Abidi, S, Piccardi, M, Tsang, IW & Williams, M-A 2019, 'Well-M$^3$N: 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.
Amiri, M, Tofigh, F, Shariati, N, Lipman, J & Abolhasan, M 2019, 'Miniature tri‐wideband Sierpinski–Minkowski fractals metamaterial perfect absorber', IET Microwaves, Antennas & 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>>
© 1963-2012 IEEE. 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.
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>>
© 2018 The Author(s). Objectives: (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. Design: Observation of simulated hand hygiene encounters between a healthcare worker and a patient. Setting: Computer laboratory in a university. Participants: Healthy volunteers. Main outcome measures: Sensitivity and specificity of automatic detection of the first moment of hand hygiene. Methods: 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. Results: 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%). Conclusions: 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, P-Y, 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
View description>>
© 1963-2012 IEEE. A design methodology for a compact millimeter-wave on-chip bandpass filter (BPF) is presented in this paper. Unlike the previously published works in the literature, the presented method is based on quasi-lumped elements, which consists of a resonator with enhanced self-coupling and metal-insulator-metal capacitors. Thus, this approach provides inherently compact designs comparing with the conventional distributed elements-based ones. To fully understand the insight of the approach, simplified LC-equivalent circuit models are developed. To further demonstrate the feasibility of using this approach in practice, the resonator and two compact BPFs are designed using the presented models. All three designs are fabricated in a standard 0.13- \mu \text{m} (Bi)-CMOS technology. The measured results show that the resonator can generate a notch at 47 GHz with the attenuation better than 28 dB due to the enhanced self-coupling. The chip size, excluding the pads, is only 0.096 \times 0.294 mm2. In addition, using the resonator for BPF designs, the first BPF has one transmission zero at 58 GHz with a peak attenuation of 23 dB. The center frequency of this filter is 27 GHz with an insertion loss of 2.5 dB, while the return loss is better than 10 dB from 26 to 31 GHz. The second BPF has two transmission zeros, and a minimum insertion loss of 3.5 dB is found at 29 GHz, while the return loss is better than 10 dB from 26 GHz to 34 GHz. Also, more than 20-dB stopband attenuation is achieved from dc to 20.5 GHz and from 48 to 67 GHz. The chip sizes of these two BPFs, excluding the pads, are only 0.076\times 0.296 mm2 and 0.096\times 0.296 mm2, respectively.
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.
Chen, S-L, Karmokar, DK, Li, Z, Qin, P-Y, 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, S-L, Karmokar, DK, Li, Z, Qin, P-Y, Ziolkowski, RW & Guo, YJ 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, 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, IB, 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, DN & Dutkiewicz, E 2019, 'Sensing OFDM Signal: A Deep Learning Approach', IEEE Transactions on Communications, vol. 67, no. 11, pp. 7785-7798.
View/Download from: Publisher's site
View description>>
© 1972-2012 IEEE. Spectrum sensing plays a critical role in dynamic spectrum sharing, a promising technology to address the radio spectrum shortage. In particular, sensing of orthogonal frequency division multiplexing (OFDM) signals, a widely accepted multi-carrier transmission paradigm, has received paramount interest. Despite various efforts, noise uncertainty, timing delay and carrier frequency offset (CFO) still remain as challenging problems, significantly degrading the sensing performance. In this work, we develop two novel OFDM sensing frameworks utilizing the properties of deep learning networks. Specifically, we first propose a stacked autoencoder based spectrum sensing method (SAE-SS), in which a stacked autoencoder network is designed to extract the hidden features of OFDM signals for classifying the user's activities. Compared to the conventional OFDM sensing methods, SAE-SS is significantly superior in the robustness to noise uncertainty, timing delay, and CFO. Moreover, SAE-SS requires neither any prior information of signals (e.g., signal structure, pilot tones, cyclic prefix) nor explicit feature extraction algorithms which however are essential for the conventional OFDM sensing methods. To further improve the sensing accuracy of SAE-SS, especially under low SNR conditions, we propose a stacked autoencoder based spectrum sensing method using time-frequency domain signals (SAE-TF). SAE-TF achieves higher sensing accuracy than SAE-SS using the features extracted from both time and frequency domains, at the cost of higher computational complexity. Through extensive simulation results, both SAE-SS and SAE-TF are shown to achieve notably higher sensing accuracy than that of state of the art approaches.
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
View description>>
© The Institution of Engineering and Technology 2019. A high-efficiency, S-band Doherty power amplifier (DPA) with wide output power back-off (OPBO) range is presented. A novel parasitic capacitance compensation approach is applied at the output of Cree's GaN high-electron-mobility transistor to achieve high saturation efficiency in a wide OPBO range. Specifically, a parallel shorting microstrip line between the transistor output and its match network is adopted to realise parasitic capacitance compensation. The measurement results indicate good Doherty behaviour with 10 dB back-off efficiency of 40.6-44.2% and saturation efficiency of 70.2-73.3% over 2.9-3.3 GHz. When stimulated by a 20-MHz LTE signal with 7.5 dB PAPR, the proposed Doherty amplifier power, combined with digital predistortion, achieved adjacent channel leakage ratios below -47.2 dBc. The DPA demonstrate superior performance in OPBO range and efficiency, which makes it an ideal component for base station communication systems.
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.
Cui, P-F, Zhang, JA, Lu, W-J, 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.
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, 'Correction: Analysis of healthcare service utilization after transport-related injuries by a mixture of hidden Markov models', PLOS ONE, vol. 14, no. 4, pp. e0214973-e0214973.
View/Download from: Publisher's site
View description>>
© 2019 Esmaili et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. After publication of this article [1], concerns were raised that the references to the software packages used for this analysis had been omitted. The authors utilized Stata Statistical Software: Release 15. The reference is StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX: StataCorp LLC. The authors also utilized the seqHMM package package in R. The reference is: Helske S, Helske J. Mixture hidden markov models for sequence data: the seqhmm package in R. arXiv preprint arXiv:1704.00543. 2017 Apr 3.
Gao, X, Du, J, Zhang, T & Guo, YJ 2019, 'High-<italic>T<sub>c</sub> </italic> 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-Tc 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.
Ghantous, GB & Gill, AQ 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>>
©2019 The authors and IJLTER.ORG. All rights reserved. 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.
Gill, AQ & Chew, E 2019, 'Configuration information system architecture: Insights from applied action design research.', Inf. Manag., 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 and Computation: Practice and Experience, vol. 31, no. 23.
View/Download from: Publisher's site
View description>>
SummaryRegarding 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.
He, W, Sun, C, Wunsch, DC & Xu, RYD 2019, 'Guest Editorial Special Issue on Intelligent Control Through Neural Learning and Optimization for Human–Machine Hybrid Systems', IEEE Transactions on Neural Networks and Learning Systems, vol. 30, no. 12, pp. 3530-3533.
View/Download from: Publisher's site
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>>
© 2019, The Author(s). 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.
Hoang, DT, Nguyen, DN, Alsheikh, MA, Gong, S, Dutkiewicz, E, Niyato, D & Han, Z 2019, ''Borrowing Arrows with Thatched Boats': The Art of Defeating Reactive Jammers in IoT Networks', IEEE Wireless Communications Magazine, vol. 27, no. 3, pp. 79-87.
