non-inverting buck-boost DC-to-DC converter application', 2019 4th IEEE International Circuits and Systems Symposium, ICSyS 2019.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Inappropriate turning-off of synchronous switches in DC-to-DC converters operating in discontinuous conduction mode (DCM) degrades the overall efficiency of the converter due to the power losses in either body-diode conduction or reverse inductor current. This paper presents a self-calibrating off-time controller using a digitally-controlled delay element (DCDE) for synchronous DC-to-DC converter. An up-down counter that is controlled by the polarity of the inductor current is utilized to drive the binary-weighted DCDE. The programmable DCDE is used to generate a self-calibrating off-time pulse to turn-off the synchronous switches very close to the zero-crossing of the inductor current. The design and simulation of the proposed method is implemented using 65nm CMOS Technology process in Synopsys Custom Designer tool. When implemented, the measured accuracy of the detection is in order of less than ±10mA. The average power consumption of the proposed system is less than 10 uW and is expected to be much lower at lighter loads.
Canning, J 2019, 'Towards optical manipulation on a chip', Optics InfoBase Conference Papers, OSA Advanced Photonics Congress, OSA, Burlingame, California, United States.
View description>>
© 2019 The Author(s). Hoovering of optical water using an optical tractor on an SPR excited sputtered gold (Au) surface is proposed and demonstrated. Direct visual observation, and measured by contact angle, shows accumulation of water towards optical light passing through a gold layered metal film (e-mail: john.canning@uts.edu.au).
Cao, Y & Veitch, D 2019, 'Where on Earth Are the Best-50 Time Servers?', Passive and Active Measurement 20th International Conference, PAM 2019, Puerto Varas, Chile, March 27–29, 2019, Proceedings, International Conference on Passive and Active Network Measurement, Spronger, Chile, pp. 101-115.
View/Download from: Publisher's site
View description>>
© 2019, Springer Nature Switzerland AG. We present a list of the Best-50 public IPv4 time servers by mining a high-resolution dataset of Stratum-1 servers for Availability, Stratum Constancy, Leap Performance, and Clock Error, broken down by continent. We find that a server with ideal leap performance, high availability, and low stratum variation is often clock error-free, but this is no guarantee. We discuss the relevance and lifetime of our findings, the scalability of our approach, and implications for load balancing and server ranking.
Chaczko, Z, Wajs-Chaczko, P, Tien, D & Haidar, Y 2019, 'Detection of Microplastics Using Machine Learning', 2019 International Conference on Machine Learning and Cybernetics (ICMLC), International Conference on Machine Learning and Cybernetics, IEEE, Kobe, Japan.
View/Download from: Publisher's site
View description>>
Monitoring the presence of micro-plastics in human and animal habitats is fast becoming an important research theme due to a need to preserve healthy ecosystems. Microplastics pollute the environment and can represent a serious threat for biological organisms including the human body, as they can be inadvertently consumed through the food chain. To perceive and understand the level of microplastics pollution threats in the environment there is a need to design and develop reliable methodologies and tools that can detect and classify the different types of microplastics. This paper presents results of our work related to the exploration of methods and techniques useful for detecting suspicious objects in their respective ecosystems captured in hyperspectral images and then classifying these objects with the use of Neural Networks technique.
Chemalamarri, VD, Braun, R, Lipman, J & Abolhasan, M 2018, 'A Multi-agent Controller to enable Cognition in Software Defined Networks', 2018 28th International Telecommunication Networks and Applications Conference (ITNAC), International Telecommunication Networks and Applications Conference, IEEE, Sydney, Australia, pp. 52-56.
View/Download from: Publisher's site
View description>>
Current SDN controllers are not cognitive. We propose a new architecture for an SDN controller to enable intelligence. The proposed new architecture is based on Multi-agent systems. As a prototype, we have built a MAS-SDN controller using the GOAL agent programming language. We highlight the motivation behind the new architecture, describe the architecture and provide some initial results
Chen, L, Liu, Y & Guo, YJ 2019, 'Efficient Frequency-invariant Beam Pattern Synthesis with Multiple Space-frequency Nulls', 2019 IEEE International Conference on Computational Electromagnetics, ICCEM 2019 - Proceedings, IEEE International Conference on Computational Electromagnetics, IEEE, Shanghai, China.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a modified Fourier transform method to synthesize a broadband array with frequency-invariant (FI) mainlobe and multiple space-frequency nulls. In this method, a iterative Fourier transform is used to design multiple optimized reference patterns which have the same mainlobe shape but with different sidelobe or null distributions. These reference patterns are used for describing different radiation requirements in different sub-bands of the whole frequency band. Then, a desired broadband spatial spectral distribution can be obtained by matching it to each reference pattern at each sub-band, and a broadband excitation distribution for generating the desired FI pattern can be efficiently found by performing an inverse fast Fourier transform (IFFT) on the broadband spatial spectral distribution. An example for synthesizing a FI beam pattern with two space-frequency nulls is provided to verify the effectiveness and efficiency of the proposed method.
Chen, SL, Karmokar, DK, Ziolkowski, RW & Guo, YJ 2019, 'Wide-angle wideband frequency-independent beam-scanning leaky wave antenna', Proceedings of the 2019 21st International Conference on Electromagnetics in Advanced Applications, ICEAA 2019, International Conference on Electromagnetics in Advanced Applications, IEEE, Granada, Spain, pp. 554-557.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Frequency-independent beam scanning leaky-wave antennas (LWAs) that can operate over a specific frequency band are highly desirable for future wireless systems. A composite right/left-handed (CRLH) LWA is developed in this paper that facilitates these functionalities. It utilizes two groups of varactor diodes to realize the frequency-independent beam scanning capability. The optimized reconfigurable CRLH LWA and its simple DC biasing network achieves a simulated frequency-independent beam that scans over 100° at each frequency point between 4.75 and 5.25 GHz. An antenna prototype was fabricated and tested. The measured results at 5.0 GHz confirm its simulated performance characteristics.
Chu, P, Zhang, JA, Wang, X, Fang, G & Wang, D 2019, 'Semi-persistent v2x resource allocation with traffic prediction in two-tier cellular networks', IEEE Vehicular Technology Conference, IEEE Vehicular Technology Conference, IEEE, Kuala Lumpur, Malaysia.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In a dense urban area, conventional cellular V2X communications require frequent and heavy resource allocation, which can lead to processing congestion and large delay. In this paper, we propose a semi- persistent resource allocation scheme using the least minimum mean square error (LMMSE) traffic prediction in a two-tier network. The two-tier network architecture includes a central macro base station (MBS) and multiple roadside units (RSU). In the proposed scheme, the MBS pre-allocates persistent resource to RSUs based on predicted traffic, and then allocate dynamic resource upon real-time requests from vehicles through RSUs. We formulate an optimization problem for minimizing the total bandwidth under latency constraints and provide an optimal solution to the problem. Simulation is conducted for both artificially generated and real-world data, and the results validate the effectiveness of the proposed semi- persistent scheme.
Cui, P, Zhang, J, Lu, W, Guo, Y & Zhu, H 2018, 'Sparse Channel Modelling Using Multi-measurement Vector Compressive Sensing', 2018 IEEE Global Communications Conference (GLOBECOM), IEEE Global Communications Conference, IEEE, Abu Dahbi, UAE.
View/Download from: Publisher's site
View description>>
Channel sparsity is well exploited for channel estimation, but there is very limited work on sparse channel modelling, which studies and characterizes the statistical properties of sparse channel coefficients. In this paper, we study sparse channel modelling using real measured channel data in off-body signal propagation. We propose multi-measurement vector based compressive sensing algorithms for extracting sparse channel coefficients, study the statistical properties of these extracted coefficients, and develop an algorithm for generating simulated channels using the statistical sparse model. The proposed method can be directly applied to other channel measurements, and is very useful for channel simulation and developing advanced sparse channel estimation schemes
Cui, P, Zhang, J, Lu, W-J, Guo, YJ & Zhu, H 2019, 'Influence of Human Body on Massive MIMO Indoor Channels', Vehicular Technology Conference, IEEE, Kuala Lumpur, Malaysia.
View/Download from: Publisher's site
View description>>
Massive MIMO can dramatically improve capacity and spectral efficiency. However, it is not very clear whether it can significantly improve the signal blockage problem that exists in single antenna systems. In this paper, we investigate the impact of the human body on indoor massive MIMO channels, using practically measured channel data for a 32x8 massive MIMO system in a complex office environment. We introduce a parameter of Power Imbalance (PI) indices to estimate the wide-sense none-stationarity in multiple domains and another parameter of Channel Popularity Indices (CPI) to predict the popularity of MIMO channel. We find that in most cases, the presence of the human body still has a non- negligible negative impact. It decreases the ergodic capacity by about 8% and increases the path loss exponent by 1. In average, the ergodic capacity for NLOS channels are 15% higher than that for LOS.
Dasgupta, A, Gill, A & Hussain, F 2019, 'A Conceptual Framework for Data Governance in IoT-enabled Digital IS Ecosystems', Proceedings of the 8th International Conference on Data Science, Technology and Applications: DATA, 8th International Conference on Data Science, Technology and Applications, SCITEPRESS – Science and Technology Publications, Prague, Czech Republic, pp. 209-216.
View/Download from: Publisher's site
Ding, C, Wang, K & Guo, YJ 2018, 'Building Antennas on Perovskite Solar Cell (PSC) for Hybrid Solar/EM Wireless Energy Harvesting and Transfer', 2018 Asian Wireless Power Transfer Workshop (AWPT), Sendai, Japan.
Dinh, H, Niyato, D, Wang, P, Domenico, AD & Strinati, EC 2018, 'Optimal Cross Slice Orchestration for 5G Mobile Services', 2018 IEEE 88th Vehicular Technology Conference (VTC-Fall), IEEE Vehicular Technology Conference, IEEE, Chicago, IL, USA, USA.
View/Download from: Publisher's site
View description>>
5G mobile networks encompass the capabilities of hosting a variety of
services such as mobile social networks, multimedia delivery, healthcare,
transportation, and public safety. Therefore, the major challenge in designing
the 5G networks is how to support different types of users and applications
with different quality-of-service requirements under a single physical network
infrastructure. Recently, network slicing has been introduced as a promising
solution to address this challenge. Network slicing allows programmable network
instances which match the service requirements by using network virtualization
technologies. However, how to efficiently allocate resources across network
slices has not been well studied in the literature. Therefore, in this paper,
we first introduce a model for orchestrating network slices based on the
service requirements and available resources. Then, we propose a Markov
decision process framework to formulate and determine the optimal policy that
manages cross-slice admission control and resource allocation for the 5G
networks. Through simulation results, we show that the proposed framework and
solution are efficient not only in providing slice-as-a-service based on the
service requirements, but also in maximizing the provider's revenue.
Dinh, TH, Alsheikh, MA, Gong, S, Niyato, D, Han, Z & Liang, Y-C 2019, 'Defend Jamming Attacks: How to Make Enemies Become Friends', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE.
View/Download from: Publisher's site
dos Santos, AP & Chaczko, Z 2018, 'Blockchain: Status-Quo, Enablers and Inhibitors', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
Blockchain has been evolving and gaining new heights over the years. The shift in the perspective is allowing new user cases beyond the cryptocurrency space. Cryptocurrencies are digital assets supported by the complexities of cryptography, game theory and peer-to-peer networks. Blockchain became a popular platform for decentralized applications, as well as a valuable tool for start-ups seeking fundraising. The aim of this research paper is to review and assess the status quo for each branch of use cases, and then analyze the enabling and inhibiting factors influencing the adoption of blockchain. These findings permit a broader comprehension over the concepts backing blockchain. It will help new users to establish strategies, develop solutions and encourage the employment of blockchain technology.
Du, A, Huang, X, Zhang, J, Yao, L & Wu, Q 2019, 'Kpsnet: Keypoint Detection and Feature Extraction for Point Cloud Registration', Proceedings - International Conference on Image Processing, ICIP, IEEE International Conference on Image Processing, IEEE, Taipei, Taiwan, pp. 2576-2580.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents the KPSNet, a KeyPoint Siamese Network to simultaneously learn task-desirable keypoint detector and feature extractor. The keypoint detector is optimized to predict a score vector, which signifies the probability of each candidate being a keypoint. The feature extractor is optimized to learn robust features of keypoints by exploiting the correspondence between the keypoints generated from two inputs, respectively. For training, the KPSNet does not require to manually annotate keypoints and local patches pairwise. Instead, we design an alignment module to establish the correspondence between the two inputs and generate positive and negative samples on-the-fly. Therefore, our method can be easily extended to new scenes. We test the proposed method on the open-source benchmark and experiments show the validity of our method.
Fan, X, Xiang, C, Gong, L, He, X, Chen, C & Huang, X 2019, 'Urbanedge: Deep learning empowered edge computing for urban IoT time series prediction', ACM International Conference Proceeding Series, ACM Turing Celebration Conference, ACM, Chengdu, China.
View/Download from: Publisher's site
View description>>
© 2019 Association for Computing Machinery. The revolution of smart city has led to rapid development and proliferation of Internet of Things (IoT) technologies, with the focus on transmitting raw sensory data into valuable knowledge. Meanwhile, the ubiquitous deployments of IoT are raising the importance of processing data in real-time at the edge of networks rather than in remote cloud data centers. Based on above, edge computing has been proposed to exploit the capabilities of edge devices in providing in-proximity computing services for various IoT applications. In this paper, we present UrbanEdge, a conceptual edge computing architecture empowered by deep learning for urban IoT time series prediction. We design a hierarchical architecture to process correlated IoT time series and illustrate the work-flow of UrbanEdge in data collection, data transmission and data processing. As a core component of UrbanEdge, a deep learning model is developed with attention-based recurrent neural networks. Composed with multiple processing layers, the deep learning model can extract feature representations from raw IoT data for monitoring and prediction. We evaluate the designed deep learning model of UrbanEdge on real-world datasets, evaluation results show that the UrbanEdge outperforms other baseline methods in time series prediction.
Galvão, JR, De Bastos, TP, Zamarreño, CR, Canning, J, Martelli, C & Da Silva, JCC 2019, 'Smart carbon fiber sensing systems applied to biomechanics', Optics InfoBase Conference Papers.
View/Download from: Publisher's site
View description>>
© 2019 The Autho r(s). This paper presents three applications of carbon fiber reinforced polymer with integrated FBG sensor systems in biomechanics. In vivo tests were performed showing that the sensors are robust for the different applications.
Gamal, M, Abolhasan, M, Jafarizadeh, S, Lipman, J & Ni, W 2019, 'Mapping and Scheduling of Virtual Network Functions using Multi Objective Optimization Algorithm', Proceedings - 2019 19th International Symposium on Communications and Information Technologies, ISCIT 2019, International Symposium on Communications and Information Technologies, IEEE, Ho Chi Minh City, Vietnam, pp. 328-333.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Within the context of Software-Defined Networking (SDN), the problem of resource allocation for a set of incoming Virtual Network Functions (VNF) service requests has been the focus of many studies. In this paper, a new optimization model has been developed to find the near to optimal mapping and scheduling for the incoming VNF service requests. This model while considering delay, aims to achieve three objectives functions, namely, minimizing the transmission delays occurring in every link, minimizing the processing capacity for every Virtual Machine (VM) and minimizing the processing delay at every VM. The resultant problem is formulated as a multi-objective optimization problem and the developed solution is based on a multi-objective evolutionary algorithm utilizing the decomposition algorithm. Simulation results illustrate that the resulting algorithm is scalable while considering delay and it outperforms the genetic bandwidth link allocation (GA-BA) and genetic non-bandwidth link allocation (GA-NBA) algorithms.
Gamal, M, Jafarizadeh, S, Abolhasan, M, Lipman, J & Ni, W 2019, 'Mapping and scheduling for non-uniform arrival of virtual network function (VNF) requests', IEEE Vehicular Technology Conference, Vehicular Technology Conference, IEEE, Honolulu, HI, USA.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. As a new research concept for both academia and industry, there are several challenges faced by the Network Function Virtualization (NFV). One such challenge is to find the optimal mapping and scheduling for the incoming service requests which is the focus of this study. This optimization has been done by maximizing the number of accepted service requests, minimizing the number of bottleneck links and the overall processing time. The resultant problem is formulated as a multi- objective optimization problem, and two novel algorithms based on genetic algorithm have been developed. Through simulations, it has been shown that the developed algorithms can converge to the near to optimal solutions and they are scalable to large networks.
Gao, X, Zhang, T, Du, J & Guo, YJ 2019, 'Ultrasensitive Terahertz High-T-c Superconducting Receivers', 2019 IEEE MTT-S INTERNATIONAL WIRELESS SYMPOSIUM (IWS 2019), 6th IEEE MTT-S International Wireless Symposium (IWS) part of China Microwave Week, IEEE, Guangzhou, PEOPLES R CHINA.
Gorski, T, Bednarski, J & Chaczko, Z 2018, 'Blockchain-based renewable energy exchange management system', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia, Australia, pp. 1-6.