View/Download from: Publisher's site
View description>>
In this article, we introduce a novel deception strategy which is inspired bythe 'Borrowing Arrows with Thatched Boats', one of the most famous militarytactics in the history, in order to defeat reactive jamming attacks forlow-power IoT networks. Our proposed strategy allows resource-constrained IoTdevices to be able to defeat powerful reactive jammers by leveraging their ownjamming signals. More specifically, by stimulating the jammer to attack thechannel through transmitting fake transmissions, the IoT system can not onlyundermine the jammer's power, but also harvest energy or utilize jammingsignals as a communication means to transmit data through using RF energyharvesting and ambient backscatter techniques, respectively. Furthermore, wedevelop a low-cost deep reinforcement learning framework that enables thehardware-constrained IoT device to quickly obtain an optimal defense policywithout requiring any information about the jammer in advance. Simulationresults reveal that our proposed framework can not only be very effective indefeating reactive jamming attacks, but also leverage jammer's power to enhancesystem performance for the IoT network.
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. 1-18.
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, J, Fei, Z, Wang, T, Wang, X, Liu, F, Zhou, H, Zhang, JA & Wei, G 2019, 'V2X-communication assisted interference minimization for automotive radars', China Communications, vol. 16, no. 10, pp. 100-111.
View/Download from: Publisher's site
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.
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. 3557-3557.
View/Download from: Publisher's site
View description>>
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement. Advanced handheld plenoptic cameras are being rapidly developed to capture information about light fields (LFs) from the 3D world. Rich LF data can be used to develop dense sub-aperture images (SAIs) that can provide a more immersive experience for users. Unlike conventional 2D images, 4D SAIs contain both the positional and directional information of light rays; the practical applications of handheld plenoptic cameras are limited by the huge volume of data required to capture this information. Therefore, an efficient LF compression method is vital for further application of the cameras. To this end, the pair of steps and depth estimation (PoS&DE) method is proposed in this paper, and the multiview video and depth (MVD) coding structure is used to relieve the LF coding burden. More specifically, a precise depth-estimation approach is presented for SAIs based on the cost function, and an SAI-guided depth optimization algorithm is designed to refine the initial depth map based on pixel variation tendency. Meanwhile, to reduce running time, intermediate SAI synthesis quality and coding bitrates, including the key SAIs selected and cost-computation steps, are set via extensive statistical experiments. In this way, only a limited number of optimally selected SAIs and their corresponding depth maps must be encoded. The experimental results demonstrate that our proposed LF compression solution using PoS&DE can obtain a satisfied coding performance.
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>>
© 2019 IEEE. 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>>
© 1992-2012 IEEE. Sufficient training data normally is required to train deeply learned models. However, due to the expensive manual process for a labeling 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 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 pre-defined training classes in the real data domain. Unlike the traditional label which usually is a single integral number, the virtual label proposed in this paper 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>>
© 2014 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 perform 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 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 to the original target signals with different signal to noise ratios. 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.
Huynh, NV, Hoang, DT, Nguyen, DN & Dutkiewicz, E 2019, 'Optimal and Fast Real-time Resources Slicing with Deep Dueling Neural Networks', IEEE Journal on Selected Areas in Communications, vol. 37, no. 6, pp. 1-1.
View/Download from: Publisher's site
View description>>
Effective network slicing requires an infrastructure/network provider to dealwith the uncertain demand and real-time dynamics of network resource requests.Another challenge is the combinatorial optimization of numerous resources,e.g., radio, computing, and storage. This article develops an optimal and fastreal-time resource slicing framework that maximizes the long-term return of thenetwork provider while taking into account the uncertainty of resource demandfrom tenants. Specifically, we first propose a novel system model which enablesthe network provider to effectively slice various types of resources todifferent classes of users under separate virtual slices. We then capture thereal-time arrival of slice requests by a semi-Markov decision process. Toobtain the optimal resource allocation policy under the dynamics of slicingrequests, e.g., uncertain service time and resource demands, a Q-learningalgorithm is often adopted in the literature. However, such an algorithm isnotorious for its slow convergence, especially for problems with largestate/action spaces. This makes Q-learning practically inapplicable to our casein which multiple resources are simultaneously optimized. To tackle it, wepropose a novel network slicing approach with an advanced deep learningarchitecture, called deep dueling that attains the optimal average reward muchfaster than the conventional Q-learning algorithm. This property is especiallydesirable to cope with real-time resource requests and the dynamic demands ofusers. Extensive simulations show that the proposed framework yields up to 40%higher long-term average return while being few thousand times faster, comparedwith state of the art network slicing approaches.
Huynh, NV, Nguyen, DN, Hoang, DT & 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
View description>>
With conventional anti-jamming solutions like frequency hopping or spreadspectrum, legitimate transceivers often tend to 'escape' or 'hide' themselvesfrom jammers. These reactive anti-jamming approaches are constrained by thelack of timely knowledge of jamming attacks. Bringing together the latestadvances in neural network architectures and ambient backscatteringcommunications, this work allows wireless nodes to effectively 'face' thejammer by first learning its jamming strategy, then adapting the rate ortransmitting information right on the jamming signal. Specifically, to dealwith unknown jamming attacks, existing work often relies on reinforcementlearning algorithms, e.g., Q-learning. However, the Q-learning algorithm isnotorious for its slow convergence to the optimal policy, especially when thesystem state and action spaces are large. This makes the Q-learning algorithmpragmatically inapplicable. To overcome this problem, we design a novel deepreinforcement learning algorithm using the recent dueling neural networkarchitecture. Our proposed algorithm allows the transmitter to effectivelylearn about the jammer and attain the optimal countermeasures thousand timesfaster than that of the conventional Q-learning algorithm. Through extensivesimulation results, we show that our design (using ambient backscattering andthe deep dueling neural network architecture) can improve the averagethroughput by up to 426% and reduce the packet loss by 24%. By augmenting theambient backscattering capability on devices and using our algorithm, it isinteresting to observe that the (successful) transmission rate increases withthe jamming power. Our proposed solution can find its applications in bothcivil (e.g., ultra-reliable and low-latency communications or URLLC) andmilitary scenarios (to combat both inadvertent and deliberate jamming).
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 & Da Xu, RY 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.
Jiao, S & Liu, RP 2019, 'A survey on physical authentication methods for smart objects in IoT ecosystem', Internet of Things, vol. 6, pp. 100043-100043.
View/Download from: Publisher's site
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, S-L, 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
View description>>
© 2019 IEEE. To support diversified quality of service demands and dynamic resource requirements of users in 5G driven VANETs, network resources need flexible and scalable resource allocation strategies. Current heterogeneous vehicular networks are designed and deployed with a connection-centric mindset with fixed resource allocation to a cell regardless of traffic conditions, static coverage, and capacity. In this paper, we propose a hybrid-fuzzy logic guided genetic algorithm (H-FLGA) approach for the software defined networking controller, to solve a multi-objective resource optimization problem for 5G driven VANETs. Realizing the service oriented view, the proposed approach formulates five different scenarios of network resource optimization in 5G VANETs. Furthermore, the proposed fuzzy inference system is used to optimize weights of multi-objectives, depending on the type of service requirements of customers. The proposed approach shows the minimized value of multi-objective cost function when compared with the GA. The simulation results show the minimized value of end-to-end delay as compared to other schemes. The proposed approach will help the network service providers to implement a customer-centric network infrastructure, depending on dynamic customer needs of users.
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, Guo, YJ & 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-B, Fan, X, Xu, RYD, Van Huffel, S, 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
View description>>
© 1992-2012 IEEE. Singular value thresholding (SVT)- or nuclear norm minimization (NNM)-based nonlocal image denoising methods often rely on the precise estimation of the noise variance. However, most existing methods either assume that the noise variance is known or require an extra step to estimate it. Under the iterative regularization framework, the error in the noise variance estimate propagates and accumulates with each iteration, ultimately degrading the overall denoising performance. In addition, the essence of these methods is still least squares estimation, which can cause a very high mean-squared error (MSE) and is inadequate for handling missing data or outliers. In order to address these deficiencies, we present a hybrid denoising model based on variational Bayesian inference and Stein's unbiased risk estimator (SURE), which consists of two complementary steps. In the first step, the variational Bayesian SVT performs a low-rank approximation of the nonlocal image patch matrix to simultaneously remove the noise and estimate the noise variance. In the second step, we modify the conventional SURE full-rank SVT and its divergence formulas for rank-reduced eigen-triplets to remove the residual artifacts. The proposed hybrid BSSVT method achieves better performance in recovering the true image compared with state-of-the-art methods.
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
View description>>
© 2013 IEEE. The propagation of a millimeter wave (mmWave) signal is dominated by its line-of-sight component. Therefore, the knowledge of angle-of-arrival and polarization state of the wave is of great importance for its reception at the receiver. In this paper, we estimate these parameters for an information-bearing signal in mmWave systems using hybrid antenna arrays with dual-polarized dipoles. The estimation is studied in the context of both the interleaved and localized arrays. Two blind adaptive algorithms, namely, the joint differential beam tracking and cross-correlation-to-power ratio polarization tracking, and the differential beam and polarization search, are developed, each tailored for an array. It is shown that the use of dual-polarized dipoles in combination with the developed algorithms effectively lead to polarization diversity which significantly enhances the signal-to-noise ratio at the decoder. The simulation results also show that the antennas with dual dipoles provide improved accuracy and convergence rate for the estimations compared with the conventional arrays.
Li, H, Wang, TQ, Huang, X & Jay Guo, Y 2019, 'Adaptive AoA and Polarization Estimation for Receiving Polarized mmWave Signals', IEEE Wireless Communications Letters, vol. 8, no. 2, pp. 540-543.
View/Download from: Publisher's site
View description>>
© 2012 IEEE. This letter 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
View description>>
© 1972-2012 IEEE. In this paper, we study the low complexity reception of multiuser signals in uplink millimeter wave (mmWave) communications using a partially connected hybrid antenna array. Exploiting the mmWave channel property, we propose a low-complexity user-directed multiuser receiver with three novel schemes for allocating subarrays to users. This receiver only requires the knowledge of angles-of-Arrival (AoAs) for dominating paths and a small amount of equivalent channel information instead of perfect channel state information. For comparison, we also derive a successive interference cancellation-based solution as a performance benchmark. We design two types of reference signals with the channel estimation method to enable efficient and simple estimation for AoA and equivalent baseband channel. Also, we provide analytical results for the performance of the AoA estimation, using the lower bounds of mean square errors in line-of-sight dominated mmWave channels. The simulation results validate that the proposed channel estimation method is effective when employed in combination with a zero-forcing equalizer.
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>>
© 2013 IEEE. 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, Z, Guo, YJ, Chen, S-L & 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.
Lin, J-Y, Wong, S-W, Wu, Y-M, 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, J-Y, Yang, Y, Wong, S-W, Chen, R-S, 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>>
© 1963-2012 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.
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, L, Amirgholipour, S, Jiang, J, Jia, W, Zeibots, M & He, X 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, 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, 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>>
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, 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
View description>>
© 1972-2012 IEEE. Joint communication and radio sensing (JCAS) in millimeter-wave (mmWave) systems requires the use of a steerable beam. For analog antenna arrays, a single beam is typically used, which limits the sensing area within the direction of the communication. Multibeam technology can overcome this limitation by separately generating package-level direction-varying sensing subbeams and fixed communication subbeams and then combine them coherently. In this paper, we investigate the optimal combination of the two subbeams and the quantization of the beamforming (BF) vector that generates the combined beam. When either the full channel matrix or only the angle of departure (AoD) of the dominating line-of-sight (LOS) path is known at the transmitter, we derive the closed-form expressions for the optimal combining coefficients that maximize the received communication signal power. For the quantization of the BF vector, we focus on the two-phase-shifter array where two phase shifters are used to represent each BF weight. We propose novel joint quantization methods by combining the codebooks of the two phase shifters. The mean squared quantization error is derived for various quantization methods. Extensive simulation results validate the accuracy of the analytical results and the effectiveness of the proposed multibeam optimization and joint quantization methods.
Luong, NC, Hoang, DT, Gong, S, Niyato, D, Wang, P, Liang, Y-C & Kim, DI 2019, 'Applications of Deep Reinforcement Learning in Communications and Networking: A Survey', IEEE Communications Surveys & 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, 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>>
© 2018 Elsevier Ltd 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', IEEE Communications Surveys & Tutorials, vol. 21, no. 2, pp. 1636-1675.
View/Download from: Publisher's site
View description>>
© 1998-2012 IEEE. 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.
Nguyen, CT, Hoang, DT, 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, D, vanSonnenberg, E, Kang, P & Mueller, PR 2019, 'Urologic and interventional radiology treatment of renal cell carcinomas—similarities and differences', Annals of Translational Medicine, vol. 7, no. S3, pp. S113-S113.
View/Download from: Publisher's site
Nguyen, HV, Nguyen, V-D, Dobre, OA, Nguyen, DN, Dutkiewicz, E & Shin, O-S 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
View description>>
This paper investigates the coexistence of non-orthogonal multiple access(NOMA) and full-duplex (FD) to improve both spectral efficiency (SE) and userfairness. In such a scenario, NOMA based on the successive interferencecancellation technique is simultaneously applied to both uplink (UL) anddownlink (DL) transmissions in an FD system. We consider the problem of jointlyoptimizing user association (UA) and power control to maximize the overall SE,subject to user-specific quality-of-service and total transmit powerconstraints. To be spectrally-efficient, we introduce the tensor model tooptimize UL users' decoding order and DL users' clustering, which results in amixed-integer non-convex problem. For practically appealing applications, wefirst relax the binary variables and then propose two low-complexity designs.In the first design, the continuous relaxation problem is solved using theinner convex approximation framework. Next, we additionally introduce thepenalty method to further accelerate the performance of the former design. Fora benchmark, we develop an optimal solution based on brute-force search (BFS)over all possible cases of UAs. It is demonstrated in numerical results thatthe proposed algorithms outperform the conventional FD-based schemes and itshalf-duplex counterpart, as well as yield data rates close to those obtained byBFS-based algorithm.