View/Download from: Publisher's site
View description>>
The paper presents the concept of renewable energy
management system. The idea behind the system is to exploit
the potential of renewable energy generation sources so as
to provide additional energy services and participation in a
competitive energy market. These actions can significantly affect
the shortening of the period of return on investment of individual
customer in renewable energy sources. The paper contains a
concept of Electricity Consumption and Supply Management
System (ECSM) with application of blockchain technology.
ECSM provides functionality to monitor and record continuously
information about inbound and outbound energy to/from power
grid. Except monitoring inbound and outbound energy, solution
will provide the possibility to manage in automatic and manual
way when energy should be sent to energy grid. Information
about inbound/outbound energy will be part of smart contract
which will be confirmed and stored in every node.
Hakim, G & Braun, R 2019, 'Agent based modeling of a flange climb derailment', 26th International Conference on Systems Engineering, ICSEng 2018 - Proceedings.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. We report on the development of an Agent Based Model of a train derailment incident, considering a number of factors including friction and flange angle. We describe the background and objectives, and use the Rushall derailment as a Case Study. We use the NetLogo modeling environment to build our model. We describe the workings of the model. Two scenarios involving frequency of maintenance are tested using the model. We observe unexpected (emergent) results in one case.
He, Y, Jayawickrama, BA & Dutkiewicz, E 2019, 'Distributed Power Allocation Algorithm for General Authorised Access in Spectrum Access System', IEEE Wireless Communications and Networking Conference, WCNC, IEEE Wireless Communications and Networking Conference, IEEE, Marrakesh, Morocco.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. To meet the capacity needs of the next generation wireless communications, U.S. Federal Communications Commission has recently introduced Spectrum Access System. Spectrum is shared between three tiers - Incumbents, Priority Access Licensees (PAL) and General Authorised Access (GAA) Licensees. When the incumbents are absent, PAL and GAA share the spectrum under the constraint that GAA ensure the aggregate interference to PAL is no more than -80 dBm within the PAL protection area. Currently GAA users are required to report their geolocations. However, geolocation is private information that GAA may not be willing to share. We propose a distributed GAA power allocation algorithm that does not require centralised coordination on sharing locations with other GAA users via SAS. We analytically proved the critical point of the interference along the PAL protection area to avoid calculating the interference on every points of the area. We proposed exclusion zone, transitional zone and open zone for GAA users to calculate the self-determined transmit power. Simulation results show that our method meets the interference requirement and achieve more than 90% of capacity approximation to the optimal centralised method, while completely masking the GAA locations.
Hesamian, MH, Jia, W, He, X & Kennedy, PJ 2019, 'Atrous Convolution for Binary Semantic Segmentation of Lung Nodule', ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Brighton. UK, pp. 1015-1019.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Accurately estimating the size of tumours and reproducing their boundaries from lung CT images provides crucial information for early diagnosis, staging and evaluating patients response to cancer therapy. This paper presents an advanced solution to segment lung nodules from CT images by employing a deep residual network structure with Atrous convolution. The Atrous convolution increases the field of view of the filters and helps to improve classification accuracy. Moreover, in order to address the significant class imbalance issue between the nodule pixels and background non-nodule pixels, a weighted loss function is proposed. We evaluate our proposed solution on the widely adopted benchmark dataset LIDC. A promising result of an average DCS of 81.24% is achieved, outperforming the state of the arts. This demonstrates the effectiveness and importance of applying the Atrous convolution and weighted loss for such problems.
Hora, JA, Arellano, AC, Zhu, X & Dutkiewicz, E 2019, 'Design of Buck Converter with Dead-time Control and Automatic Power-Down System for WSN Application', 2019 IEEE Wireless Power Transfer Conference, WPTC 2019, pp. 582-586.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A buck converter design with an automatic power-down technique and dead-time control system intended for low power application such as a wireless sensor network is proposed. With an input voltage range of 1V to 1.2V, the buck converter regulated the output voltage at 0.8V. This buck converter operates in a pulse-width modulation technique at load current range of 1mA-100mA. The output voltage ripple measured is 7.5 m V with the peak efficiency is 94.98 %. The quiescent current (mathrm{I}-{mathrm{q}}) of this proposed design is about 5mu mathrm{A}. The line and load regulation is 0.195 mV/V and 0.61mV/mA, respectively. The circuit core layout dimension is 179 mumathrm{m} and 120mu mathrm{m} 65nm CMOS technology.
Hora, JA, Darell Ang, J, Zhu, X & Dutkiewicz, E 2019, 'A Highly Linear OTA with 759 μs gm for RF Transceiver Application', Proceedings - 2019 19th International Symposium on Communications and Information Technologies, ISCIT 2019, International Symposium on Communications and Information Technologies, IEEE, Ho Chi Minh City, Vietnam, Vietnam, pp. 595-598.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a design of a highly linear fully differential operational transconductance amplifier (OTA) in 65nm CMOS Technology Process for RF transceiver application. To improve the linearity and the output common-mode voltage of the design OTA, a cross-coupled differential pair, and differential active loading were applied, respectively. The cross-coupled differential pair was taken from a single-ended OTA and utilized to form a fully differential OTA topology with two current mirrors. The presented OTA has a constant transconductance (gm) of 759 μS at Vin of 0V, a VDD of 1.2V and is linear at input voltages of -0.5V to 0.5V. The OTA core circuit has a gain of 22.4 dB, and unity-gain bandwidth of 280 MHz at a load of 0.2 pF and control voltage of 0.5V. The output common-mode voltage is kept close to Vdd/2 of 0.6V. The final circuit layout of the transconductance core with negative resistance circuit has dimensions of 37.22 μm x 64.59 μm. This paper also presents a pseudo-differential topology OTA for comparison purposes.
Hora, JA, Piandong, DL, Empas, PEG, Gerasta, OJL, Zhu, X & Dutkiewicz, E 2018, 'A CMOS implemented transimpedance amplifier design for optical communications', 2018 IEEE 10th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment and Management, HNICEM 2018, International Conference on Humanoid, Nanotechnology, Information Technology,Communication and Control, Environment and Management, Baguio City, Philippines.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. A transimpedance amplifier for optical communication system is presented in this study. The design includes a regulated cascode and an interleaving active feedback to improve the bandwidth of the transimpedance amplifier. Multiple gain stages are also employed to greatly improve the output voltage. This is implemented in 32 nm CMOS technology using Custom Designer from Synopsys. The circuit is designed to compete with existing transimpedance amplifiers implemented in other technologies in the field of optical communications. The transimpedance amplifier design in this study has a gain of 54 dB and a bandwidth of 9.39 GHz. The layout measures 0.0011mm2 in area and the total power dissipated is 2.94 mW.
Hora, JA, Zhu, X & Dutkiewicz, E 2019, '2.4 GHz CMOS Design RF-to-DC Energy Harvesting with Charge Control System for WSN Application', 2019 IEEE Wireless Power Transfer Conference, WPTC 2019, pp. 49-54.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents an RF-to-DC energy harvester in the Wi-Fi band. The energy harvester is meant to charge the 1.2V battery of the wireless sensor node device. The system design consists of three main circuit blocks: A low-dropout (LDO) voltage regulator, a charge control circuit and multistage differential-drive rectifier. The maximum PCE attained by the rectifier alone is 31.43%. The charge control circuit maintains the voltage within 1.3V-l.4V, while the LDO provides a stable and regulated output of 1.2V. The designed energy harvester has a minimum RF input power of -2.04 dBm. The chip layout of the overall design has a dimension of 1.2mm × 1.1mm.
Hora, JA, Zhu, X & Dutkiewicz, E 2019, 'Design of High Voltage Output for CMOS Voltage Rectifier for Energy Harvesting Design', 2019 IEEE Wireless Power Transfer Conference, WPTC 2019, pp. 40-44.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a modified design of CMOS differential voltage multiplier circuit block for energy harvesting circuit for wireless sensor networks (WSN) application. The design simulation and layout was carried out using 65nm CMOS process. The extraction of high DC voltage from rectifier block is always a severe bottleneck for energy harvesting. In this work, a simple mechanism to eliminate (Vth) of the MOS transistor by adding an auxiliary PMOS transistor is proposed. Also, an additional two capacitor (Cs) is split and connected to the differential output. Moreover, the conventional and modified voltage multiplier was simulated and implemented with three stages with a load capacitance of 100pF. The simulation result shows that the modified voltage multiplier obtain a higher voltage conversion ratio (Gv) of 3.96, while the conventional voltage multiplier only obtained a Gv of 2.96. Accordingly, the proposed modified rectifier circuit achieved a peak efficiency of 22.41 % and can able to operate a device with a power requirement of 1.2V to 1.8V and with a continuous output current of 3mA.
Hu, Z, Chung, YY, Zandavi, SM, Ouyang, W, He, X & Gao, Y 2019, 'High-Performance Light Field Reconstruction with Channel-wise and SAI-wise Attention', Communications in Computer and Information Science, International Conference on Neural Information Processing, Springer, Sydney, Australia, pp. 118-126.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. Light field (LF) images provide rich information and are suitable for high-level computer vision applications. To acquire capabilities of modeling the correlated information of LF, most of the previous methods have to stack several convolutional layers to improve the feature representation and result in heavy computation and large model sizes. In this paper, we propose channel-wise and SAI-wise attention modules to enhance the feature representation at a low cost. The channel-wise attention module helps to focus on important channels while the SAI-wise attention module guides the network to pay more attention to informative SAIs. The experimental results demonstrate that the baseline network can achieve better performance with the aid of the attention modules.
Huang, CL, Shih, YC, Lai, CM, Ying Chung, VY, Zhu, WB, Yeh, WC & He, X 2019, 'Optimization of a Convolutional Neural Network Using a Hybrid Algorithm', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Budapest, Hungary.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In recent years, Convolutional Neural Networks (CNNs) have been widely used in image recognition due to their aptitude in large scale image processing. The CNN uses Back-propagation (BP) to train weights and biases, which in turn makes the error consistently smaller. The most common optimizers that uses a BP algorithm are Stochastic Gradient Decent (SGD), Adam, and Adadelta. These optimizers, however, have been proved to fall easily into the regional optimal solution. Little research has been conducted on the application of Soft Computing in CNN to fix the above problem, and most studies that have been conducted focus on Particle Swarm Optimization. Among them, the hybrid algorithm combined with SGD proposed by Albeahdili improves the image classification accuracy over that achieved by the original CNN. This study proposes the amalgamation of Improved Simplified Swarm Optimization (iSSO) with SGD, hence culminating in the iSSO-SGD which is intended train CNNs more efficiently to establish a better prediction model and improve the classification accuracy. The performance of the proposed iSSO-SGD can be affirmed through a comparison with the PSO-SGD, the Adam, Adadelta, rmsprop and momentum optimizers and their abilities in improving the accuracy of image classification.
Huang, H, Liu, Y, Chen, L, Qin, PY & Guo, YJ 2019, 'Synthesis of a Dipole Array with Optimally End-fire Directive Pattern', 2019 IEEE International Conference on Computational Electromagnetics, ICCEM 2019 - Proceedings, IEEE International Conference on Computational Electromagnetics, IEEE, Shanghai, China.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this work, a formula is derived for generating optimally directive pattern for an arbitrary antenna array including mutual coupling effect, and it is then applied to design the optimally directive four-element end-fire dipole array. Synthesis results show that the obtained directivity coefficient is about 2dB and 3.5dB higher than those of ordinary and hansen-woodyard four-element end-fire dipole arrays.
Huang, H, Xu, J, Zhang, J, Wu, Q & Kirsch, C 2018, 'Railway Infrastructure Defects Recognition using Fine-grained Deep Convolutional Neural Networks', 2018 Digital Image Computing: Techniques and Applications (DICTA), Digital Image Computing: Techniques and Applications, IEEE, Canberra, Australia.
View/Download from: Publisher's site
View description>>
Railway power supply infrastructure is one of the most important components of railway transportation. As the key step of railway maintenance system, power supply infrastructure defects recognition plays a vital role in the whole defects inspection sub-system. Traditional defects recognition task is performed manually, which is time-consuming and high-labor costing. Inspired by the great success of deep neural networks in dealing with different vision tasks, this paper presents an end-to-end deep network to solve the railway infrastructure defects detection problem. More importantly, this paper is the first work that adopts the idea of deep fine-grained classification to do railway defects detection. We propose a new bilinear deep network named Spatial Transformer And Bilinear Low-Rank (STABLR) model and apply it to railway infrastructure defects detection. The experimental results demonstrate that the proposed method outperforms both hand-craft features based machine learning methods and classic deep neural network methods.
Huang, H, Zheng, J, Zhang, J, Wu, Q & Xu, J 2019, 'Compare more nuanced: Pairwise alignment bilinear network for few-shot fine-grained learning', Proceedings - IEEE International Conference on Multimedia and Expo, IEEE International Conference on Multimedia and Expo, IEEE, Shanghai, China, pp. 91-96.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The recognition ability of human beings is developed in a progressive way. Usually, children learn to discriminate various objects from coarse to fine-grained with limited supervision. Inspired by this learning process, we propose a simple yet effective model for the Few-Shot Fine-Grained (FSFG) recognition, which tries to tackle the challenging fine-grained recognition task using meta-learning. The proposed method, named Pairwise Alignment Bilinear Network (PABN), is an end-to-end deep neural network. Unlike traditional deep bilinear networks for fine-grained classification, which adopt the self-bilinear pooling to capture the subtle features of images, the proposed model uses a novel pairwise bilinear pooling to compare the nuanced differences between base images and query images for learning a deep distance metric. In order to match base image features with query image features, we design feature alignment losses before the proposed pairwise bilinear pooling. Experiment results on four fine-grained classification datasets and one generic few-shot dataset demonstrate that the proposed model outperforms both the state-of-the-art few-shot fine-grained and general few-shot methods.
Huang, X 2019, 'Towards Terabit Wireless Communications', 2019 Australian Communications Theory Workshop, Sydney, Australia.
Huang, X, Fan, L, Wu, Q, Zhang, J & Yuan, C 2019, 'Fast registration for cross-source point clouds by using weak regional affinity and pixel-wise refinement', Proceedings - IEEE International Conference on Multimedia and Expo, IEEE International Conference on Multimedia and Expo, IEEE, Shanghai, China, pp. 1552-1557.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Many types of 3D acquisition sensors have emerged in recent years and point cloud has been widely used in many areas. Accurate and fast registration of cross-source 3D point clouds from different sensors is an emerged research problem in computer vision. This problem is extremely challenging because cross-source point clouds contain a mixture of various variances, such as density, partial overlap, large noise and outliers, viewpoint changing. In this paper, an algorithm is proposed to align cross-source point clouds with both high accuracy and high efficiency. There are two main contributions: firstly, two components, the weak region affinity and pixel-wise refinement, are proposed to maintain the global and local information of 3D point clouds. Then, these two components are integrated into an iterative tensor-based registration algorithm to solve the cross-source point cloud registration problem. We conduct experiments on a synthetic cross-source benchmark dataset and real cross-source datasets. Comparison with six state-of-the-art methods, the proposed method obtains both higher efficiency and accuracy.
Huang, X, Zhang, H, Zhang, JA, Guo, YJ, Song, RL, Xu, XF, Wang, CT, Lu, Z & Wu, W 2019, 'Dual pulse shaping transmission with complementary nyquist pulses', IEEE Vehicular Technology Conference, Vehicular Technology Conference, IEEE, Honolulu, HI, USA.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The concept of complementary Nyquist pulse is introduced in this paper. Making use of a half rate Nyquist pulse and its complementary one, a dual pulse shaping transmission scheme is proposed, which achieves full Nyquist rate transmission with only a half of the sampling rate required by conventional Nyquist pulse shaping. This is essential for realizing high-speed digital communication systems with available and affordable data conversion devices. The condition for cross-symbol interference free transmission with the proposed dual pulse shaping is proved in theory, and two classes of ideal complementary Nyquist pulses are formulated assuming raised-cosine pulse shaping. Simulation results are also presented to demonstrate the improved spectral efficiency with dual pulse shaping and compare other system performance against conventional Nyquist pulse shaping.
Huang, X, Zhang, JA, Liu, R & Guo, YJ 2019, 'Novel Architecture and Key Technologies for Achieving High Capacity and Low Cost Space and Terrestrial Integrated Networks', 2019 19th Australian Space Research Conference, Adelaide, SA, Australia.