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
View description>>
© 1963-2012 IEEE. A novel conformal transmitarray with beam steering ability is presented. First, an ultra-thin transmitarray element consisting of three layers of identical square ring slots is developed. The element has a thickness of 0.508 mm (0.04 wavelength in the free space at 25 GHz), achieving a transmission phase range of 330° with a maximum 3.6 dB loss. The element is then applied to a curved transmitarray conformal to a cylindrical surface fed by a standard gain horn with about a 10-dBi gain. A prototype is fabricated radiating a boresight beam with a peak measured gain of 19.6 dBi and an aperture efficiency of 25.1%. Second, when the transmitting surface of the above array is divided into two parts from the middle with different main beam directions, the combined beam can be radiated to an oblique angle with respect to the boresight direction. Using this method, a mechanical beam scanning conformal transmitarray antenna is designed. Its size is about 2.5 times larger than the fixed-beam one and consists of six transmitting surfaces with main beams directed to different angles. By rotating the feed horn to different positions, the main beam of the array can be switched to ±15°, ±10°, ±5°, and 0°. A prototype is fabricated with a stable gain of about 18.7 dBi at all beam angles.
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.
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, S & 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.
Seifollahi, S, Bagirov, A, Zare Borzeshi, E & Piccardi, M 2019, 'A simulated annealing‐based maximum‐margin clustering algorithm', Computational Intelligence, vol. 35, no. 1, pp. 23-41.
View/Download from: Publisher's site
View description>>
AbstractMaximum‐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, pp. 163112-163112.
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. 2355-2355.
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.
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-$\mu$ 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, H-H, 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>>
© 1998-2012 IEEE. 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.
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, no. 10, pp. 6451-6459.
View/Download from: Publisher's site
View description>>
© 2019, The Minerals, Metals & Materials Society. 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.
Usman, M, He, X, Lam, K-M, Xu, M, Bokhari, SMM, Chen, J & 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.
Vu, TT, Nguyen, DN, Hoang, DT, Dutkiewicz, E & Nguyen, TV 2019, 'Optimal Energy Efficiency with Delay Constraints for Multi-layer Cooperative Fog Computing Networks'.
View description>>
We develop a joint offloading and resource allocation framework for amulti-layer cooperative fog computing network, aiming to minimize the totalenergy consumption of multiple mobile devices subject to their service delayrequirements. The resulting optimization involves both binary (offloadingdecisions) and real variables (resource allocations), making it an NP-hard andcomputationally intractable problem. To tackle it, we first propose an improvedbranch-and-bound algorithm (IBBA) that is implemented in a centralized manner.However, due to the large size of the cooperative fog computing network, thecomputational complexity of the proposed IBBA is relatively high. To speed upthe optimal solution searching as well as to enable its distributedimplementation, we then leverage the unique structure of the underlying problemand the parallel processing at fog nodes. To that end, we propose a distributedframework, namely feasibility finding Benders decomposition (FFBD), thatdecomposes the original problem into a master problem for the offloadingdecision and subproblems for resource allocation. The master problem (MP) isthen equipped with powerful cutting-planes to exploit the fact of resourcelimitation at fog nodes. The subproblems (SP) for resource allocation can findtheir closed-form solutions using our fast solution detection method. These(simpler) subproblems can then be solved in parallel at fog nodes. Thenumerical results show that the FFBD always returns the optimal solution of theproblem with significantly less computation time (e.g., compared with thecentralized IBBA approach). The FFBD with the fast solution detection method,namely FFBD-F, can reduce up to $60\%$ and $90\%$ of computation time,respectively, compared with those of the conventional FFBD, namely FFBD-S, andIBBA.
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>>
© 1972-2012 IEEE. Due to the limited dynamic range of the off-the-shelf 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 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, pp. 1-9.
View/Download from: Publisher's site
View description>>
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, pp. 100109-100109.
View/Download from: Publisher's site
Wang, X, Zha, X, Ni, W, Liu, RP, Guo, YJ, Niu, X & Zheng, K 2019, 'Game Theoretic Suppression of Forged Messages in Online Social Networks', IEEE Transactions on Systems, Man, and Cybernetics: Systems, pp. 1-11.
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', Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 3, no. 1, pp. 1-30.
View/Download from: Publisher's site
View description>>
Capturing RFID tags in the region of interest (ROI) is challenging. Many issues, such as multipath interference, frequency-dependent hardware characteristics and phase periodicity, make RF phase difficult to accurately indicate the tag-to-antenna distance for RFID tag localization. In this paper, we propose a comprehensive solution, called RF-Focus, which fuses RFID and computer vision (CV) techniques to recognize and locate moving RFID-tagged objects within ROI. Firstly, we build a multipath propagation model and propose a dual-antenna solution to minimize the impact of multipath interference on RF phase. Secondly, by extending the multipath model, we estimate phase shifts due to hardware characteristics at different operating frequencies. Thirdly, to minimize the tag position uncertainty due to RF phase periodicity, we leverage CV to extract image regions of being likely to contain ROI RFID-tagged objects, and then associate them with the processed RF phase after the removal of the phase shifts due to multipath interference and hardware characteristics for recognition and localization. Our experiments demonstrate the effectiveness of multipath modelling and hardware-related phase shift estimation. When five RFID-tagged objects are moving in the ROI, RF-Focus achieves the average recognition accuracy of 91.67% and localization accuracy of 94.26% given a false positive rate of 10%.
Wei, F, Yang, Z-J, Qin, P-Y, Guo, YJ, Li, B & Shi, X-W 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.
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, 'Exploiting Spatial-Wideband Effect for Fast AoA Estimation at Lens Antenna Array', IEEE Journal of 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, '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, Y, Wong, S, Lin, J, Yang, Y, Zhang, L, Choi, W, Zhu, L & He, Y 2019, 'Design of triple‐band and triplex slot antenna using triple‐mode cavity resonator', IET Microwaves, Antennas & 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 of 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
View description>>
© 2019 IEEE. With the emergence of the mission-critical Internet of Things applications, ultra-reliable low-latency communications are attracting a lot of attentions. Non-orthogonal multiple access (NOMA) with multiple-input multiple-output (MIMO) is one of the promising candidates to enhance connectivity, reliability, and latency performance of the emerging applications. In this paper, we derive a closed-form upper bound for the delay target violation probability in the downlink MIMO-NOMA, by applying stochastic network calculus to the Mellin transforms of service processes. A key contribution is that we prove that the infinite-length Mellin transforms resulting from the non-negligible interferences of NOMA are Cauchy convergent and can be asymptotically approached by a finite truncated binomial series in the closed form. By exploiting the asymptotically accurate truncated binomial series, another important contribution is that we identify the critical condition for the optimal power allocation of MIMO-NOMA to achieve consistent latency and reliability between the receivers. The condition is employed to minimize the total transmit power, given a latency and reliability requirement of the receivers. It is also used to prove that the minimal total transmit power needs to change linearly with the path losses, to maintain latency and reliability at the receivers. This enables the power allocation for mobile MIMO-NOMA receivers to be effectively tracked. The extensive simulations corroborate the accuracy and effectiveness of the proposed model and the identified critical condition.
Xu, J-X, Li, H-Y, 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.
Yang, T, Ding, C & Guo, YJ 2019, 'A Highly Birefringent and Nonlinear AsSe<inline-formula> <tex-math notation='LaTeX'>$_2$</tex-math> </inline-formula>–As<inline-formula> <tex-math notation='LaTeX'>$_2$</tex-math> </inline-formula>S<inline-formula> <tex-math notation='LaTeX'>$_5$</tex-math> </inline-formula> Photonic Crystal Fiber With Two Zero-Dispersion Wavelengths', IEEE Photonics Journal, vol. 11, no. 1, pp. 1-7.
View/Download from: Publisher's site
View description>>
© 2018 IEEE. A hybrid AsSe2-As2S5 photonic crystal fiber (PCF) with a solid elliptical core is proposed and studied theoretically by the full-vector finite element method. The core and cladding of the PCF are made of AsSe2 and As2S5 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-1m-1 for the X- A nd Y-polarized (X and Y-pol) modes, respectively. Moreover, it is able to achieve two zero-dispersion wavelengths 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 the 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, pp. 2442-2442.
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.
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, pp. 111635-111635.
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
View description>>
© 1967-2012 IEEE. This paper analyzes the secrecy capacity of a cooperative relaying system using non-orthogonal multiple access (NOMA). A new cooperative NOMA scheme is proposed, where the source actively sends jamming signals while the relay is forwarding, thereby enhancing the security of intended communication links. Closed-form expressions for the ergodic secrecy rate are derived in the presence of an eavesdropper. Asymptotic approximate expressions for the ergodic secrecy rate are established in high signal-to-noise ratio (SNR) regime, which provides insights on secure NOMA transmission. Numerical results reveal the critical condition, under which NOMA is able to outperform orthogonal multiple access (OMA) in terms of secrecy rate. The proposed NOMA scheme can improve the secrecy rate by about 78.1rm%.
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. 6952-6965.
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, NC & 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
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
View description>>
© 2019 Elsevier Ltd The automatic image annotation can provide semantic illustrations to understand image contents, and builds a foundation to develop algorithms that can search images within a large database. However, most current methods focus on solving the annotation problem by modeling the image visual content and tag semantic information, which overlooks the additional information, such as scene descriptions and locations. Moreover, the majority of current annotation datasets are visually consistent and only annotated by common visual objects and attributes, which makes the classic methods vulnerable to handle the more diverse image annotation. To address above issues, we propose to annotate images via collective knowledge, that is, we uncover relationships between the image and its neighbors by measuring similarities among metadata and conduct the metric learning to obtain the representations of image contents, we also generate semantic representations for images given collective semantic information from their neighbors. Two representations from different paradigms are embedded together to train an annotation model. We ground our model on the heritage image collection we collected from the library online open data. Annotations on the heritage image collection are not limited to common visual objects, and are highly relevant to historical events, and the diversity of the heritage image content is much larger than the current datasets, which makes it more suitable for this task. Comprehensive experimental results on the benchmark dataset indicate that the proposed model achieves the best performance compared to baselines and state-of-the-art methods.
Zhang, R, Mu, C, Xu, M, Xu, L & Xu, 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
View description>>
© 2019 Elsevier B.V. Graphene based three-layer compound film on the silicon substrate is formed by gold deposition of electron beam evaporation (EBE) and graphene transfer. Processed with different high temperature annealing in nitrogen, the film with residual tensile stress of 52.58 MPa at 500 ℃ can be achieved by using an X-ray diffraction (XRD) method. Based on this low stress film, a series of long lifecycle MEMS double-clamped beams (DCBs) are fabricated by the standard MEMS manufacturing technology. The achieved beam can be turned on/off for up to 100 million times at the pull-in voltage of 30 V, which is compatible with the conventional, complementary metal-oxide-semiconductor (CMOS) circuit requirements.
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
View description>>
© 2013 IEEE. Recently, large scale seismic data acquisition has been a critical method for scientific research and industrial production. However, due to the bottleneck on data transmission and the limitation of energy storage, it is hard to conduct large seismic data acquisition in a real-time way. So, in this paper, an efficient seismic data acquisition method, namely, compressed sensing architecture with generative adversarial networks (CSA-GAN), is proposed to tackle the two restrictions of collecting large scale seismic data. In the CSA-GAN, a data collection architecture based on compressed sensing theory is applied to reduce data traffic load of the whole system, as well as balance the data transmission. To make the compressed sensing procedure perform well in both data quality and compression ratio, a kind of generative adversarial networks is designed to learn the recovering map. According to our experiment results, a high data quality (about 30 dB) at the compression ratio of 16 is achieved by the proposed approach, which enables the system to afford 15 times more sensors and reduces the energy cost by means of data collection from N(N + 1)/2 to N2/16. These results show that the CSA-GAN can afford more sensors with the same bandwidth and consume less energy, via improving the efficiency seismic data acquisition.
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>>
© 1999-2012 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 meth...
Zhou, W, Sutton, GJ, Andrew Zhang, J, 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
View description>>
© 2012 IEEE. Licensed-assisted-access (LAA) is used to extend the LTE link into the unlicensed band. How to guarantee the quality-of-service (QoS) for LTE devices in the unlicensed band is a challenging problem due to the listen-before-talk contention access in 5-GHz unlicensed bands. In this letter, we quantitatively analyze the medium access control delay for tagged LAA eNBs and propose a delay-guaranteed admission control scheme. We consider the freezing time of busy slots caused by collision or successful transmission, and introduce the exponential backoff mechanism for delay analysis. Validated by simulation results, our method provides important insights into the system admission performance and fairness of access.
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, J-Y & 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, H, 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>>
© 1963-2012 IEEE. Wideband multibeam 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, analog 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 multibeam arrays.
Abdollahi, M, Abolhasan, M, Shariati, N, Lipman, J, Jamalipour, A & Ni, W 1970, 'A Routing Protocol for SDN-based Multi-hop D2D Communications', 2019 16th IEEE Annual Consumer Communications & Networking Conference (CCNC), 2019 16th IEEE Annual Consumer Communications & Networking Conference (CCNC), 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 1970, 'An Adaptive UAV Network for Increased User Coverage and Spectral Efficiency', 2019 IEEE Wireless Communications and Networking Conference (WCNC), 2019 IEEE Wireless Communications and Networking Conference (WCNC), 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.
Acut, RVP, Hora, JA, Gerasta, OJL, Zhu, X & Dutkiewicz, E 1970, 'PV-TEG- WiFi Multiple Sources Design Energy Harvesting System for WSN Application', 2019 IEEE International Circuits and Systems Symposium (ICSyS), 2019 IEEE International Circuits and Systems Symposium (ICSyS), IEEE.
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.