Huang, Y, Wu, Q, Xu, J & Zhong, Y 2019, 'Celebrities-ReID: A Benchmark for Clothes Variation in Long-Term Person Re-Identification', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Budapest, Hungary.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper considers person re-identification (re-ID) in the case of long-time gap (i.e., long-term re-ID) that concentrates on the challenge of clothes variation of each person. We introduce a new dataset, named Celebrities-reID to handle that challenge. Compared with current datasets, the proposed Celebrities-reID dataset is featured in two aspects. First, it contains 590 persons with 10,842 images, and each person does not wear the same clothing twice, making it the largest clothes variation person re-ID dataset to date. Second, a comprehensive evaluation using state of the arts is carried out to verify the feasibility and new challenge exposed by this dataset. In addition, we propose a benchmark approach to the dataset where a two-step fine-tuning strategy on human body parts is introduced to tackle the challenge of clothes variation. In experiments, we evaluate the feasibility and quality of the proposed Celebrities-reID dataset. The experimental results demonstrate that the proposed benchmark approach is not only able to best tackle clothes variation shown in our dataset but also achieves competitive performance on a widely used person re-ID dataset Market1501, which further proves the reliability of the proposed benchmark approach.
Huang, Y, Wu, Q, Xu, J & Zhong, Y 2019, 'SBSGAN: Suppression of Inter-Domain Background Shift for Person Re-Identification', IEEE International Conference on Computer Vision (ICCV), Seoul, South Korea.
Islam, MR, Lu, H, Fang, G, Li, L & Hossain, MJ 2019, 'Optimal Dispatch of Electrical Vehicle and PV Power to Improve the Power Quality of an Unbalanced Distribution Grid', 2019 Proceedings of International Conference on High Performance Big Data and Intelligent Systems (HPBD&IS), 2019 International Conference on High Performance Big Data and Intelligent Systems (HPBD&IS), IEEE, Shenzhen, China.
View/Download from: Publisher's site
Jauregi Unanue, I, Zare Borzeshi, E, Esmaili, N & Piccardi, M 2019, 'ReWE: Regressing Word Embeddings for Regularization of Neural Machine Translation Systems', Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), Conference of the North American Chapter of the Association for Computational Linguistics, Association for Computational Linguistics, Minneapolis, USA, pp. 430-436.
View/Download from: Publisher's site
View description>>
Regularization of neural machine translation is still a significant problem, especially in low-resource settings. To mollify this problem, we propose regressing word embeddings (ReWE) as a new regularization technique in a system that is jointly trained to predict the next word in the translation (categorical value) and its word embedding (continuous value).
Such a joint training allows the proposed system to learn the distributional properties represented by the word embeddings, empirically improving the generalization to unseen sentences. Experiments over three translation datasets have showed a consistent improvement over a strong baseline, ranging between 0.91 and 2.54 BLEU points, and also a marked
improvement over a state-of-the-art system.
Ji, LY, Qin, PY, Guo, YJ, Genovesi, S, Zhu, HL & Zong, Y 2019, 'A Reconfigurable Partially Reflective Surface Antenna with Enhanced Beam Steering Capability', 13th European Conference on Antennas and Propagation, EuCAP 2019, European Conference on Antennas and Propagation, IEEE, Krakow, Poland.
View description>>
© 2019 European Association on Antennas and Propagation. A reconfigurable partially reflective surface (PRS) antenna with improved beam steering capability is proposed in this paper. Compared with our previous paper, the beam-steering angle can be enhanced from ±5° to ±17° with less active elements and a much smaller gain variation. It is realized by employing a compact reconfigurable metasurface as the PRS structure, which is located atop a probe-fed square patch antenna. A prototype antenna operating at 5.5 GHz is fabricated and measured. Good agreement between the simulated and measured results for the input reflection coefficients and radiation patterns is achieved, which validates the feasibility of the design principle.
Jimenez, KJP, Hora, JA, Gerasta, OJL, Zhu, X & Dutkiewicz, E 2019, 'Self-biased 2.4 GHz CMOS RF-to-DC converter with 80% efficiency and -22.04 dBm sensitivity for Wi-Fi energy harvesting', 2019 4th IEEE International Circuits and Systems Symposium, ICSyS 2019.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. One of the significant disadvantages of RF energy harvesting is having a low power density in comparison to other ambient energy sources. The rectifier is the core of an RF energy harvesting system since it converts and boosts weak RF power to usable DC power. This study introduces a design of an efficiency enhanced RF-to-DC power converter with impedance matched to ±50Ω. The circuit design is based on two circuit design architectures, namely: the fully cross-coupled rectifier and self-biased technique, and the combined with LC matching circuit to obtain high power conversion efficiency. The design simulation is implemented using 65 nm CMOS Technology process. The performance of the proposed circuit design has achieved a peak power conversion efficiency of 80% at -14.4 dBm with a minimum input power of -22.04 dBm at 2.4 GHz for a load resistance of 20 kΩ and 10 pF load.
Karmokar, DK, Chen, SL & Guo, YJ 2018, 'Novel Continuous Beam Scanning Leaky-Wave Antennas Using 1-D Mushroom Structure', 2018 International Symposium on Antennas and Propagation (ISAP), International Symposium on Antennas and Propagation, IEEE, Korea, pp. 753-754.
View description>>
© 2018 KIEES. Novel continuous backward-to-forward beam-scanning leaky-wave antennas (LWAs) are designed using a 1-D mushroom structure. An effective method is proposed to suppress the bandgap of a mushroom structure. A smooth transition between the backward and forward leaky modes is achievable by choosing a suitable value of the via inductance, and hence the antenna has design flexibility. The study starts from an equivalent circuit of a unit cell and is verified through simulation and measurement. The measured results confirm a continuous 126° beam scan, starting from -60°, with less than 3 dB gain variation. Moreover, the measured 3dB gain bandwidth is over 58%, which is better than most of the reported LWAs.
Karmokar, DK, Chen, S-L, Qin, P-Y & Guo, YJ 2019, 'Open-Stopband Suppression and Cross-Polarization Reduction of a Substrate Integrated Waveguide Leaky-Wave Antenna', 2019 URSI ASIA-PACIFIC RADIO SCIENCE CONFERENCE (AP-RASC), URSI Asia-Pacific Radio Science Conference (AP-RASC), IEEE, New Delhi, INDIA.
Kempegowda, SM & Chaczko, Z 2018, 'Essential Skill of Enterprise Architect Practitioners for Digital Era', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia, pp. 1-5.
View/Download from: Publisher's site
View description>>
The technology landscape has evolved from
Mainframe to Digital platform. In this paper, we are proposing
the skills that are essential for an Enterprise Architect to be
successful in the Digital Era.
Kempegowda, SM & Chaczko, Z 2018, 'Industry 4.0 Complemented with EA Approach: A Proposal for Digital Transformation Success', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
Manufacturing industry based on steam know as
Industry 1.0 is evolving to Industry 4.0 a digital ecosystem
consisting of an interconnected automated system with realtime
data. This paper investigates and proposes, how
the digital ecosystem complemented with Enterprise
Architecture practice will ensure the success of digital
transformation.
Kempegowda, SM & Chaczko, Z 2018, 'The optimum number of Principles ideal for Enterprise Architecture practice', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia, pp. 1-4.
View/Download from: Publisher's site
View description>>
Every organization defines Principles for Enterprise Architecture (EA) practice. As there is no set standard, the principles identified exceeds the recommended number 20 by TOGAF. More the number of Principles defined it will be ignored by the Enterprise Architects instead referring for their decision making. In this paper, we identify the ideal number of principles that will motivate Architects to refer to perform their task
Kong, X, Fang, G, Liu, L & Tran, T 2019, 'Low Computational Data Fusion Approach Using INS and UWB for UAV Navigation Tasks in GPS Denied Environments', 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies (PDCAT), International Conference on Parallel and Distributed Computing, Applications and Technologies, IEEE, Gold Coast, Australia.
View/Download from: Publisher's site
View description>>
This paper presents a low computational approach for unmanned aerial vehicles (UAV) navigation in GPS-denied environments. This approach is aiming to reduce computation load for UAV flying mission constraints. Small size, light weight on board hardware are constraints for UAV deployment and flying missions. The on board processor should not be built with high complexity and should consume as little computing as possible. Most existing approaches use Kalman filter, extended Kalman filter, Unscented filter, or particle filter to fuse different types of onboard sensor data to estimate UAV position. We developed a data fusion architecture that does not use these filters. We use an ultra-light-coupling fusion architecture. In this architecture, primary sensor and secondary sensor data are fused. When the secondary sensor is unavailable in most of the time, the UAV navigation uses the output of the primary sensor. When the secondary sensor signal is available, the primary sensor is re-aligned using the secondary sensor signal to bond the errors. In our approach, the primary sensor is Inertial Measurement Unit (IMU), and the secondary sensor inputs are from Ultra-wideband system (UWB). This approach is validated using demonstration of comparison of computing load, and simulation results for accuracy and reliability testing using UAV flying mission scenario.
Le, AT, Nan, Y, Tran, LC, Huang, X, Guo, YJ & Vardaxoglou, JCY 2018, 'Analog Least Mean Square Loop for Self-Interference Cancellation in Generalized Continuous Wave SAR', In Proceedings of the IEEE 88th Vehicular Technology Conference, IEEE Vehicular Technology Conference, IEEE, Chicago, US, pp. 1-5.
View/Download from: Publisher's site
View description>>
Generalized continuous wave synthetic aperture radar (GCW-SAR) is a promising new imaging radar system since it applies the full-duplex (FD) transmission technique to achieve continuous signaling in order to overcome several fundamental limitations of the conventional pulsed SARs. As in any FD wireless communication system, self-interference (SI) is also a key problem which can impact on the GCW-SAR system. In this paper, the analog least mean square (ALMS) loop in the radio frequency domain is adopted to cancel the SI for a GCW-SAR system with periodic chirp signaling. The average residual SI power after the ALMS loop is analyzed theoretically by a stationary analysis. It is found that the ALMS loop not only works with random signals in general FD communication systems, but also works well with the periodic signal in GCW-SAR systems. Simulation results show that over 45 dB SI cancellation can be achieved by the ALMS loop which ensures the proper operation of the GCW-SAR system.
Le, AT, Tran, LC, Huang, X & Guo, YJ 2019, 'Beam-Based Analog Self-Interference Cancellation with Auxiliary Transmit Chains in Full-Duplex MIMO Systems', The IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) 2019, IEEE, Cannes, France, pp. 1-5.
Li, D, Tang, MC, Chen, X, Wang, Y, Hu, KZ & Ziolkowski, RW 2019, 'Design of Planar, Wideband, Multi-Polarization Reconfigurable Filtering Antenna', 2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Proceedings, IEEE MTT-S International Wireless Symposium, IEEE, Guangzhou, China.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A planar, wideband, multi-polarization reconfigurable filtering antenna is proposed. By controlling the state of PIN diodes, the operation states of the filtering antenna can be switched among linear polarization (LP), left-hand circular polarization (LHCP) and right-hand circular polarization (RHCP). The filtering antenna consists of a driven patch, a parasitic patch, and an established reconfigurable feed network. Two radiation nulls can be generated at the lower and upper band-edge in all operation states through interaction between the driven patch and feeding network. A pair of U-slots is etched on the driven patch to obtain the extra radiation null to further improve the sharp roll-off rate at the upper band-edge. The optimized prototype was fabricated and tested. The measured results, in good agreement with the simulated values, demonstrate that the filtenna has a wide -10dB fractional bandwidth: 15.6%, a realized gain value over operational bandwidth of 7.7 ± 0.5 dBi, and stable radiation performance. Furthermore, the filtenna presents a good out-of-band rejection level.
Li, H, Wang, TQ, Huang, X & Zhang, JA 2019, 'Enhanced AoA estimation using localized hybrid dual-polarized arrays', IEEE Vehicular Technology Conference, Vehicular Technology Conference, IEEE, Honolulu, HI, USA.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. With balanced system performance, implementation complexity and hardware cost, hybrid antenna array is regarded as an enabling technology for massive multiple-input and multiple-output communication systems in millimeter wave (mmWave) frequencies. Angle-of-arrival (AoA) estimation using a localized hybrid array faces the challenges of the phase ambiguity problem due to its localized nature of array structure and susceptibility to noises. This paper discusses AoA estimation in an mmWave system employing dual-polarized antennas. We propose an enhanced AoA estimation algorithm using a localized hybrid dual-polarized array for a polarized mmWave signal. First, the use of dual-polarized arrays effectively strengthens the calibration of differential signals and resulting signal-to-noise ratio with coherent polarization combining, leading to an enhanced estimate of the phase offset between adjacent subarrays. Second, given the phase offset, an initial AoA estimate can be obtained, which is used to update the phase offset. By employing the updated one, the AoA is re- estimated with improved accuracy. The closed-form mean square error (MSE) lower bounds of AoA estimation are derived and compared with simulated MSEs. The simulation results show that the proposed algorithm in combination with hybrid dual- polarized arrays significantly improves the estimation accuracy compared with the state of the art.
Li, Q, Wu, Q & Liu, X 2019, 'Multi-scale and Hierarchical Embedding for Polarity Shift Sensitive Sentiment Classification', Artificial Intelligence and Security, International Conference on Artificial Intelligence and Security, Springer, New York, NY, USA, pp. 227-238.
View/Download from: Publisher's site
View description>>
© 2019, Springer Nature Switzerland AG. Appropriate paragraph embedding is critical for sentiment classification. However, the embedding for paragraph with polarity shift is very challenging and insufficiently explored. In this paper, a MUlti-Scale and Hierarchical embedding method, MUSH, is proposed to learn a more accurate paragraph embedding for polarity shift sensitive sentiment classification. MUSH adopts CNN with multi-size filters to reveal multi-scale sentiment atoms and utilizes hierarchical multi-line CNN-RNN structures to simultaneously capture polarity shift in both sentence level and paragraph level. Extensive experiments on four large real-world data sets demonstrate that the MUSH-enabled sentiment classification significantly enhances the accuracy compared with three state-of-the-art and four baseline competitors.
Li, Q, Wu, Q, Zhu, C & Zhang, J 2019, 'Bi-level Masked Multi-scale CNN-RNN Networks for Short Text Representation', 2019 International Conference on Document Analysis and Recognition (ICDAR), International Conference on Document Analysis and Recognition, IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
Representing short text is becoming extremely important for a variety of valuable applications. However, representing short text is critical yet challenging because it involves lots of informal words and typos (i.e. the noise problem) but only a few vocabularies in each text (i.e. the sparsity problem). Most of the existing work on representing short text relies on noise recognition and sparsity expansion. However, the noises in short text are with various forms and changing fast, but, most of the current methods may fail to adaptively recognize the noise. Also, it is hard to explicitly expand a sparse text to a high-quality dense text. In this paper, we tackle the noise and sparsity problems in short text representation by learning multi-grain noise-tolerant patterns and then embedding the most significant patterns in a text as its representation. To achieve this goal, we propose a bi-level multi-scale masked CNN-RNN network to embed the most significant multi-grain noise-tolerant relations among words and characters in a text into a dense vector space. Comprehensive experiments on five large real-world data sets demonstrate our method significantly outperforms the state-of-the-art competitors.
Li, Q, Wu, Q, Zhu, C, Zhang, J & Zhao, W 2019, 'An Inferable Representation Learning for Fraud Review Detection with Cold-start Problem', 2019 International Joint Conference on Neural Networks (IJCNN), International Joint Conference on Neural Networks, IEEE, Budapest, Hungary.
View/Download from: Publisher's site
View description>>
Fraud review significantly damages the business reputation and also customers' trust to certain products. It has become a serious problem existing on the current social media. Various efforts have been put in to tackle such problems. However, in the case of cold-start where a review is posted by a new user who just pops up on the social media, common fraud detection methods may fail because most of them are heavily depended on the information about the user's historical behavior and its social relation to other users, yet such information is lacking in the cold-start case. This paper presents a novel Joint-bEhavior-and-Social-relaTion-infERable (JESTER) embedding method to leverage the user reviewing behavior and social relations for cold-start fraud review detection. JESTER embeds the deep characteristics of existing user behavior and social relations of users and items in an inferable user-item-review-rating representation space where the representation of a new user can be efficiently inferred by a closed-form solution and reflects the user's most probable behavior and social relations. Thus, a cold-start fraud review can be effectively detected accordingly. Our experiments show JESTER (i) performs significantly better in detecting fraud reviews on four real-life social media data sets, and (ii) effectively infers new user representation in the cold-start problem, compared to three state-of-the-art and two baseline competitors.
Li, Q, Wu, Q, Zhu, C, Zhang, J & Zhao, W 2019, 'Unsupervised User Behavior Representation for Fraud Review Detection with Cold-Start Problem', Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, China, pp. 222-236.
View/Download from: Publisher's site
Li, Z, Gong, Y, Zhang, J, Yi, J, Wu, Q & Kirsch, C 2019, 'Sample adaptive multiple kernel learning for failure prediction of railway points', Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, ACM, Anchorage AK USA, pp. 2848-2856.