Al-Doghman, F, Chaczko, Z, Brooke, W & Gordon, LC 1970, 'Social Consensus-inspired Aggregation Algorithms for Edge Computing', 2019 3rd Cyber Security in Networking Conference (CSNet), 2019 3rd Cyber Security in Networking Conference (CSNet), IEEE, 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 & Beydoun, G 1970, 'Using Adaptive Enterprise Architecture Framework for Defining the Adaptable Identity Ecosystem Architecture', https://aisel.aisnet.org/acis2019/, Australasian Conference on Information Systems, AIS, Perth, pp. 1-11.
View description>>
Digital identity management is often used to handle fraud detection and hence reduce identity thefts. However, using digital identity management presents additional challenges in terms of privacy of the identity owner meanwhile managing the security of the verification. In this paper, drawing on adaptive enterprise architecture (EA) with an ecosystem approach to digital identity, we describe an identity ecosystem (IdE) architecture to handle identity management (IdM) while safeguarding security and privacy. This study is a part of the larger action design research project with our industry partner DZ. We have used Adaptive EA as a baseline to define a privacy aware adaptive IdE to make ID operations more efficient and improve the delivery of services in the public and private sector. The value of the anticipated architecture is in its generic yet comprehensive structure, component orientation and layered approach which aim to enable the contemporary IdM
Anwar, MJ & Gill, AQ 1970, 'A Review of the Seven Modelling Approaches for Digital Ecosystem Architecture.', CBI (1), IEEE Conference on Business Informatics, IEEE, Moscow, Russia, pp. 94-103.
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.
Anwar, MJ, Gill, AQ & Beydoun, G 1970, 'Using Adaptive Enterprise Architecture Framework for Defining the Adaptable Identity Ecosystem Architecture', ACIS 2019 Proceedings - 30th Australasian Conference on Information Systems, pp. 890-900.
View description>>
Digital identity management is often used to handle fraud detection and hence reduce identity thefts. However, using digital identity management presents additional challenges in terms of privacy of the identity owner meanwhile managing the security of the verification. In this paper, drawing on adaptive enterprise architecture (EA) with an ecosystem approach to digital identity, we describe an identity ecosystem (IdE) architecture to handle identity management (IdM) while safeguarding security and privacy. This study is a part of the larger action design research project with our industry partner DZ. We have used adaptive EA as a theoretical lens to define a privacy aware adaptive IdM with a view to improve the Id operations and delivery of services in the public and private sector. The value of the anticipated architecture is in its generic yet comprehensive structure, component orientation and layered approach which aim to enable the contemporary IdM.
Ashtari, S, Tofigh, F, Abolhasan, M, Lipman, J & Ni, W 1970, 'Efficient Cellular Base Stations Sleep Mode Control Using Image Matching', 2019 IEEE 89th Vehicular Technology Conference (VTC2019-Spring), 2019 IEEE 89th Vehicular Technology Conference (VTC2019-Spring), IEEE, Kuala Lumpur, MALAYSIA.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Green cellular network helps to decrease environmental pollution. In contrast, massive connectivity and demand for higher data rate promise the presence of new generation of cellular system (5G) and small cell networks. Hence, expectation on increasing the number of base stations (BSs), which leads to increase in energy usage. One way to improve energy consumption is by shutting down the redundant BSs while sustaining the Quality-of-Service (QoS) for each user. In this paper, we propose a dynamic structural algorithm based on transportation problem, to switch on/off the BSs in cellular networks without compromising its coverage, and maintain the networks load by neighboring cells. We use weighted graphs to translate our problem as a transportation problem and then use linear programming to solve it. The cost of transport, turning a BS into sleep mode, is illustrated as a function of energy usage,coverage area and load on the BSs. Running the propose method consecutively provides the maximum number of BSs whom are at sleep mode. The methodology explained in this paper reduces energy consumption to almost 40%, whereas maintaining all the existing loads in the network.
Bah, AO, Bird, TS & Qin, P 1970, 'A Low Profile Tightly Coupled Antenna Array with 80° Scanning for Multifunctional Applications', 2019 IEEE International Symposium on Phased Array System & Technology (PAST), 2019 IEEE International Symposium on Phased Array System & Technology (PAST), IEEE, Waltham, MA, USA.
View/Download from: Publisher's site
View description>>
A wideband wide scanning antenna array for application in multifunctional phased arrays is presented. The dipoles and balun are printed on both sides of a single RT/Duroid™ 6010 substrate with a relative dielectric constant of 10.2. Optimized designs of two thicknesses of a metasurface-based wide angle impedance matching layer are presented, facilitating the highest figure of merit values in phased array antennas. The feed network, composed of meandered impedance transformer and balun sections, are constructed from Klopfenstein tapered microstrip lines. The overall height of the array above the ground plane is 0.087 $\lambda_{\mathrm{L}}$ , where $\lambda_{\mathrm{L}}$ is the wavelength at the lowest frequency of operation. For the single sided metasurface design, scanning to 80° along the E-plane and 55° along the H-plane over a 5.5:1 impedance bandwidth (0.77 GHz-4.2 GHz) was achieved assuming an active VSWR value of 3.1.
Bautista, M, Zhu, H, Zhu, X & Yang, Y 1970, 'Design of Self-Coupling Enhanced Resonator in $0.13-\mu\mathrm{m}$ (Bi)-CMOS Technology', 2019 International Conference on Microwave and Millimeter Wave Technology (ICMMT), 2019 International Conference on Microwave and Millimeter Wave Technology (ICMMT), IEEE, Guangzhou, China, pp. 1-27.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Design of a compact on-chip resonator operating at mm-wave region is presented in this paper. Unlike the previously published ones in the literature, this work is implemented using a planar structure, which requires no broadside coupling. Instead, the resonance is generated through enhanced self-coupling. To further demonstrate the feasibility of using this approach in practice, the designed resonator is fabricated in a standard 0.13- mumathrm{m} (Bi)-CMOS technology. The measured results show that it can generate a notch at 47 GHz with the attenuation better than 28 dB due to the enhanced self-resonant frequency. The chip size, excluding the pads, is only 0.096times 0.294 text{mm}{2}.
Bautista, MG, Zhu, H, Zhu, X, Yang, Y, Sun, Y, Dutkiewicz, E & Zhang, F 1970, 'Millimeter-Wave BPFs Design using Quasi-Lumped Elements in 0.13-μm (Bi)-CMOS Technology', 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 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.
Bożejko, W, Chaczko, Z, Nadybski, P & Wodecki, M 1970, '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 International Publishing, 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 1970, 'Studios in DE and EE at UTS: Structure and Rationale.', ITHET, International Conference on Information Technology Based Higher Education and Training, IEEE, Magdeburg, Germany, pp. 1-6.
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 1970, 'Assessment Design for Studio-Based Learning', Proceedings of the Twenty-First Australasian Computing Education Conference, ACE'19: Twenty-First Australasian Computing Education Conference, ACM, Sydney, Australia, pp. 106-111.
View/Download from: Publisher's site
View description>>
© 2019 Association for Computing Machinery. 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 studiobased 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.
Broomhead, T, Cremean, L, Ridoux, J & Veitch, D 1970, 'Virtualize everything but time', Proceedings of the 9th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2010, pp. 451-464.