View/Download from: Publisher's site
View description>>
© 2019 Association for Computing Machinery. Railway points are among the key components of railway infrastructure. As a part of signal equipment, points control the routes of trains at railway junctions, having a significant impact on the reliability, capacity, and punctuality of rail transport. Meanwhile, they are also one of the most fragile parts in railway systems. Points failures cause a large portion of railway incidents. Traditionally, maintenance of points is based on a fixed time interval or raised after the equipment failures. Instead, it would be of great value if we could forecast points' failures and take action beforehand, min-imising any negative effect. To date, most of the existing prediction methods are either lab-based or relying on specially installed sensors which makes them infeasible for large-scale implementation. Besides, they often use data from only one source. We, therefore, explore a new way that integrates multi-source data which are ready to hand to fulfil this task. We conducted our case study based on Sydney Trains rail network which is an extensive network of passenger and freight railways. Unfortunately, the real-world data are usually incomplete due to various reasons, e.g., faults in the database, operational errors or transmission faults. Besides, railway points differ in their locations, types and some other properties, which means it is hard to use a unified model to predict their failures. Aiming at this challenging task, we firstly constructed a dataset from multiple sources and selected key features with the help of domain experts. In this paper, we formulate our prediction task as a multiple kernel learning problem with missing kernels. We present a robust multiple kernel learning algorithm for predicting points failures. Our model takes into account the missing pattern of data as well as the inherent variance on different sets of railway points. Extensive experiments demonstrate the superiority of our algorithm compare...
Liao, Q, Holewa, H, Xu, M & Wang, D 2018, 'Fine-Grained Categorization by Deep Part-Collaboration Convolution Net', 2018 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2018, Digital Image Computing: Techniques and Applications, IEEE, Australia.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. In part-based categorization context, the ability to learn representative feature from quantitative tiny object parts is of similar importance as to exactly localize the parts. We propose a new deep net structure for fine-grained categorization that follows the taxonomy workflow, which makes it interpretable and understandable for humans. By training customized sub-nets on each manually annotated parts, we increased the state-of-the-art part-based classification accuracy for general fine-grained CUB-200-2011 dataset by 2.1%. Our study shows the proposed method can produce more activation to discriminate detail part difference while maintaining high computing performance by applying a set of strategies to optimize the deep net structure.
Liao, Y, Qiu, B, Su, Z, Wang, R & He, X 2019, 'Learning transmission filtering network for image-based pm2.5 estimation', Proceedings - IEEE International Conference on Multimedia and Expo, IEEE International Conference on Multimedia and Expo, IEEE, Shanghai, China, pp. 266-271.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. PM2.5 is an important indicator of the severity of air pollution and its level can be predicted through hazy photographs caused by its degradation. Image-based PM2.5 estimation is thus extensively employed in various multimedia applications but is challenging because of its ill-posed property. In this paper, we convert it to the problem of estimating the PM2.5-relevant haze transmission and propose a learning model called the transmission filtering network. Different from most methods that generate a transmission map directly from a hazy image, our model takes the coarse transmission map derived from the dark channel prior as the input. To obtain a transmission map that satisfies the local smoothness constraint without regional boundary degradation, our model performs the edge-preserving smoothing filtering as the refinement on the map. Moreover, we introduce the attention mechanism to the network architecture for more efficient feature extraction and smoothing effects in the transmission estimation. Experimental results prove that our model performs favorably against the state-of-the-art dehazing methods in a variety of hazy scenes.
Liberal, I, Ederra, I & Ziolkowski, RW 2019, 'Designing the bandwidth of single-photon sources with classical antenna techniques', 2019 13th European Conference on Antennas and Propagation (EuCAP), European Conference on Antennas and Propagation, IEEE, Krakow, Poland.
View description>>
© 2019 European Association on Antennas and Propagation. We discuss the role of the classical electromagnetic theory concept of reactive interactions on determining the bandwidth of a single-photon source. Typically, quantum emitters operate in the weak-coupling regime where the bandwidth of the emission spectrum is simply proportional to their decay rates. However, we introduce a first-order correction to the emission spectrum, demonstrating that its bandwidth is also directly affected by the dispersion properties of the reactive interactions of the quantum emitter with its environment. This correction is particularly important in the intermediate region bridging the weak and strong coupling regimes. As an example of the applicability of this theory, we study the behaviour of a quantum emitter decaying through a coupled two-cavity system. Our results suggests that this setup could be utilized for the design of efficient, but narrowband single-photon sources.
Lin, JY, Wong, SW, Yang, Y & Zhu, L 2019, 'Cavity Balanced-to-Unbalanced Magic-T with Filtering Response', IEEE MTT-S International Microwave Symposium Digest, IEEE MTT-S International Microwave Symposium, IEEE, Boston, MA, USA, pp. 444-447.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this paper, a design of balanced-to-unbalanced magic-T with filtering response is proposed for the first time. The function of the proposed magic-T is composed of an in-phase balanced-to-unbalanced power divider and an out-of-phase balanced-to-unbalanced power divider. Three fundamental modes, namely, TE011, TE101, and TM110, of the triple-mode resonators (TMRs) are excited to provide the odd- and even-symmetric field distributions so that in-phase and out-of-phase responses can be achieved. It is noteworthy that the common-mode suppression can be achieved at the balanced ports, while high isolation is achieved at the single-end ports. To verify the concept, the proposed balanced-to-unbalanced magic-T structure is fabricated and measured. Good matching between simulated and measured results shows the validity and accuracy of the proposed design methodology.
Lin, W & Ziolkowski, RW 2019, 'A new approach to design high directivity, compact omnidirectional CP antenna arrays', 2019 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting, APSURSI 2019 - Proceedings, pp. 2157-2158.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A new approach to design high directivty, compact omnidirectional circularly polarized (OCP) antenna arrays is presented in this paper. An array of collinear electric (E-) and magnetic (M-) radiators is realized by simply cascading copper loops and vertical strips to form an electrically long antenna structure. This array has two identical half sections that is excited in its center. The circumference of each loop and the length of each vertical strip are about a half-wavelength. The configuration facilitates the currents on all of the loop and the strip radiators to achieve the same phase and, hence, the array is a set of in-phase E-and M-radiators. Due to the fact that the phases of the magnetic radiator currents are 90° ahead of the electric ones, the 90° phase difference between the two subsets of radiators required to achieve OCP radiation is realized. An optimized prototype was fabricated and measured. The whole structure is compact and easily fabricated. It covers a 130 MHz bandwidth from 2.34 to 2.47 GHz and produces OCP radiation. The peak measured LHCP realized gain for the four-stage version is 5.1 dBic.
Lin, W & Ziolkowski, RW 2018, 'Dual-Functional Electrically Small Huygens Antenna System', ISAP 2018 - 2018 International Symposium on Antennas and Propagation, International Symposium on Antennas and Propagation, IEEE, South Korea.
View description>>
© 2018 KIEES. A dual-functional electrically small Huygens antenna system is introduced for wireless power transfer (WPT) and communication applications at 915 MHz. This dual-functional system is facilitated by using two orthogonally-oriented electrically small Huygens linearly polarized (HLP) dipole subsystems. Each HLP antenna consists of two metamaterial-inspired near field resonant parasitic (NFRP) elements, i.e., the capacitive loaded loop (CLL) and the Egyptian axe dipole (EAD). A rectifier circuit is integrated with one of the HLP antennas to facilitate its function as a rectenna for WPT applications. The other HLP antenna serves the communication applications. Due to the large isolation (> 30 dB) between these two HLP subsystems, their functionalities are independent. A successfully fabricated and measured prototype demonstrates that this highly-integrated dual-functional antenna system has excellent performance characteristics. It is electrically-small (ka < 0.77) and produces unidirectional Huygens (cardioid) broadside realized gain patterns with broad beamwidths. It has a high AC to DC conversion efficiency. The antenna system is an excellent, practical candidate for wireless Internet-of-Things (IoT) applications.
Lin, W & Ziolkowski, RW 2019, 'Electrically Small, Highly Efficient, Huygens Circularly Polarized Rectenna for Wireless Power Transfer Applications', 13th European Conference on Antennas and Propagation, EuCAP 2019, European Conference on Antennas and Propagation, IEEE, Krakow, Poland.
View description>>
© 2019 European Association on Antennas and Propagation. This paper introduces the first realized electrically small, highly efficient, Huygens circularly polarized (HCP) rectenna for wireless power transfer applications. It is designed to operate at the 915 MHz in the corresponding ISM band. It is realized through the seamless integration of an electrically small HCP antenna and a highly efficient rectifier circuit. The electrically small HCP antenna consists of four electrically small near-field resonant parasitic (NFRP) elements: two Egyptian axe dipoles (EADs) and two capacitively loaded loops (CLLs). The rectifier is a full-wave rectifying circuit based on HSMS286C diodes. It is integrated with the HCP antenna on its bottom layer via a coplanar stripline (CPS) without occupying any additional space. A HCP rectenna prototype was successfully fabricated and tested. It is electrically small (ka < 0.77) with a radius R = λ0 / 8 and a height H = λ0 / 25. Excellent CP radiation absorption capacity is observed. The measured peak AC to DC conversion efficiency reaches 82%. Being electrically small and highly efficient, the reported HCP rectenna is an ideal candidate for wirelessly powering internet-of-things (IoT) devices in many emerging IoT applications.
Liu, F, Liu, Y, Guo, YJ & Liu, QH 2018, 'Synthesis of Rotated Sparse Linear Dipole Array with Shaped Power Pattern', 2018 International Applied Computational Electromagnetics Society Symposium in China, ACES-China 2018, International Applied Computational Electromagnetics Society Symposium - China, IEEE, Beijing, China.
View/Download from: Publisher's site
View description>>
© 2018 ACES. A new shaped pattern synthesis method is presented in which element rotations, positions and phases are co-optimized to produce a shaped beam pattern for a sparse dipole array. Compared with conventional shaped pattern synthesis using excitation amplitude and phase optimization, the proposed method can not only reduce the number of elements But also avoid the usage of unequal power dividers. A synthesis example is provided to verify the performance of the proposed method.
Liu, H, Zhu, X, Lu, M & Yeo, KS 2019, 'Design of a Voltage-Controlled Programmable-Gain Amplifier in 65-nm CMOS Technology', 2019 IEEE MTT-S International Microwave Symposium (IMS), 2019 IEEE/MTT-S International Microwave Symposium, IEEE, Boston, USA, pp. 87-90.
View/Download from: Publisher's site
View description>>
A voltage-controlled programmable-gain amplifier (VC-PGA) is designed in this work. The power consumption of the VC-PGA is binary-weighted. In contrast to conventional PGAs, the gain step of the designed PGA can be continuously tuned by a control voltage. To prove the concept, an analog baseband chain is implemented in 65 nm CMOS technology, which consists of a switchable-order filter with the VC-PGA. The measurements show that the frequency responses can be configured as either 5 th or 7 th order with 16 gain steps. The bandwidth is approximately 50 MHz for all cases and the gain step can be continuously tuned between 0 and 3 dB. The core area is only 0.18 μm 2 .
Liu, J, Chaczko, Z, Braun, R & Gudzbeler, G 2019, 'Collaborative RFID agent simulation in dynamic environment', 2019 18th International Conference on Information Technology Based Higher Education and Training, ITHET 2019, International Conference on Information Technology Based Higher Education and Training, IEEE, Magdeburg, Germany, Germany, pp. 1-4.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper discusses the design and implementation process of applying morphogenetic and reflexive agents models into the software architecture of radio frequency identification (RFID) agent simulation environment framework (RMAP), as well as, illustrate the agent internal interaction mechanism, and thus allowing multiple agents to work collaboratively to achieve the same goal at least effort. The RMAP can be extensively used for training of medical professionals, developers, students, first responders, emergency services, etc. The RMAP simulation framework is an attempt to synthesize on technological issues related to RFID based solutions in order to 'train' and 'support' medical practitioners and developers.
Liu, Y, Luo, Q, Li, M & Guo, YJ 2019, 'Thinned Massive Antenna Array for 5G Millimeter-Wave Communications', 13th European Conference on Antennas and Propagation, EuCAP 2019, European Conference on Antennas and Propagation, IEEE, Krakow, Poland.
View description>>
© 2019 European Association on Antennas and Propagation. massive antenna array is one of the key technologies for 5G millimeter-wave communications. In this paper, a modified iterative FFT is introduced to obtain thinned massive arrays. An example is given for synthesizing a 128-element thinned array with U-slot microstrip antenna working at 27.5-28.5 GHz. Simulated results show that the thinned array has improved beam resolution and sidelobe performance than those for a conventional 128-element array.
Luo, Q, Liu, Y, Liu, F, Ren, Y & Guo, YJ 2018, 'Fast Synthesis Algorithm for Uniformly Spaced Circular Array with Low Sidelobe Pattern', 2018 International Applied Computational Electromagnetics Society Symposium in China, ACES-China 2018, International Applied Computational Electromagnetics Society Symposium - China, IEEE, Beijing, China.
View/Download from: Publisher's site
View description>>
© 2018 ACES. In this paper, a highly efficient approach is proposed to synthesize the low sidelobe pattern of uniformly spaced circular array. The proposed approach can be generalized to deal with the pattern synthesis for the circular array with directional elements. Numerical examples are given to verify the effectiveness and advantage of this approach.
Luo, Y, Yan, B, Canning, J, Tafti, G, Wang, S, Wang, W, Tian, Y, Cook, K, Dhasarathan, V & Peng, GD 2019, 'A novel spun photonic crystal fibre with amoeba shape', Optics InfoBase Conference Papers.
View description>>
Asia Communications and Photonics Conference (ACP) © OSA 2019 © 2019 The Author(s) A novel spun photonic crystal fibre with Amoeba shape has been fabricated based on asymmetric self-pressurization. Asymmetry growth dynamics with fibre drawing has been investigated.
Luo, Y, Zhang, JA, Huang, S, Pan, J & Huang, X 2019, 'Quantization with Combined Codebook for Hybrid Array using Two-Phase-Shifter Structure', IEEE International Conference on Communications, IEEE International Conference on Communications, IEEE, Shanghai, China.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. We propose a novel joint quantization scheme for hybrid antenna array systems using the two-phase-shifter (2-PS) structure, where two phase shifters are combined to represent one beamforming weight. Conventional quantization using a single phase shifter for each beamforming weight cannot represent the magnitude. We propose a new codebook design that combines the two codebooks of the two phase shifters in the recently proposed 2-PS structure. We also study the scaling problem of the beamforming vector and propose a low-complexity searching algorithm for finding a near-optimal scalar based on element-wise quantization. The mean squared quantization error and signal-to-noise ratio (SNR) degradation are derived analytically. Simulation results validate the accuracy of the analytical results and the effectiveness of the proposed quantization methods.
Makhdoom, I, Zhou, I, Abolhasan, M, Lipman, J & Ni, W 2019, 'PrivySharing: A Blockchain-based framework for integrity and privacy-preserving data sharing in Smart Cities', Proceedings of the 16th International Joint Conference on e-Business and Telecommunications - (Volume 2), International Conference on Security and Cryptography, Scitepress, Prague, Czech Republic, pp. 363-371.
View/Download from: Publisher's site
View description>>
Copyright © 2019 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved The ubiquitous use of Internet of Things (IoT) ranges from industrial control systems to e-Health, e-commerce, smart cities, supply chain management, smart cars, cyber-physical systems and a lot more. However, the data collected and processed by IoT systems especially the ones with centralized control are vulnerable to availability, integrity, and privacy threats. Hence, we present “PrivySharing,” a blockchain-based innovative framework for integrity and privacy-preserving IoT data sharing in a smart city environment. The proposed scheme is distinct from existing technologies on many aspects. The data privacy is preserved by dividing the blockchain network into various channels, where every channel processes a specific type of data such as health, smart car, smart energy or financial data. Moreover, access to user data within a channel is controlled by embedding access control rules in the smart contracts. In addition, users' data within a channel is further isolated and secured by using private data collection. Likewise, the REST API that enables clients to interact with the blockchain network has dual security in the form of an API Key and OAuth 2.0. The proposed solution also conforms to some of the significant requirements outlined in the European Union General Data Protection Regulation. Lastly, we present a system of reward in the form of a digital token “PrivyCoin” for the users for sharing their data with the stakeholders/third parties.
Malik, N, Nanda, P, He, X & Liu, RP 2019, 'Trust and Reputation in Vehicular Networks: A Smart contract-based approach', 18th IEEE International Conference on Trust, Security and Privacy in Computing and Communications, Trustcom 2019, Rotorua, New Zealand.
Mohammed, F, He, X, Chen, J & Lin, Y 2019, 'A Novel Model for Classification of Parkinson’s Disease: AccuratelyIdentifying Patients for Surgical Therapy', HICSS, Hawaii International Conference on System Sciences, USA, pp. 3741-3750.
View/Download from: Publisher's site
View description>>
Parkinson’s disease (PD) is a neurodegenerative disorder and a global health problem that has no curative therapies. Surgery is a well-established therapy for controlling symptoms of advanced PD patients. This paper proposes a streamlined model to classify PD and to identify appropriate patients for surgical therapy. The data was gathered from the
Parkinson’s Progressive Markers Initiative consisting of 1080 subjects. Multilayer Perceptron (MLP), Decision trees, Support Vector Machine and Na¨ıve Bayes are used as classifiers. MLP achieves the highest accuracy as compared to other three classifiers. The dataset used in our experiments is from the Parkinson Progressive Markers Initiative. With feature selection, it is observed that the same classification accuracy is achieved with 60% of the attributes as that using all attributes. It
is demonstrated that our classification model for PD patients produces the most accurate results and achieves the highest accuracy of 98.13%
More, FJ & Chaczko, Z 2018, 'Non-invasive Methods in the Detection of Coronary Artery Disease', 2018 26th International Conference on Systems Engineering (ICSEng), International Conference on Systems Engineering, IEEE, Sydney, Australia, pp. 1-5.
View/Download from: Publisher's site
View description>>
Coronary Artery Disease (CAD) is the prime causal factor in cardiovascular disease in the 21st century throughout the world. In Australia, CAD related diseases result in 12% morbidity and mortality rate. This paper summarizes the non-invasive methods of diagnosis of CAD. The association between medical science and biomedical engineering has led to the development of non-invasive methods of diagnosis of CAD. The use of new technology that exploits IoT and Body Area Networks using wearable sensor devices over the patient’s body and medical experts to diagnose CAD. Progression of clinical assessment, diagnosis, and evaluation of CAD have been achieved in the last decade. The current treatment plan for CAD focused on clinical prevention, surgical or a combination of both depending on the severity of disease. The analysis of coronary artery disease, chest pain, and various things involved in the assessment of patient’s history with relieving factors such as risk stratification and non-invasive tests used in diagnosis of CAD.
Nguyen, MH, Hà, MH, Hoang, DT, Nguyen, DN, Dutkiewicz, E & Tran, TT 2019, 'An efficient algorithm for the k-dominating set problem on very large-scale networks (extended abstract)', Computational Data and Social Networks (LNCS), International Conference on Computational Data and Social Networks, Springer, Ho Chi Minh City, Vietnam, pp. 74-76.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. The minimum dominating set problem (MDSP) aims to construct the minimum-size subset $$D \subset V$$ of a graph $$G = (V, E)$$ such that every vertex has at least one neighbor in D. The problem is proved to be NP-hard [5]. In a recent industrial application, we encountered a more general variant of MDSP that extends the neighborhood relationship as follows: a vertex is a k-neighbor of another if there exists a linking path through no more than k edges between them. This problem is called the minimum k-dominating set problem (MkDSP) and the dominating set is denoted as $$D:k$$. The MkDSP can be used to model applications in social networks [2] and design of wireless sensor networks [3]. In our case, a telecommunication company uses the problem model to supervise a large social network up to 17 millions nodes via a dominating subset in which k is set to 3.
Nguyen, MH, Ha, MH, Nguyen, D, Dinh, H, Dutkiewicz, E & Tran, TT 2019, 'An Efficient Algorithm for the k-Dominating Set Problem on Very Large-Scale Networks', International Conference on Computational Data and Social Networks (CSoNet 2019), Ho Chi Minh city.
Nguyen, N-T, Van Huynh, N, Hoang, DT, Nguyen, DN, Nguyen, N-H, Nguyen, Q-T & Dutkiewicz, E 2019, 'Energy Management and Time Scheduling for Heterogeneous IoT Wireless-Powered Backscatter Networks', ICC 2019 - 2019 IEEE International Conference on Communications (ICC), ICC 2019 - 2019 IEEE International Conference on Communications (ICC), IEEE, Shanghai.
View/Download from: Publisher's site
Ni, Z, Zhang, JA, Yang, K, Gao, F & Gao, Z 2018, 'Codebook Based Minimum Subspace Distortion Hybrid Precoding for Millimeter Wave Systems', IEEE Globecom Workshops, Abu Dhabi, United Arab Emirates.
View/Download from: Publisher's site
Pham, T, Takalkar, M, Xu, M, Hoang, DT, Truong, HA, Dutkiewicz, E & Perry, S 2019, 'Airborne Object Detection Using Hyperspectral Imaging: Deep Learning Review', Computational Science and Its Applications – ICCSA 2019, International Conference on Computational Science and Its Applications, Springer, Saint Petersburg, Russia, pp. 306-321.
View/Download from: Publisher's site
View description>>
Hyperspectral images have been increasingly important in object detection applications especially in remote sensing scenarios. Machine learning algorithms have become emerging tools for hyperspectral image analysis. The high dimensionality of hyperspectral images and the availability of simulated spectral sample libraries make deep learning an appealing approach. This report reviews recent data processing and object detection methods in the area including hand-crafted and automated feature extraction based on deep learning neural networks. The accuracy performances were compared according to existing reports as well as our own experiments (i.e., re-implementing and testing on new datasets). CNN models provided reliable performance of over 97% detection accuracy across a large set of HSI collections. A wide range of data were used: a rural area (Indian Pines data), an urban area (Pavia University), a wetland region (Botswana), an industrial field (Kennedy Space Center), to a farm site (Salinas). Note that, the Botswana set was not reviewed in recent works, thus high accuracy selected methods were newly compared in this work. A plain CNN model was also found to be able to perform comparably to its more complex variants in target detection applications.
Pham, TT & Dutkiewicz, E 2019, 'Quantify Physiologic Interactions Using Network Analysis', Computational Science and Its Applications – ICCSA 2019, International Conference on Computational Science and Its Applications, Springer, Saint Petersburg, Russia, pp. 142-151.
View/Download from: Publisher's site
View description>>
© 2019, Springer Nature Switzerland AG. To better understand the neural interactions amongst human organ systems, this work provides a framework of data analysis to quantify forms of neural signalling. We explore network interactions among the human brain and motor controlling. The main objective of this work is to provoke unique challenges in the emerging Network Physiology field. The proposed method applies network analysis techniques including power coherence for connectivity discovering and correlation measurement for profiling relationships. We used a well-designed dataset of 50 subjects over 14 different scenarios for each individual. We found network models for these interactions and observed informative network behaviours. The information can be used to study impaired communications that can lead to dysfunction of organs or the entire system such as sepsis.
Poostchimohammadabadi, H & Piccardi, M 2019, 'A multi-constraint structured hinge loss for named-entity recognition', Proceedings of The 17th Annual Workshop of the Australasian Language Technology Association, Annual Workshop of the Australasian Language Technology Association, ACLWEB, Sydney, NSW, Australia, pp. 41-46.
View description>>
The negative log-likelihood or cross entropy is the usual training objective of NLP models owing to its versatility and empirical performance. However, training objectives which directly target the performance measure used
to evaluate the task have the potential to lead to higher empirical accuracy. For this reason, in this short paper we propose using a multi-constraint structured hinge loss as the training objective of a contemporary name identity recognition (NER) model. Experimental results over the challenging OntoNotes 5.0 dataset have shown that the proposed objective has been able to achieve an improvement of 0.62 CoNLL score points at a complete parity of testing set-up.
Qashlan, A, Nanda, P & He, X 2019, 'Automated Ethereum Smart Contract for Block chain based Smart Home Security', Automated Ethereum Smart Contract for Block Chain Based Smart Home Security, International Conference on Artificial Intelligence and Security, Springer, Jaipur, India, pp. 313-326.
View/Download from: Publisher's site
View description>>
Presence of Internet of Things (IoT) based applications has been increasing in various domains including transportation, logistics, health care, and smart homes. Such applications involve deploying an enormous number of IoT devices, which generally lacks from security and often associates several vulnerabilities. These IoT devices need to communicate and synchronize with each other, which also increase the security and privacy challenges. Traditional security models are based on centralized and often include complicated approaches which, tend to be inapplicable and have some limitations. Therefore, one proposed solution is to use blockchain technology which could provide decentralize, secure, and peer-to-peer networks. In this paper, private blockchain implementation using Ethereum smart contract is developed for the smart home to ensure only the home owner can access and monitor home appliances. Simple smart contracts are designed to allow devices to communicate without the need for trusted third party. Our prototype demonstrates three key elements of blockchain-based smart security solution for smart home applications such as smart contract, blockchain-based access control and performance evaluation of the proposed scheme.
Qin, C, Zhang, JA, Huang, X & Guo, YJ 2019, 'Angle-of-arrival acquisition and tracking via virtual subarrays in an analog array', IEEE Vehicular Technology Conference, Vehicular Technology Conference, IEEE, Honolulu, HI, USA.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Angle-of-arrival (AoA) estimation is a challenging problem for analog antenna arrays. Typical algorithms use beam scanning and sweeping, which can be time-consuming, and the resolution is limited to the scanning step. In this paper, we propose a virtual-subarray based AoA estimation scheme, which divides an analog array into two virtual subarrays and can obtain a direct AoA estimate from every two temporal measurements. We propose different subarray constructions which lead to different range and accuracy of estimation. We provide detailed beamforming vector designs for these constructions and provide a performance lower bound for the estimator. We also present how to apply the estimator to AoA acquisition and tracking. Simulation results demonstrate that the proposed scheme significantly outperforms existing ones when the signal-to-noise ratio is not very low.
Qureshi, S & Braun, R 2018, 'Dynamic light path establishment in switch fabric using openFlow', 26th International Conference on Systems Engineering, ICSEng 2018 - Proceedings, International Conference on Systems Engineering, IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. In today's optical networks, Light paths are established through Network Management System and Element Management System, which is a manual and cumbersome process. Paths are computed and pre-setup according to the known traffic d emands, a nd p rovisioning a ny new service takes time. Several efforts have recently been made to make the path establishment process dynamic, including the Software Defined N etworking a pproach. H owever all of this work has assumed fully interconnected fabrics of the network devices which is generally not the case. The task of end to end path establishment between network elements and through fabric are interrelated, and this task remains incomplete, without consideration of the switch's limitations. This work highlights for the first time the issue of path establishment dynamically through a switch fabric. The paper briefly e xplains t he p roblem a nd s uggests an SDN based solution using the OpenFlow protocol. This work contributes a represention of the basic framework which will be required to make the complete path setup process dynamic. The paper also includes an operational description.
Rahman, ML, Cui, PF, Zhang, JA, Huang, X, Guo, YJ & Lu, Z 2019, 'Joint Communication and Radar Sensing in 5G Mobile Network by Compressive Sensing', Proceedings - 2019 19th International Symposium on Communications and Information Technologies, ISCIT 2019, pp. 599-604.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. There is growing interest in integrating communication and radar sensing into one system. However, very limited results are reported on how to realize sensing using complicated mobile signals when joint communication and radar sensing (JCAS) is applied to mobile networks. This paper studies radar sensing using one-dimension (1D) to 3D compressive sensing (CS) techniques, referring to signals compatible with latest fifth generation (5G) new radio (NR) standard. We demonstrate that radio sensing using both downlink and uplink 5G signals can be realized with reasonable performance using these CS techniques, and highlight the respective advantages and disadvantages of these techniques.1
Saki, M, Abolhasan, M & Lipman, J 2019, 'A big sensor data offloading scheme in rail networks', IEEE Vehicular Technology Conference, IEEE Vehicular Technology Conference, IEEE, Kuala Lumpur, Malaysia, Malaysia.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this paper, we propose an offloading scheme to transfer massive stored sensor data from rolling stock to railway data centers. We apply a delayed offloading strategy for non-critical stored data assuming that the critical data has been already separated through an appropriate edge processing task and has been sent via a real-time communication such as cellular networks. We propose train stations as potential and feasible spots for data offloading via available wireless local area networks (WLAN) such as existing WiFi network at stations. Thus, stations will not only be the places of passenger exchange but also data exchange. We develop an analytical model customized for the proposed offloading strategy in rail applications. Then we validate the performance of our model through simulation in various scenarios in Omnet. The simulation results shows an accuracy of %98.67 for the proposed analytical model with reference to the simulation results in Omnetpp. Additionally, by using our proposed scheme, we can theoretically offload up to 5.43 GB per each stopping station.
Sang, L, Xu, M, Qian, S & Wu, X 2019, 'AAANE: Attention-based adversarial autoencoder for multi-scale network embedding', Advances in Knowledge Discovery and Data Mining 23rd Pacific-Asia Conference, PAKDD 2019, Macau, China, April 14-17, 2019, Proceedings, Part III, Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer, China, pp. 3-14.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. Network embedding represents nodes in a continuous vector space and preserves structure information from a network. Existing methods usually adopt a one-size-fits-all approach when concerning multi-scale structure information, such as first- and second-order proximity of nodes, ignoring the fact that different scales play different roles in embedding learning. In this paper, we propose an Attention-based Adversarial Autoencoder Network Embedding (AAANE) framework, which promotes the collaboration of different scales and lets them vote for robust representations. The proposed AAANE consists of two components: (1) an attention-based autoencoder that effectively capture the highly non-linear network structure, which can de-emphasize irrelevant scales during training, and (2) an adversarial regularization guides the autoencoder in learning robust representations by matching the posterior distribution of the latent embeddings to a given prior distribution. Experimental results on real-world networks show that the proposed approach outperforms strong baselines.
Seifollahi, S, Piccardi, M, Borzeshi, EZ & Kruger, B 2018, 'Taxonomy-augmented features for document clustering', Communications in Computer and Information Science, Australasian Conference on Data Mining, Springer, Bathurst, NSW, Australia, pp. 241-252.
View/Download from: Publisher's site
View description>>
© Springer Nature Singapore Pte Ltd. 2019. In document clustering, individual documents are typically represented by feature vectors based on term-frequency or bag-of-word models. However, such feature vectors intrinsically dismiss the order of the words in the document and suffer from very high dimensionality. For these reasons, in this paper we present novel taxonomy-augmented features that enjoy two promising characteristics: (1) they leverage semantic word embeddings to take the word order into account, and (2) they reduce the feature dimensionality to a very manageable size. Our feature extraction approach consists of three main steps: first, we apply a word embedding technique to represent the words in a word embedding space. Second, we partition the word vocabulary into a hierarchy of clusters by using k-means hierarchically. Lastly, the individual documents are projected to the hierarchy and a compact feature vector is extracted. We propose two methods for generating the features: the first uses all the clusters in the hierarchy and results in a feature vector whose dimensionality is equal to the number of the clusters. The second uses a small set of user-defined words and results in an even smaller feature vector whose dimensionality is equal to the size of the set. Numerical experiments on document clustering show that the proposed approach is capable of achieving comparable or even higher accuracy than conventional feature vectors with a much more compact representation.
Shen, J, Wang, Y & Zhang, J 2018, 'Memory optimized Deep Dense Network for Image Super-resolution', 2018 Digital Image Computing: Techniques and Applications (DICTA), Digital Image Computing: Techniques and Applications, IEEE, Canberra, Australia.
View/Download from: Publisher's site
View description>>
CNN methods for image super-resolution consume a large number of training-time memory, due to the feature size will not decrease as the network goes deeper. To reduce the memory consumption during training, we propose a memory optimized deep dense network for image super-resolution. We first reduce redundant features learning, by rationally designing the skip connection and dense connection in the network. Then we adopt share memory allocations to store concatenated features and Batch Normalization intermediate feature maps. The memory optimized network consumes less memory than normal dense network. We also evaluate our proposed architecture on highly competitive super-resolution benchmark datasets. Our deep dense network outperforms some existing methods, and requires relatively less computation.
Shi, Z, Zhang, JA, Xu, R & Cheng, Q 2019, 'Deep Learning Networks for Human Activity Recognition with CSI Correlation Feature Extraction', IEEE International Conference on Communications, IEEE International Conference on Communications, IEEE, Shanghai, China.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Device free WiFi Sensing using channel state information (CSI) has been shown great potentials for human activity recognition (HAR). However, extracting reliable and concise feature signals remains as a challenging problem, especially in a dynamic and complex environment. In this paper, we propose a novel scheme for CSI-based HAR using deep learning network (CH-DLN), with an innovative CSI correlation feature extraction (CCFE) method. The CCFE method pre-processes the signals input to the DLN in two steps. Firstly, it uses a recursive algorithm to reduce non-activity-related information from the signal and hence enhance the activity-dependent signals. Secondly, it computes the correlation over both the time and frequency domain to disclose better signal structure and compress the signal. From such enhanced and compressed signals, we utilize the recurrent neural networking (RNN) to automatically extract deeper features, and then apply the softmax regression algorithm for classifying activities. Through extensive experimental results, our proposed scheme is shown to outperform state-of-the-art methods in recognition accuracy, with much less training time.
Shi, Z, Zhang, JA, Xu, Y & Fang, G 2018, 'Human Activity Recognition Using Deep Learning Networks with Enhanced Channel State information', 2018 IEEE Globecom Workshops (GC Wkshps), IEEE Globecom Workshops, IEEE, Abu Dhabi, United Arab Emirates.
View/Download from: Publisher's site
View description>>
Channel State Information (CSI) is widely used for device free human activity recognition. Feature extraction remains as one of the most challenging tasks in a dynamic and complex environment. In this paper, we propose a human activity recognition scheme using Deep Learning Networks with enhanced Channel State information (DLN-eCSI). We develop a CSI feature enhancement scheme (CFES), including two modules of background reduction and correlation feature enhancement, for preprocessing the data input to the DLN. After cleaning and compressing the signals using CFES, we apply the recurrent neural networking (RNN) to automatically extract deeper features and then the softmax regression algorithm for activity classification. Extensive experiments are conducted to validate the effectiveness of the proposed scheme.