View description>>
We propose a new timekeeping architecture for virtualized systems, in the context of Xen. Built upon a feed-forward based RADclock synchronization algorithm, it ensures that the clocks in each OS sharing the hardware derive from a single central clock in a resource effective way, and that this clock is both accurate and robust. A key advantage is simple, seamless VM migration with consistent time. In contrast, the current Xen approach for timekeeping behaves very poorly under live migration, posing a major problem for applications such as financial transactions, gaming, and network measurement, which are critically dependent on reliable timekeeping. We also provide a detailed examination of the HPET and Xen Clocksource counters. Results are validated using a hardware-supported testbed.
Calam, RCM, Hora, JA, Gerasta, OJL, Zhu, X & Dutkiewicz, E 1970, 'A Self-Calibrating Off-Time Controller for WSN/IoT Synchronous Non-Inverting Buck-Boost DC-to-DC Converter Application', 2019 IEEE International Circuits and Systems Symposium (ICSyS), 2019 IEEE International Circuits and Systems Symposium (ICSyS), IEEE.
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.
Cao, Y & Veitch, D 1970, 'Where on Earth Are the Best-50 Time Servers?', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Passive and Active Network Measurement, Springer International Publishing, 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, Klempous, R, Rozenblit, J, Chiu, C, Kluwak, K & Smutnicki, C 1970, 'Enabling Design of Middleware for Massive Scale IOT-based Systems', 2019 IEEE 23rd International Conference on Intelligent Engineering Systems (INES), 2019 IEEE 23rd International Conference on Intelligent Engineering Systems (INES), IEEE, Gödöllő, Hungary.
View/Download from: Publisher's site
View description>>
Recently, the Internet of Things (IoT) technology has rapidly advanced to the stage where it is feasible to discover, locate and identify various smart sensors and devices based on the context, situation, characteristics, and relevancy to query for their data or control actions. Taking things a step further when developing Large Scale Applications requires that two serious issues be overcome. The first issue is to find a solution for data sensing and collection from a massive number of various ubiquitous devices when converging these into the next generation networks. The second important issue is to deal with the “Big Data” that arrive from a very large number of sources. This research emphasizes the need for finding a solution for a large scale data aggregation and delivery. The paper introduces biomimetic design methods for data aggregation in the context of large scale IoT-based systems.
Chaczko, Z, Wajs-Chaczko, P, Tien, D & Haidar, Y 1970, 'Contents', 2019 International Conference on Machine Learning and Cybernetics (ICMLC), 2019 International Conference on Machine Learning and Cybernetics (ICMLC), 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.
Chen, L, Liu, Y & Guo, YJ 1970, 'Efficient Frequency-invariant Beam Pattern Synthesis With Multiple Space-frequency Nulls', 2019 IEEE International Conference on Computational Electromagnetics (ICCEM), 2019 IEEE International Conference on Computational Electromagnetics (ICCEM), 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, L, Wang, Y, Liu, Y & Guo, YJ 1970, 'Synthesis of Frequency-invariant Beam Patterns under Accurate Sidelobe Control by Second-order Cone Programming', 2019 Photonics & Electromagnetics Research Symposium - Fall (PIERS - Fall), 2019 Photonics & Electromagnetics Research Symposium - Fall (PIERS - Fall), IEEE, Xiamen, China, pp. 2260-2263.
View/Download from: Publisher's site
View description>>
It is shown in this work that the FI pattern synthesis can be treated as an optimization problem for minimizing the mainlobe frequency variation. To control both the mainlobe and sidelobe regions, we introduce several constraints imposed on the broadband pattern, called the look-direction constraint, the spatial response variation constraint and the sidelobe constraint, respectively. The whole optimization process needs to perform the SOCP solver. A synthesis of FI pattern with low sidelobe level (SLL) is given to validate the accuracy and effectiveness of the proposed method.
Chen, S-L, Karmokar, DK, Ziolkowski, RW & Guo, YJ 1970, 'Wide-Angle Wideband Frequency-Independent Beam-Scanning Leaky Wave Antenna', PROCEEDINGS OF THE 2019 INTERNATIONAL CONFERENCE ON ELECTROMAGNETICS IN ADVANCED APPLICATIONS (ICEAA), 21st International Conference on Electromagnetics in Advanced Applications (ICEAA), IEEE, SPAIN, Granada, pp. 554-557.
View/Download from: Publisher's site
Chen, S-L, Karmokar, DK, Ziolkowski, RW & Guo, YJ 1970, 'Wide-Angle Wideband Frequency-Independent Beam-Scanning Leaky Wave Antenna', 2019 International Conference on Electromagnetics in Advanced Applications (ICEAA), 2019 International Conference on Electromagnetics in Advanced Applications (ICEAA), 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.
Chen, Y, An, P, Huang, X, Meng, C & Wu, Q 1970, 'Modified Baseline for Light Field Stitching', 2019 IEEE Visual Communications and Image Processing (VCIP), 2019 IEEE Visual Communications and Image Processing (VCIP), IEEE, Australia, pp. 1-4.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In traditional 2D image stitching, the baseline method usually means global homography via Direct Linear Transformation (DLT) on inliers. In this paper, a modified baseline method for light field (LF) stitching is proposed to stitch two LFs. The depth map and the center sub-Aperture image (SAI) are used to filter the feature points of the entire LF. The global 4D homography is then calculated by DLT to align all SAIs corresponding to the same angular domain coordinates of two LFs. Finally, the improved Markov Random Field (MRF) energy considering the global LF is used to find the seam of 2D SAIs instead of computational 4D graph cut. Experimental results show that the proposed method can effectively stitch the 4D LFs, and preserve the consistency of the angular and spatial domains of the stitched LF compared with implementing 2D image stitching to the corresponding SAIs. Moreover, the method proposed in this paper can easily extend all advanced 2D image stitching methods to 4D LF, so that the acquired LF can have larger field of view and wider applications.
Chen, Z, Zhu, X & Xu, L 1970, 'Integration of mm-wave Antennas on Fan-Out Wafer Level Packaging (FOWLP) for Automotive Radar Applications', 2019 IEEE Asia-Pacific Microwave Conference (APMC), 2019 IEEE Asia-Pacific Microwave Conference (APMC), IEEE, Singapore, pp. 1607-1609.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents the integration of mm-wave antennas on fan-out wafer level packaging (FOWLP) for automotive radar applications. The size of the package is 24x24.5x0.3 mm3, Three microstrip grid array antennas are used as transmitting antennas and four patch sub arrays are used as receiving antennas. The antennas are fed by chip signal which is coupled through a slot in the ground plane. The simulation results show maximal peak realized gain of 14.28 dBi and 11.1 dBi for transmitting and receiving antennas, respectively. The angle resolution of the MIMO radar can be improved to 9.6°.