Shu, Z, Cheng, M, Yang, B, Su, Z & He, X 2019, 'Residual magnifier: A dense information flow network for super resolution', Proceedings - IEEE International Conference on Multimedia and Expo, IEEE International Conference on Multimedia and Expo, IEEE, Shanghai, China, pp. 646-651.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Recently, deep learning methods have been successfully applied to single image super-resolution tasks. However, some networks with extreme depth failed to achieve better performance because of the insufficient utilization of the local residual information extracted at each stage. To solve the above question, we propose a Dense Information Flow Network (DIF-Net), which can fully extract and utilize the local residual information at each stage to accomplish a better reconstruction. Specifically, we present a Two-stage Residual Extraction Block (TREB) to extract the shallow and deep local residual information at each stage. The dense connection mechanism is introduced throughout the model and within TREBs to dramatically increase the information flow. Meanwhile this mechanism prevents the shallow features extracted earlier from being diluted. Finally, we propose a lightweight subnet (residual enhancer) to efficiently recycle the overflow residual information from the backbone net for detail enhancement of the residual image. Experimental results demonstrate that the proposed method performs favorably against the state-of-the-art methods with relatively-less parameters.
Sinha, S & Chaczko, Z 2019, 'Data visualisation of complex adaptive systems', 2019 18th International Conference on Information Technology Based Higher Education and Training, ITHET 2019, International Conference on Information Technology Based Higher Education and Training, IEEE, Magdeburg, Germany, Germany, pp. 1-4.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In Complex Adaptive Systems, the complexity of a construct is predicated by its heterogenic entities and the dynamic nature of their interactions. These interactions are often non-probablistic and triggered as a consequence of potentially multiple previous effects, which can lead to emergent behaviours of the overall system not present in the individual sub-systems. These systems are ubiquitous, not only in natural habitats and within living organisms but also in human-made constructs such as communities, organisations and technology that often have a life of their own. Human-made constructs are particularly interesting due to the data generated that can be unstructured, multidimensional and dynamic, often containing outlier data that may be considered anomalous but can still have critical impact. Complex adaptive data or data generated from Complex Adaptive Systems have interesting ramifications, particularly for data visualisation. Due to the unpredictable nature of complex adaptive data well known forms of data visualisation such as bar charts, scatter plots and heatmaps are inadequate in effectively communicating the data stories in Complex Adaptive Systems. There is a need for more advanced data visualisations that caters for complex adaptive data as it can serve as an educational and decision-making tool for experimentation of ideas in systems that are complex and adaptive.
Song, LZ, Qin, PY & Guo, YJ 2019, 'Conformal Transmitarray and Its Beam Scanning', 2019 International Symposium on Antennas and Propagation, ISAP 2019 - Proceedings, International Symposium on Antennas and Propagation, IEEE, Xi'an, China, pp. 1-3.
View description>>
© 2019 Antenna Branch of Chinese Institute of Electronics. A mechanically beam-scanning conformal transmitarray is developed in this paper. Firstly, a transmitarray element with three layers of identical square ring slots is proposed and its performance for different element thickness is studied. A transmission phase range of 330° with a maximum 3.6 dB loss can be achieved when the thickness is 0.508mm (only 0.04 wavelength at 25GHz). Secondly, a cylindrically conformal transmitarray is designed using the above antenna elements, realizing a 45.3% simulated antenna efficiency. Finally, the above fixed-beam conformal transmitarray is expanded to a beam scanning one. By rotating the feed horn to different positions, the main beam of the array can be switched to ±15°, ±10°, ±5° and 0°, while the whole size of this array is only 2.5 times larger than the fixed beam one. A prototype is fabricated and measured with a stable gain of about 18.7 dBi at all beam angles.
Song, X, Fan, X, He, X, Xiang, C, Ye, Q, Huang, X, Fang, G, Chen, LL, Qin, J & Wang, Z 2019, 'Cnnloc: Deep-learning based indoor localization with wifi fingerprinting', Proceedings - 2019 IEEE SmartWorld, Ubiquitous Intelligence and Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Internet of People and Smart City Innovation, SmartWorld/UIC/ATC/SCALCOM/IOP/SCI 2019, IEEE SmartWorld, Ubiquitous Intelligence and Computing, Advanced and Trusted Computing, Scalable Computing and Communications, Internet of People and Smart City Innovation, IEEE, Leicester, United Kingdom, pp. 589-595.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. With the ubiquitous deployment of wireless systems and pervasive availability of smart devices, indoor localization is empowering numerous location-based services. With the established radio maps, WiFi fingerprinting has become one of the most practical approaches to localize mobile users. However, most fingerprint-based localization algorithms are computationintensive, with heavy dependence on both offline training phase and online localization phase. In this paper, we propose CNNLoc, a Convolutional Neural Network (CNN) based indoor localization system with WiFi fingerprints for multi-building and multifloor localization. Specifically, we devise a novel classification model by combining a Stacked Auto-Encoder (SAE) with a onedimensional CNN. The SAE is utilized to precisely extract key features from sparse Received Signal Strength (RSS) data while the CNN is trained to effectively achieve high success rates in the positioning phase. We evaluate the proposed system on the UJIIndoorLoc dataset and Tampere dataset with several stateof-the-art methods. The results show CNNLoc outperforms the existing solutions with 100% and 95% success rates on buildinglevel localization and floor-level localization, respectively.
Suarez-Rodriguez, C, He, Y, Jayawickrama, BA & Dutkiewicz, E 2019, 'Low-Overhead Handover-Skipping Technique for 5G Networks', IEEE Wireless Communications and Networking Conference, WCNC.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Network densification has been one of the principal causes of performance gain in cellular networks, and 5G networks will not be any different. As cell sizes shrink, handovers become more frequent incurring extra delays that bury all the prospective gains. Mobility in multi-tier dense cellular networks calls for a change in the way it has been traditionally handled in an always-on world, where users take universal data access for granted. Invisible to them, mobile network operators need to provision backhauling to include advanced interference mitigation techniques. In this paper, we propose a spectrum database-aided handover management technique that aims to mitigate the number of disconnections without overloading the backhaul unnecessarily. The proposed technique exploits a spectrum database that stores reception information along with geolocation data, commercially available on any handheld device. Moreover, we have benchmarked several state-of-the-art handover schemes for 5G networks against ours in a realistic urban environment with user mobility trace data. The results highlight that our method can deliver the same downstream traffic with 33% decrease in disconnections when compared to the conventional approach. At the same time, backhaul traffic is reduced up to 68% against our counterparts.
Sun, F, Zhu, H, Zhu, X, Yang, Y, Sun, Y & Xue, Q 2019, 'Design of ultra-wideband on-chip millimter-wave bandpass filter in 0.13-µm (Bi)-CMOS technology', Proceedings - IEEE International Symposium on Circuits and Systems, IEEE International Symposium on Circuits and Systems, IEEE, Sapporo, Japan.
View/Download from: Publisher's site
View description>>
© 2019 IEEE In this work, an on-chip bandpass filter (BPF) with ultra-wideband, low insertion loss, sharp selectivity and excellent in-band flatness is achieved using a novel design approach based on a quasi-lumped-element method. This approach simply utilizes folded metal strip lines with metal-insulator-metal (MIM) capacitors. To understand the principle of the presented design approach, theoretical analysis is given by means of a simplified equivalent LC-circuit model. Using the analyzed results with a full-wave electromagnetic (EM) simulator to guide the design, a BPF is implemented and fabricated in a standard 0.13-µm (Bi)-CMOS technology. The measurements show that a return loss of better than 10 dB is obtained from 13.5 to 32 GHz. Furthermore, the insertion loss of less than 2.3 dB is achieved with less than 0.1 dB in-band magnitude ripple. The BPF size without measurement pads is only 0.148 mm2 (0.37 × 0.4 mm2).
Sun, F, Zhu, X, Zhu, H, Yang, Y & Gomez-Garcia, R 2019, 'On-Chip Millimeter-Wave Bandpass Filter Design Using Multi-Layer Modified-Ground-Ring Structure', IEEE MTT-S International Microwave Symposium Digest, IEEE MTT-S International Microwave Symposium, IEE, Boston, MA, USA, pp. 853-856.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A design approach for compact on-chip bandpass filters (BPFs) is presented in this work. It exploits a novel multi-layer modified-ground-ring structure (ML-MGRS) with additional ground plates that allows to generate a transmission zero for selectivity increase without extra occupied area. A simplified LC-equivalent circuit model is provided to understand the operational principles of this ML-MGRS concept. Moreover, to validate the experimental feasibility of this transmission-zero-creation technique for on-chip BPFs, a millimeter-wave compact BPF is designed and fabricated in a standard 0.13- CMOS technology. The measured results show that the filter exhibits out-of-band power-suppression levels above 40 dB beyond 40 GHz. The center frequency of this filter is 23.5 GHz with a power-insertion-loss level of 3.8 dB, while the input power-matching levels are higher than 10 dB from 19 GHz to 28 GHz. The size of the BPF, excluding the pads, is only 0.06 × 0.284 mm.
Takalkar, MA, Zhang, H & Xu, M 2019, 'Improving Micro-expression Recognition Accuracy Using Twofold Feature Extraction', MultiMedia Modeling (LNCS), International Conference on Multimedia Modeling, Springer, Thessaloniki, Greece, pp. 652-664.
View/Download from: Publisher's site
View description>>
© 2019, Springer Nature Switzerland AG. Micro-expressions are generated involuntarily on a person’s face and are usually a manifestation of repressed feelings of the person. Micro-expressions are characterised by short duration, involuntariness and low intensity. Because of these characteristics, micro-expressions are difficult to perceive and interpret correctly, and they are profoundly challenging to identify and categorise automatically. Previous work for micro-expression recognition has used hand-crafted features like LBP-TOP, Gabor filter, HOG and optical flow. Recent work also has demonstrated the possible use of deep learning for micro-expression recognition. This paper is the first work to explore the use of hand-craft feature descriptor and deep feature descriptor for micro-expression recognition task. The aim is to use the hand-craft and deep learning feature descriptor to extract features and integrate them together to construct a large feature vector to describe a video. Through experiments on CASME, CASME II and CASME+2 databases, we demonstrate our proposed method can achieve promising results for micro-expression recognition accuracy with larger training samples.
Tang, MC & Ziolkowski, RW 2019, 'Multifunctional Huygens Dipole Antennas', 13th European Conference on Antennas and Propagation, EuCAP 2019, European Conference on Antennas and Propagation, IEEE, Krakow, Poland.
View description>>
© 2019 European Association on Antennas and Propagation. Two electrically small, multifunctional Huygens dipole antennas that operate in the L-band are reviewed briefly. In both designs, two pairs of magnetic and electric near-field resonant parasitic (NFRP) elements are combined. Egyptian axe dipoles (EADs) generate the electric dipoles; capacitively loaded loops (CLLs) generate the magnetic dipoles. These NFPR elements are excited with coax-fed driven dipole elements. Both systems are low profile and radiate cardioid patterns pointed in the broadside direction. One Huygens antenna is a dual linearly polarized (dual-LP) system. The other one produces parallel, LP fields at two operating frequencies (dual-band LP).
Tofigh, F, Mao, G, Lipman, J & Abolhasan, M 2018, 'Crowd Density Mapping Based on Wi-Fi Measurements on Train Platforms', 2018 12TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING AND COMMUNICATION SYSTEMS (ICSPCS), International Conference on Signal Processing and Communication Systems, IEEE, Cairns, Australia.
View/Download from: Publisher's site
View description>>
Crowd distribution is a challenging issue in the management and design levels. This paper provides a passive method to derive the crowd density distribution using Wi-Fi measurements on a real scenario. Six WiFi access points (AP) are deployed in the platform 2/3 of Redfern station, Sydney to monitor the platform for a week. Based on the probability maps that are built using RSSI measurements and prior knowledge, the crowd distribution is calculated on the platform and its results are compared with distributions acquired from CCTV images. Final density heat maps are in good agreement with the acquired results from CCTV cameras.
Trede, F, Braun, R & Brookes, W 2019, 'Studio-based learning in a first year engineering curriculum: Exploring students' learning experiences and reflections using the rich picture method', 2019 18th International Conference on Information Technology Based Higher Education and Training, ITHET 2019, International Conference on Information Technology Based Higher Education and Training, IEEE, Magdeburg, Germany.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. We have described engineering students in their first year participating in a 'studio' based experience. We used a rich picture method imbedded in research interviews to explore student's attitudes to, and understandings of their studio experience. Our findings demonstrate that this research method produces an enriched understanding of and deep insights into student experiences in the studio.
Trinh, CD, Kien, VC, Bac, DH, Dutkiewicz, E, Hanh, T & Trung, NL 2019, 'Message from the ISCIT'19 General Chairs', Proceedings - 2019 19th International Symposium on Communications and Information Technologies, ISCIT 2019, p. XXI.
View/Download from: Publisher's site
Van Huynh, N, Hoang, DT, Nguyen, DN & Dutkiewicz, E 2019, 'Real-Time Network Slicing with Uncertain Demand: A Deep Learning Approach', ICC 2019 - 2019 IEEE International Conference on Communications (ICC), ICC 2019 - 2019 IEEE International Conference on Communications (ICC), IEEE, Shanghai.
View/Download from: Publisher's site
Vu, L, Cao, VL, Nguyen, QU, Nguyen, DN, Hoang, DT & Dutkiewicz, E 2019, 'Learning Latent Distribution for Distinguishing Network Traffic in Intrusion Detection System', ICC 2019 - 2019 IEEE International Conference on Communications (ICC), ICC 2019 - 2019 IEEE International Conference on Communications (ICC), IEEE, Shanghai.
View/Download from: Publisher's site
Wang, F, Lin, S, Wu, H, Li, H, Wang, R, Luo, X & He, X 2019, 'SPFusionNet: Sketch segmentation using multi-modal data fusion', Proceedings - IEEE International Conference on Multimedia and Expo, Shanghai, pp. 1654-1659.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The sketch segmentation problem remains largely unsolved because conventional methods are greatly challenged by the highly abstract appearances of freehand sketches and their numerous shape variations. In this work, we tackle such challenges by exploiting different modes of sketch data in a unified framework. Specifically, we propose a deep neural network SPFusionNet to capture the characteristic of sketch by fusing from its image and point set modes. The image modal component SketchNet learns hierarchically abstract ro-bust features and utilizes multi-level representations to produce pixel-wise feature maps, while the point set-modal component SPointNet captures local and global contexts of the sampled point set to produce point-wise feature maps. Then our framework aggregates these feature maps by a fusion network component to generate the sketch segmentation result. The extensive experimental evaluation and comparison with peer methods on our large SketchSeg dataset verify the effectiveness of the proposed framework.
Wang, Q, Jia, W, He, X, Lu, Y, Blumenstein, M, Huang, Y & Lyu, S 2019, 'ReELFA: A Scene Text Recognizer with Encoded Location and Focused Attention', 2019 International Conference on Document Analysis and Recognition Workshops (ICDARW), 2019 International Conference on Document Analysis and Recognition Workshops (ICDARW), IEEE, Australia.
View/Download from: Publisher's site
Wang, X, Yu, P, Yu, G, Zha, X, Ni, W, Liu, RP & Guo, YJ 2019, 'A High-Performance Hybrid Blockchain System for Traceable IoT Applications', Network and System Security, International Conference on Network and System Security, Springer, Sapporo, Japan, pp. 721-728.
View/Download from: Publisher's site
View description>>
© 2019, Springer Nature Switzerland AG. Blockchain, as an immutable distributed ledger, can be the key to realize secure and trustworthy IoT applications. However, existing blockchains can hardly achieve high-performance and high-security for large-scale IoT applications simultaneously. In this paper, we propose a hyper blockchain architecture combining the security of public blockchains with the efficiency of private blockchains. An IoT anchoring smart contract is proposed to anchor private IoT blockchains into a public blockchain. An IoT device management smart contract is also designed to trace sensory data. A comprehensive analysis reveals that the proposed hybrid blockchain system can achieve the performance of private blockchains and resist tampering.
Wang, X, Zha, X, Yu, G, Ni, W, Liu, RP, Guo, YJ, Niu, X & Zheng, K 2018, 'Attack and Defence of Ethereum Remote APIs', 2018 IEEE Globecom Workshops, GC Wkshps 2018 - Proceedings, IEEE Globecom Workshops, IEEE, Abu Dhabi, United Arab Emirates, United Arab Emirates.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. Ethereum, as the first Turing-complete blockchain platform, provides various application program interfaces for developers. Although blockchain has highly improved security, faulty configuration and usage can result in serious vulnerabilities. In this paper, we focus on the security vulnerabilities of the official Go-version Ethereum client (geth). The vulnerabilities are because of the insecure API design and the specific Ethereum wallet mechanism. We demonstrate attacks exploiting these vulnerabilities in an Ethereum testbed. The vulnerabilities are confirmed by the scanning results on the public Internet. Finally, corresponding countermeasures against attacks are provided to enhance the security of the Ethereum platform.
Wang, Y, Shen, J & Zhang, J 2018, 'Deep Bi-Dense Networks for Image Super-Resolution', 2018 Digital Image Computing: Techniques and Applications (DICTA), Digital Image Computing: Techniques and Applications, IEEE, Canberra.
View/Download from: Publisher's site
View description>>
This paper proposes Deep Bi-Dense Networks (DBD-N) for single image super-resolution. Our approach extends previous intra-block dense connection approaches by including novel inter-block dense connections. In this way, feature information propagates from a single dense block to all subsequent blocks, instead of to a single successor. To build a DBDN, we firstly construct intra-dense blocks, which extract and compress abundant local features via densely connected convolutional layers and compression layers for further feature learning. Then, we use an inter-block dense net to connect intra-dense blocks, which allow each intra-dense block propagates its own local features to all successors. Additionally, our bi-dense construction connects each block to the output, alleviating the vanishing gradient problems in training. The evaluation of our proposed method on five benchmark data sets shows that our DBDN outperforms the state of the art in SISR with a moderate number of network parameters.
Webb, BA & Ziolkowski, RW 2018, 'Enabling Transmission Through Reentry Plasmas: Simulation of Plasma-SRR Layered Materials', 2018 6TH IEEE International Conference On Wireless For Space And Extreme Environments (WISEE), IEEE International Conference on Wireless for Space and Extreme Environments, IEEE, Huntsville, AL, pp. 101-103.
View/Download from: Publisher's site
View description>>
As a spacecraft enters a planetary atmosphere, severe attenuation of communication signals can occur due to the formation of a plasma sheath around its airframe. The resulting communications blackout has known negative consequences. Early studies have shown that such plasma can be modeled as a negative permittivity material. This feature suggests that a solution may be found in the realm of metamaterials, notably a negative permeability material. In particular, a double negative (DNG) material (i.e., a material with both negative permittivity and permeability) will accommodate travelling waves. We demonstrate that it is possible to introduce a negative permeability metamaterial into the presence of the plasma to form an artificial composite DNG material in order to ameliorate the communication blackout.
Wilson, K, Alabd, R, Abolhasan, M, Franklin, D & Safavi-Naeini, M 2019, 'Localisation of the Lines of Response in a Continuous Cylindrical Shell PET Scanner', 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), IEEE Engineering in Medicine and Biology Conference, IEEE, Berlin, Germany.
View description>>
This work presents a technique for localising the endpoints of the lines of response in a PET scanner based on a continuous cylindrical shell scintillator. The technique is demonstrated by applying it to a simulation of a sensitivity-optimised continuous cylindrical shell PET system using two novel scintillator materials - a transparent ceramic garnet, GLuGAG:Ce, and a LuF$_3$:Ce-polystyrene nanocomposite. Error distributions for the endpoints of the lines of response in the axial, tangential and radial dimension as well as overall endpoint spatial error are calculated for three source positions; the resultant distribution of error in the placement of the lines of response is also estimated.
Xie, HB, Li, C, Xu, RYD & Mengersen, K 2019, 'Robust Kernelized Bayesian Matrix Factorization for Video Background/Foreground Separation', Machine Learning, Optimization, and Data Science (LNCS), International Conference on Machine Learning, Optimization, and Data Science, Springer, Siena, Italy, pp. 484-495.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. Development of effective and efficient techniques for video analysis is an important research area in machine learning and computer vision. Matrix factorization (MF) is a powerful tool to perform such tasks. In this contribution, we present a hierarchical robust kernelized Bayesian matrix factorization (RKBMF) model to decompose a data set into low rank and sparse components. The RKBMF model automatically infers the parameters and latent variables including the reduced rank using variational Bayesian inference. Moreover, the model integrates the side information of similarity between frames to improve information extraction from the video. We employ RKBMF to extract background and foreground information from a traffic video. Experimental results demonstrate that RKBMF outperforms state-of-the-art approaches for background/foreground separation, particularly where the video is contaminated.
Xu, J-X, Zhang, XY & Yang, Y 2019, 'High-Q-Factor Dual-Band Bandpass Filter and Filtering Switch Using Stub-Loaded Coaxial Resonators', 2019 IEEE MTT-S INTERNATIONAL WIRELESS SYMPOSIUM (IWS 2019), 6th IEEE MTT-S International Wireless Symposium (IWS) part of China Microwave Week, IEEE, Guangzhou, PEOPLES R CHINA.
Yan, B, Zhao, Q, Zhang, JA, Li, Y & Wang, Z 2019, 'Convergence acceleration for multiobjective sparse reconstruction via knowledge transfer', Evolutionary Multi-Criterion Optimization (LNCS), International Conference on Evolutionary Multi-Criterion Optimization, Springer, East Lansing, MI, USA, pp. 475-487.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. Multiobjective sparse reconstruction (MOSR) methods can potentially obtain superior reconstruction performance. However, they suffer from high computational cost, especially in high-dimensional reconstruction. Furthermore, they are generally implemented independently without reusing prior knowledge from past experiences, leading to unnecessary computational consumption due to the re-exploration of similar search spaces. To address these problems, we propose a sparse-constraint knowledge transfer operator to accelerate the convergence of MOSR solvers by reusing the knowledge from past problem-solving experiences. Firstly, we introduce the deep nonlinear feature coding method to extract the feature mapping between the search of the current problem and a previously solved MOSR problem. Through this mapping, we learn a set of knowledge-induced solutions which contain the search experience of the past problem. Thereafter, we develop and apply a sparse-constraint strategy to refine these learned solutions to guarantee their sparse characteristics. Finally, we inject the refined solutions into the iteration of the current problem to facilitate the convergence. To validate the efficiency of the proposed operator, comprehensive studies on extensive simulated signal reconstruction are conducted.
Yang, T, Ding, C, Ziolkowski, RW & Jay, YG 2019, 'A THz Single-Polarization-Single-Mode (SPSM) Photonic Crystal Fiber Based on Epsilon-Near-Zero Material', AOS AUSTRALIAN CONFERENCE ON OPTICAL FIBRE TECHNOLOGY (ACOFT) AND AUSTRALIAN CONFERENCE ON OPTICS, LASERS, AND SPECTROSCOPY (ACOLS) 2019, AOS Australian Conference on Optical Fibre Technology (ACOFT) / Australian Conference on Optics, Lasers, and Spectroscopy (ACOLS), SPIE-INT SOC OPTICAL ENGINEERING, Melbourne, AUSTRALIA.
View/Download from: Publisher's site
Yang, Y, Hou, Z, Zhu, X, Che, W & Xue, Q 2019, 'A Millimeter-Wave Reconfigurable On-Chip Coupler with Tunable Power-Dividing Ratios', 2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Proceedings.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a millimeter-wave on-chip tunable coupler with tunable power dividing ratios and constant phase responses. Composed by two coupled-lines, two capacitors and two series-connected varactors, the proposed tunable coupler offers wideband frequency responses. Theoretical analysis for wideband operation is provided with design parameters. For demonstration, a millimeter-wave tunable coupler is implemented in a standard 0.13-μm SiGe (Bi) CMOS technology and measured through an on-wafer probing system. From 25 to 31 GHz, the proposed tunable coupler shows a power-dividing ratio tuned from 0 to 5 dB, while maintaining an in-band return loss of better than 10 dB and an output isolation of 20 dB, simultaneously. The phase imbalance is better than ±4° with a measured insertion loss of 1.6 dB across the entire tuning range.
Yang, Y, Xu, M, Wu, W, Zhang, R & Peng, Y 2018, '3D Multiview Basketball Players Detection and Localization Based on Probabilistic Occupancy', 2018 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2018, Digital Image Computing: Techniques and Applications, IEEE, Australia, pp. 267-274.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. This paper addresses the issue of 3D multiview basketball players detection and localization. Existing methods for this problem typically take background subtraction as input, which limits the accuracy of localization and the performance of further object tracking. Moreover, the performance of background subtraction based methods is heavily impacted by the occlusions in crowded scenes. In this paper, we propose an innovative method which jointly implements deep learning based player detection and occupancy probability based player localization. What's more, a new Bayesian model of the localization algorithms is developed, which uses foreground information from fisheye cameras to setup meaningful initialization values in the first step of iteration, in order to not only eliminate ambiguous detection, but also accelerate computational processes. Experimental results on real basketball game data demonstrate that our methods significantly improve the performance compared with current methods, by eliminating missed and false detection, as well as increasing probabilities of positive results.
Yang, Y, Zhu, X, Che, W & Xue, Q 2019, 'A Millimeter-Wave On-Chip Bandpass Filter with All-Pole Characteristics', 2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Proceedings.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a millimeter-wave on-chip bandpass filter (BPF) with all-pole characteristics. It is believed to be the first time that an all-pole frequency response has been successfully implemented by using a standard 0.13-μm Silicon-Germanium (SiGe) technology. To further demonstrate the feasibility of using this approach in practice, the designed resonator is fabricated. The measured results show that the BPF has an insertion loss of 2.2 dB at the center frequency of 31 GHz. The return loss is better than 10 dB from 26.5 GHz to 37.5 GHz. In addition, the out-of-band suppression of this filter is superior, which is better than 20 dB beyond 50.6 GHz. The chip size, excluding the pads, is only 0.091 × 0.268 mm2.
Yao, L, Kusakunniran, W, Wu, Q, Zhang, J & Tang, Z 2018, 'Robust CNN-based Gait Verification and Identification using Skeleton Gait Energy Image', 2018 Digital Image Computing: Techniques and Applications (DICTA), Digital Image Computing: Techniques and Applications, IEEE, Canberra, Australia.
View/Download from: Publisher's site
View description>>
As a kind of behavioral biometric feature, gait has been widely applied for human verification and identification. Approaches to gait recognition can be classified into two categories: model-free approaches and model-based approaches. Model-free approaches are sensitive to appearance changes. For model-based approaches, it is difficult to extract the reliable body models from gait sequences. In this paper, based on the robust skeleton points produced from a two-branch multi-stage CNN network, a novel model-based feature, Skeleton Gait Energy Image (SGEI), has been proposed. Relevant experimental performances indicate that SGEI is more robust to the cloth changes. Another contribution is that two different CNN-based architectures have been separately proposed for gait verification and gait identification. Both these two architectures have been evaluated on the datasets. They have presented satisfying performances and increased the robustness for gait recognition in the unconstrained environments with view variances and cloth variances.
Ye, P, Wang, Y, Xia, Y, An, P & Zhang, J 2018, 'Enhanced saliency prediction via free energy principle', Digital TV and Multimedia Communication, International Forum on Digital TV and Wireless Multimedia Communications, Springer, Shanghai, China, pp. 31-44.
View/Download from: Publisher's site
View description>>
© Springer Nature Singapore Pte Ltd 2019. Saliency prediction can be treated as the activity of human brain. Most saliency prediction methods employ features to determine the contrast of an image area relative to its surroundings. However, only few studies have investigated how human brain activities affect saliency prediction. In this paper, we propose an enhanced saliency prediction model via free energy principle. A new AR-RTV model, which combines the relative total variation (RTV) structure extractor with autoregressive (AR) operator, is firstly utilized to decompose an original image into the predictable component and the surprise component. Then, we adopt the local entropy of ‘surprise’ map and the gradient magnitude (GM) map to estimate the component saliency maps-sub-saliency respectively. Finally, inspired by visual error sensitivity, a saliency augment operator is designed to enhance the final saliency combined two sub-saliency maps. Experimental results on two benchmark databases demonstrate the superior performance of the proposed method compared to eleven state-of-the-art algorithms.
Yu, G, Wang, X, Zha, X, Zhang, JA & Liu, R 2018, 'An Optimized Round-Robin Scheduling of Speakers for Peers-to-Peers-based Byzantine Faulty Tolerance', 2018 IEEE Globecom Workshops (GC Wkshps), IEEE Globecom Workshops, Abu Dhabi, United Arab Emirates.
View/Download from: Publisher's site
Yue, P, Xin, J, Zhao, H, Liu, D, Shan, M & Zhang, J 2019, 'Experimental research on deep reinforcement learning in autonomous navigation of mobile robot', Proceedings of the 14th IEEE Conference on Industrial Electronics and Applications, ICIEA 2019, IEEE Conference on Industrial Electronics and Applications, IEEE, Xi'an, China, pp. 1612-1616.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The paper is concerned with the autonomous navigation of mobile robot from the current position to the desired position only using the current visual observation, without the environment map built beforehand. Under the framework of deep reinforcement learning, the Deep Q Network (DQN) is used to achieve the mapping from the original image to the optimal action of the mobile robot. Reinforcement learning requires a large number of training examples, which is difficult to directly be applied in a real robot navigation scenario. To solve the problem, the DQN is firstly trained in the Gazebo simulation environment, followed by the application of the well-trained DQN in the real mobile robot navigation scenario. Both simulation and real-world experiments have been conducted to validate the proposed approach. The experimental results of mobile robot autonomous navigation in the Gazebo simulation environment show that the trained DQN can approximate the state action value function of the mobile robot and perform accurate mapping from the current original image to the optimal action of the mobile robot. The experimental results in real indoor scenes demonstrate that the DQN trained in the simulated environment can work in the real indoor environment, and the mobile robot can also avoid obstacles and reach the target location even with dynamics and the presence of interference in the environment. It is therefore an effective and environmentally adaptable autonomous navigation method for mobile robots in an unknown environment.
Zhang, H, Huang, X & Andrew Zhang, J 2019, 'Comparison of OTFS diversity performance over slow and fast fading channels', 2019 IEEE/CIC International Conference on Communications in China, ICCC 2019, IEEE/CIC International Conference on Communications in China, IEEE, Changchun, China, pp. 828-833.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Orthogonal time frequency space (OTFS) modulation shows great performance improvement over high-mobility wireless channels compared with traditional orthogonal frequency division multiplexing (OFDM). In this paper, we first derive the input and output relationship of OTFS signal in the delay-time domain, which shows that OTFS can be regarded as a combination of OFDM and single-carrier frequency-division multiple access (SC-FDMA). We then examine the diversity order of an OTFS system through received signal-to-noise ratio analysis and predict that this modulation technique can potentially achieve full diversity in both delay and Doppler domains. Finally, we simulate the OTFS performance based on 5G tapped-delay-line channel models under both slow and fast fading conditions. Extensive simulation results confirm that OTFS performs significantly better than other modulation techniques in fast fading channels.
Zhang, H, Huang, X, Zhang, JA, Jay Guo, Y, Song, RL, Wang, CT, Wu, W, Xu, XF & Lu, Z 2019, 'A high-speed low-cost millimeter wave system with dual pulse shaping transmission and symbol rate equalization techniques', Proceedings - IEEE International Symposium on Circuits and Systems, IEEE International Symposium on Circuits and Systems, Sapporo, Japan.
View/Download from: Publisher's site
View description>>
© 2019 IEEE A millimeter wave system with commercially available and affordable data conversion devices is presented in this paper for achieving high-speed and low-cost wireless communications. By adopting the proposed dual pulse shaping (DPS) transmission scheme, the system can achieve full Nyquist rate transmission with only half of the sampling rate required by conventional Nyquist pulse shaping. Structures of the DPS transmitter and receiver are described and effective symbol rate equalization techniques suitable for DPS transmission are presented. Simulation results with two sets of practical dual spectral shaping pulses are also provided to compare system performance with the conventional Nyquist pulse shaping system.
Zhang, H, Huang, X, Zhang, T, Zhang, JA & Jay Guo, Y 2019, 'A 30 Gbps Low-Complexity and Real-Time Digital Modem for Wireless Communications at 0.325 THz', Proceedings - 2019 19th International Symposium on Communications and Information Technologies, ISCIT 2019, pp. 260-264.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A high-speed wideband terahertz (THz) communication system with low-complexity and real-time digital signal processing (DSP) is presented in this paper. The architectures of baseband platform, intermediate frequency (IF) module and radio frequency (RF) frontend are described. For real-time DSP implementation with affordable field programmable gate array (FPGA) device, some effective strategies are discussed to reduce resource usage and ensure that the clock constraints are met. Adopting these strategies, all physical layer DSP modules are implemented in two FPGAs with more than 300 MHz system clock. The experimental test results using the developed real-time digital modem prototype demonstrate the superb performance for THz wireless communications.
Zhang, J, Wu, Q, Zhang, J, Shen, C & Lu, J 2019, 'Mind Your Neighbours: Image Annotation With Metadata Neighbourhood Graph Co-Attention Networks', 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.
View/Download from: Publisher's site
Zhang, L, Xu, J, Zhang, J & Gong, Y 2018, 'Information Enhancement for Travelogues via a Hybrid Clustering Model', 2018 Digital Image Computing: Techniques and Applications (DICTA), Digital Image Computing: Techniques and Applications, IEEE, Canberra, ACT, Australia, pp. 1-8.
View/Download from: Publisher's site
View description>>
Travelogues consist of textual information shared by tourists through web forums or other social media which often lack illustrations (images). In image sharing websites like Flicker, users can post images with rich textual information: `title', `tag' and `description'. The topics of travelogues usually revolve around beautiful sceneries. Corresponding landscape images recommended to these travelogues can enhance the vividness of reading. However, it is difficult to fuse such information because the text attached to each image has diverse meanings/views. In this paper, we propose an unsupervised Hybrid Multiple Kernel K-means (HMKKM) model to link images and travelogues through multiple views. Multi-view matrices are built to reveal the correlations between several respects. For further improving the performance, we add a regularisation based on textual similarity. To evaluate the effectiveness of the proposed method, a dataset is constructed from TripAdvisor and Flicker to find the related images for each travelogue. Experiment results demonstrate the superiority of the proposed model by comparison with other baselines.
Zhang, P, Wu, Q & Xu, J 2019, 'VN-GAN: Identity-preserved Variation Normalizing GAN for Gait Recognition', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Budapest, Hungary.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Gait is recognized as a unique biometric characteristic to identify a walking person remotely across surveillance networks. However, the performance of gait recognition severely suffers challenges from view angle diversity. To address the problem, an identity-preserved Variation Normalizing Generative Adversarial Network (VN-GAN) is proposed for learning purely identity-related representations. It adopts a coarse-to-fine manner which firstly generates initial coarse images by normalizing view to an identical one and then refines the coarse images by injecting identity-related information. In specific, Siamese structure with discriminators for both camera view angles and human identities is utilized to achieve variation normalization and identity preservation of two stages, respectively. In addition to discriminators, reconstruction loss and identity-preserving loss are integrated, which forces the generated images to be the same in view and to be discriminative in identity. This ensures to generate identity-related images in an identical view of good visual effect for gait recognition. Extensive experiments on benchmark datasets demonstrate that the proposed VN-GAN can generate visually interpretable results and achieve promising performance for gait recognition.
Zhang, P, Wu, Q & Xu, J 2019, 'VT-GAN: View Transformation GAN for Gait Recognition across Views', Proceedings of the International Joint Conference on Neural Networks, International Joint Conference on Neural Networks, IEEE, Budapest, Hungary.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Recognizing gaits without human cooperation is of importance in surveillance and forensics because of the benefits that gait is unique and collected remotely. However, change of camera view angle severely degrades the performance of gait recognition. To address the problem, previous methods usually learn mappings for each pair of views which incurs abundant independently built models. In this paper, we proposed a View Transformation Generative Adversarial Networks (VT-GAN) to achieve view transformation of gaits across two arbitrary views using only one uniform model. In specific, we generated gaits in target view conditioned on input images from any views and the corresponding target view indicator. In addition to the classical discriminator in GAN which makes the generated images look realistic, a view classifier is imposed. This controls the consistency of generated images and conditioned target view indicator and ensures to generate gaits in the specified target view. On the other hand, retaining identity information while performing view transformation is another challenge. To solve the issue, an identity distilling module with triplet loss is integrated, which constrains the generated images inheriting identity information from inputs and yields discriminative feature embeddings. The proposed VT-GAN generates visually promising gaits and achieves promising performances for cross-view gait recognition, which exhibits great effectiveness of the proposed VT-GAN.
Zhang, T, Gong, C, Jia, W, Song, X, Sun, J & Wu, X 2018, 'Supervised image classification with self-paced regularization', IEEE International Conference on Data Mining Workshops, ICDMW, International Conference on Data Mining Workshops, IEEE, Singapore, Singapore, pp. 411-414.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. In this paper, we present a new scheme for image classification that is robust to samples noises. The proposed scheme depicts a novel sparse classification model with self-paced learning mechanism. First, inspired by the outstanding performance of curriculum learning, we integrate the idea of self-paced learning into supervised class-specific dictionary learning to select appropriate training samples. Secondly, we design a novel sparse representation model associated with self-paced learning regularization, which employs locally linear reconstruction to improve the accuracy of the classifier and exploit the manifold structure of data. By using the designed model, a classification scheme integrating self-paced learning is proposed to exploit more discriminative image information. The experimental results on two typical datasets indicate that our constructed model achieves the competitive performance when compared with the state-of-the-art methods.
Zhang, X, Liu, J, Li, Y, Cui, Q, Tao, X & Liu, RP 2019, 'Blockchain Based Secure Package Delivery via Ridesharing', 2019 11th International Conference on Wireless Communications and Signal Processing, WCSP 2019.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Delivery service via ridesharing is a promising service to share travel costs and improve vehicle occupancy. Existing ridesharing systems require participating vehicles to periodically report individual private information (e.g., identity and location) to a central controller, which is a potential central point of failure, resulting in possible data leakage or tampering in case of controller break down or under attack. In this paper, we propose a Blockchain secured ridesharing delivery system, where the immutability and distributed architecture of the Blockchain can effectively prevent data tampering. However, such tamper-resistance property comes at the cost of a long confirmation delay caused by the consensus process. A Hash-oriented Practical Byzantine Fault Tolerance (PBFT) based consensus algorithm is proposed to improve the Blockchain efficiency and reduce the transaction confirmation delay from 10 minutes to 15 seconds. The Hash-oriented PBFT effectively avoids the double-spending attack and Sybil attack. Security analysis and simulation results demonstrate that the proposed Blockchain secured ridesharing delivery system offers strong security guarantees and satisfies the quality of delivery service in terms of confirmation delay and transaction throughput.
Zhang, X, Shi, J, Zhu, X, Wang, Y, Chen, J, Ding, G & Yang, Z 2019, 'Heterogeneous Integrated MEMS Inertial Switch with Electrostatic Locking and Compliant Cantilever Stationary Electrode for Holding Stable ‘on’-State', 2019 IEEE 32nd International Conference on Micro Electro Mechanical Systems (MEMS), 2019 IEEE 32nd International Conference on Micro Electro Mechanical Systems (MEMS), IEEE.
View/Download from: Publisher's site
Zhang, Z, Wang, Y, Wu, Q & Chen, F 2019, 'Visual Relationship Attention for Image Captioning', Proceedings of the International Joint Conference on Neural Networks.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Visual attention mechanisms have been broadly used by image captioning models to attend to related visual information dynamically, allowing fine-grained image understanding and reasoning. However, they are only designed to discover the region-level alignment between visual features and the language feature. The exploration of higher-level visual relationship information between image regions, which is rarely researched in recent works, is beyond their capabilities. To fill this gap, we propose a novel visual relationship attention model based on the parallel attention mechanism under the learnt spatial constraints. It can extract relationship information from visual regions and language and then achieve the relationship-level alignment between them. Using combined visual relationship attention and visual region attention to attend to related visual relationships and regions respectively, our image captioning model can achieve state-of-the-art performances on the MSCOCO dataset. Both quantitative analysis and qualitative analysis demonstrate that our novel visual relationship attention model can capture related visual relationship and further improve the caption quality.
Zhang, Z, Wu, Q, Wang, Y & Chen, F 2018, 'Size-Invariant Attention Accuracy Metric for Image Captioning with High-Resolution Residual Attention', 2018 Digital Image Computing: Techniques and Applications (DICTA), Digital Image Computing: Techniques and Applications, IEEE, Canberra, Australia, pp. 1-8.
View/Download from: Publisher's site
View description>>
Spatial visual attention mechanisms have achieved significant performance improvements for image captioning. To quantitatively evaluate the performances of attention mechanisms, the "attention correctness" metric has been proposed to calculate the sum of attention weights generated for ground truth regions. However, this metric cannot consistently measure the attention accuracy among the element regions with large size variance. Moreover, its evaluations are inconsistent with captioning performances across different fine-grained attention resolutions. To address these problems, this paper proposes a size-invariant evaluation metric by normalizing the "attention correctness" metric with the size percentage of the attended region. To demonstrate the efficiency of our size-invariant metric, this paper further proposes a high-resolution residual attention model that uses RefineNet as the Fully Convolutional Network (FCN) encoder. By using the COCO-Stuff dataset, we can achieve pixel-level evaluations on both object and "stuff" regions. We use our metric to evaluate the proposed attention model across four high fine-grained resolutions (i.e., 27×27, 40×40, 60×60, 80×80). The results demonstrate that, compared with the "attention correctness" metric, our size-invariant metric is more consistent with the captioning performances and is more efficient for evaluating the attention accuracy.
Zhao, M, Zhang, J, Zhang, C & Zhang, W 2019, 'Leveraging Heterogeneous Auxiliary Tasks to Assist Crowd Counting', 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.
View/Download from: Publisher's site
Zhao, M, Zhang, J, Zhang, C & Zhang, W 2018, 'Towards Locally Consistent Object Counting with Constrained Multi-stage Convolutional Neural Networks', ACCV 2018: Computer Vision, Asian Conference on Computer Vision, Springer, Perth, Australia, pp. 247-261.
View/Download from: Publisher's site
View description>>
High-density object counting in surveillance scenes is challenging mainly due to the drastic variation of object scales. The prevalence of deep learning has largely boosted the object counting accuracy on several benchmark datasets. However, does the global counts really count? Armed with this question we dive into the predicted density map whose summation over the whole regions reports the global counts for more in-depth analysis. We observe that the object density map generated by most existing methods usually lacks of local consistency, i.e., counting errors in local regions exist unexpectedly even though the global count seems to well match with the ground-truth. Towards this problem, in this paper we propose a constrained multi-stage Convolutional Neural Networks (CNNs) to jointly pursue locally consistent density map from two aspects. Different from most existing methods that mainly rely on the multi-column architectures of plain CNNs, we exploit a stacking formulation of plain CNNs. Benefited from the internal multi-stage learning process, the feature map could be repeatedly refined, allowing the density map to approach the ground-truth density distribution. For further refinement of the density map, we also propose a grid loss function. With finer local-region-based supervisions, the underlying model is constrained to generate locally consistent density values to minimize the training errors considering both the global and local counts accuracy. Experiments on two widely-tested object counting benchmarks with overall significant results compared with state-of-the-art methods demonstrate the effectiveness of our approach.
Zhu, H & Jay Guo, Y 2018, 'Modified wideband tandem couplers with arbitrary coupling coefficient and its implementation in beam-forming networks', Asia-Pacific Microwave Conference Proceedings, APMC, Asia-Pacific Microwave Conference, Kyoto, Japan, pp. 542-544.
View/Download from: Publisher's site
View description>>
© 2018 IEICE This paper presents a wideband quadrature coupler using a modified Tandem structure with two stages of cascaded coupled-lines. The proposed design is built in the stripline configuration, which can achieve wide operating bandwidth and excellent matching as well as high isolation across the whole band range. The proposed design with coupling coefficient of 3-dB and 1.77-dB is applied in the design of a wideband beam-forming network for wideband applications. Experimental result has been carried out, verifying that the design approach is useful for wideband applications.
Zhu, H, Cao, Y, Ding, C, Wei, G & Guo, YJ 2018, 'Main Beam Manipulation of Patch Antenna Using Non-Uniform Meta-Surface', International Symposium on Antennas and Propagation (ISAP), International Symposium on Antennas and Propagation, IEEE, Busan, South Korea.
View description>>
A method to manipulate the main beam of patch antenna using non-uniform meta-surface (MS) is proposed in this paper. The proposed antenna is composed of a non-uniform MS placed directly atop of a patch antenna with an area of 100*100 mm2 (0.82 λ 0 * 0.82 λ 0 ), making it compact and low profile. After adding the MS to the patch antenna, the mainbeam direction can be tilted by an angle of 30°from the boresight direction. The proposed antenna is studied and designed to operate around 2.45 GHz. Simulated results show that the antenna has an operating bandwidth from 2.372.51GHz and peak realized gain of 7.3dBi.
Zhu, H, Lin, JY & Guo, YJ 2019, 'Wideband Filtering Out-of-Phase Power Dividers Using Slotline Resonators and Microstrip-to-Slotline Transitions', IEEE MTT-S International Microwave Symposium Digest, IEEE MTT-S International Microwave Symposium, IEEE, Boston, MA, USA, pp. 919-922.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Two wideband filtering out-of-phase power dividers are designed and analyzed. The initial design is composed of a quarter-wavelength slotline resonator and two microstrip-to-slotline transitions. Three transmission poles are produced to form the passband. Excellent matching and isolation are achieved using extra matching network. Based on the initial design, a pair of stepped-impedance open stubs are shunted at the output ports to improve the passband selectivity. The designs have been verified through experiment. The tested results show that the proposed devices have achieved 60% operating bandwidth, 1.8 dB insertion loss, 1.2° phase error and excellent in-band matching and isolation.
Zhu, H, Sun, H, Ding, C & Guo, YJ 2019, 'Butler Matrix Based Multi-Beam Base Station Antenna Array', 13th European Conference on Antennas and Propagation, EuCAP 2019, European Conference on Antennas and Propagation, IEEE, Krakow, Poland,.
View description>>
© 2019 European Association on Antennas and Propagation. In this paper, a three-beam Butler matrix as well as its antenna arrays is presented for cellular base stations. The three-beam Butler matrix is able to generate three beams in the azimuth plane, which can increase the capacity of base stations. Striplines are used for developing the 3 and times; 3 Butler matrix, which is compose of directional couplers and phase shifters. To extend the 3 and times; 3 Butler matrix to a 3 and times; 5 one, unequal power dividers are also require. To verify the beam-forming network, 5-element dual-polarized antenna arrays covering LTE band are developed. Multiple beams are obtained by feeding the antenna array with the augmented 3 and times; 5 Butler matrix. The design is verified by both simulation and experiments.
Zhu, H, Zhu, X, Yang, Y, Sun, Y, Le, VH & Zhang, F 2019, 'Design of miniaturized on-chip bandpass filters using inverting-coupled structure for millimter-wave applications', Proceedings - IEEE International Symposium on Circuits and Systems, IEEE International Symposium on Circuits and Systems, IEEE, Sapporo, Japan.
View/Download from: Publisher's site
View description>>
© 2019 IEEE In this work, a new type of miniaturized on-chip resonator using an inductively-coupled structure is presented. The resonator is constructed by two spiral conductors that are implemented using two different metal layers. Since the two conductors are identical but placed in different rotating pattern, a kind of inductive coupling called inverting coupling will be introduced in addition to the broadside capacitive coupling. To fully understand the working mechanism of the resonator, simplified LC equivalent-circuit models and thorough analysis are provided. To further demonstrate the feasibility of the proposed miniaturized resonator in practice, two bandpass filters, namely a 1st-order and 2nd-order, are designed and fabricated in a standard 0.13-µm (Bi)-CMOS technology. Good agreements between simulation and measurement have obtained, which verify that the presented design approach is suitable for miniaturized on-chip passive design.
Zhu, J, Yang, Y, Chu, C, Li, S, Liao, S & Xue, Q 2019, '60-GHz High Gain Planar Aperture Antenna Using Low-Temperature Cofired Ceramics (LTCC) Technology', 2019 IEEE MTT-S International Wireless Symposium, IWS 2019 - Proceedings.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a new single-ended-fed planar aperture antenna using low-temperature co-fired ceramics (LTCC) process technology. The proposed antenna not only inherits the merits of the aperture antennas including high gain, wide bandwidth, but also exhibits advantages of low profile and compact size. The aperture is excited by a cross-shaped patch and a loop-shaped balun structure placing below the patch. In this way, the energy can propagate on the patch in a traveling wave form and illuminate the aperture with uniform E-field distributions. Therefore, the antenna achieves good electrical and radiation performances, which are comparable to its balancedfed counterparts, while processing a simplified structure. Measured results demonstrate that the impedance bandwidth of the antenna covers the 60-GHz license-free band (57-64GHz) and the maximum gain can reach 11.5 dBi with a cavity size of only about 27 mm2.
Ziolkowski, RW 2018, 'Metamaterial-inspired Electrically Small Platforms: Enhanced Directivity Properties', 2018 IEEE Antennas and Propagation Society International Symposium and USNC/URSI National Radio Science Meeting, APSURSI 2018 - Proceedings, IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science Meeting, IEEE, Boston, MA, USA, pp. 873-874.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. A variety of near-field resonant parasitic (NFRP) antennas have been developed as electrically small platforms to realize high directivity. These include compact arrays and Huygens dipole and multipole radiating systems. A brief review of these developments and their scattering equivalents will be presented.
Ziolkowski, RW 2019, 'Metamaterial-inspired solution to lackluster on-chip antenna performance', 2019 13th International Congress on Artificial Materials for Novel Wave Phenomena, Metamaterials 2019, pp. X495-X497.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. An electric or magnetic dipole antenna located on the interface between a low and high permittivity dielectric faces the problem that the physics tells us that the majority of the power it emits will be radiated into the high dielectric region. This effect is a significant problem for an on-chip antenna associated with systems-on-chip applications such as mobile computing and embedded systems. It is demonstrated that one can use metamaterial-inspired Huygens antennas to overcome this very practical problem.