Cheng, Q, Shi, Z, Nguyen, DN & Dutkiewicz, E 1970, 'An OFDM Sensing Algorithm in Full-Duplex Systems with Self-Interference and Carrier Frequency Offset', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Hawaii.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Full duplex (FD) wireless technology, which enables simultaneous transmission and reception on the same frequency, has shown its great potential for doubling the spectral efficiency as well as spectrum sensing while transmitting in cognitive radio networks (CRNs). However, the self interference (SI) suppression, the underlying technique of FD, is often imperfect, resulting in non-negligible residual SI that severely affects the test statistics of sensing methods. The residual SI thereby significantly deteriorates the spectrum sensing accuracy. In this work, we aim to address this issue by proposing a novel sensing approach in FD systems leveraging the Pilot-Tone (PT) structure of Orthogonal Frequency Division Modulation (OFDM) signals. In comparison with the conventional sensing methods in FD systems, the developed sensing approach holds the advantage in the robustness not only to residual SI but also the carrier frequency offset (CFO). Besides, the proposed sensing method is able to accomplish sensing tasks in low SNR conditions with much lower computational complexity. Numerical simulations results demonstrate that the probability of detection of our proposed approach can be improved up to 34.9%, compared with state- of-the-art sensing methods in FD systems, suffering from residual SI and CFO.
Dasgupta, A, Gill, A & Hussain, F 1970, 'A Conceptual Framework for Data Governance in IoT-enabled Digital IS Ecosystems', Proceedings of the 8th International Conference on Data Science, Technology and Applications, 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
View description>>
Copyright © 2019 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved There is a growing interest in the use of Internet of Things (IoT) in information systems (IS). Data or information governance is a critical component of IoT enabled digital IS ecosystem. There is insufficient guidance available on how to effectively establish data governance for IoT enabled digital IS ecosystem. The introduction of new regulations related to privacy such as General Data Protection Regulation (GDPR) as well as existing regulations such as Health Insurance Portability and Accountability Act (HIPPA) has added complexity to this issue of data governance. This could possibly hinder the effective IoT adoption in healthcare digital IS ecosystem. This paper enhances the 4I framework, which is iteratively developed and updated using the design science research (DSR) method to address this pressing need for organizations to have a robust governance model to provide the coverage across the entire data lifecycle in IoT-enabled digital IS ecosystem. The 4I framework has four major phases: Identify, Insulate, Inspect and Improve. The application of this framework is demonstrated with the help of a Healthcare case study. It is anticipated that the proposed framework can help the practitioners to identify, insulate, inspect and improve governance of data in IoT enabled digital IS ecosystem.
Dasgupta, A, Gill, AQ & Hussain, FK 1970, 'A Review of General Data Protection Regulation for Supply Chain Ecosystem.', IMIS, International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing, Springer, Sydney, Australia, pp. 456-465.
View/Download from: Publisher's site
View description>>
© 2020, Springer Nature Switzerland AG. The data-intensive digital supply chain management (SCM) ecosystems seem to be impacted by the recent changes in the regulations and advancement in technologies such as Artificial Intelligence, Big Data, Analytics, Networking, IoT including proliferation of less expensive hardware devices. There is limited guidance available on how to govern the logistics sector, particularly from a regulatory compliance perspective. Through this paper, we investigate the impact of General Data Protection Regulation (GDPR) on digitized SCM. The key questions are: What are the GPDR specific legal obligations? What is the best approach to manage data access, quality, privacy, security and ownership effectively in SCM? This research paper aims to assist researchers and practitioners to understand the impact of GDPR on SCM, provide the 4I (Identify, Insulate, Inspect, Improve) Framework and its applicability to streamline the GDPR compliance activities.
Ding, C, Sun, H-H, Jay Guo, Y & Jones, B 1970, 'Enabling the Co-Existence of Multiband Antenna Arrays', 2019 IEEE International Symposium on Phased Array System & Technology (PAST), 2019 IEEE International Symposium on Phased Array System & Technology (PAST), IEEE, Waltham, MA USA.
View/Download from: Publisher's site
View description>>
This paper identifies a kind of interaction mechanism that has not been well addressed before, the crossband scattering in multi-band antenna arrays. First the crossband scattering effect is demonstrated on an interleaved dual- band base station antenna array section as an example. Then an effective de-scattering method is proposed, which is to insert chokes on low band antennas. The working mechanism and principle of the chokes are also presented in this paper. Finally, this method is applied on the base station antenna array section to demonstrate its effectiveness.
Ding, C, Wang, K & Guo, YJ 1970, 'Building Antennas on Perovskite Solar Cell (PSC) for Hybrid Solar/EM Wireless Energy Harvesting and Transfer', 2018 Asian Wireless Power Transfer Workshop (AWPT), 2018 Asian Wireless Power Transfer Workshop (AWPT), Sendai, Japan.
Dinh, TH, Alsheikh, MA, Gong, S, Niyato, D, Han, Z & Liang, Y-C 1970, 'Defend Jamming Attacks: How to Make Enemies Become Friends', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Waikoloa, HI.
View/Download from: Publisher's site
View description>>
In this paper, we consider a smart jammer that only attacks the channel if it detects activities of legitimate devices on that channel. To cope with smart jamming attacks, we propose an intelligent deception strategy in which the legitimate device will send fake transmissions to lure the jammer. Then, if the jammer launches attacks to the channel, the legitimate device can either backscatter the jamming signals to transmit data or harvest energy from the jamming signals for future active transmission. In this way, we can not only undermine the attack ability of the jammer, but also leverage jamming attacks as means to enhance system performance. In addition, to find an optimal defense strategy for the legitimate device under uncertainty of wireless environment as well as incomplete information from the jammer, we develop Q-learning and deep Q-learning algorithms based on the Markov decision process. Through simulation results, we demonstrate that our proposed solution is able to not only deal with smart jamming attacks, but also successfully leverage jamming attacks to improve the system performance.
Du, A, Huang, X, Zhang, J, Yao, L & Wu, Q 1970, 'Kpsnet: Keypoint Detection and Feature Extraction for Point Cloud Registration', 2019 IEEE International Conference on Image Processing (ICIP), 2019 IEEE International Conference on Image Processing (ICIP), 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.
Gamal, M, Abolhasan, M, jafarizadeh, S, Lipman, J & Ni, W 1970, 'Mapping and Scheduling of Virtual Network Functions using Multi Objective Optimization Algorithm', 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 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 1970, 'Mapping and Scheduling for Non-Uniform Arrival of Virtual Network Function (VNF) Requests', 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 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 1970, 'Ultrasensitive Terahertz High-Tc Superconducting Receivers', 2019 IEEE MTT-S International Wireless Symposium (IWS), 2019 IEEE MTT-S International Wireless Symposium (IWS), IEEE, Guangzhou, PEOPLES R CHINA.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper reviews our major technical achievement in high-Tc superconducting (HTS) terahertz (THz) receivers or heterodyne mixers in recent years. By virtue of innovative on-chip antenna/circuit designs, accurate device modelling and simulation, and advanced YBa2Cu3O7-x (YBCO) step-edge junction technology, we have successfully developed a series of HTS Josephson THz mixers with superior performance in terms of operating temperature, intermediate-frequency (IF) bandwidth, conversion gain and noise temperature. These mixers serve as promising receiver frontends for THz wireless communication and sensing systems.
He, Y, Jayawickrama, BA & Dutkiewicz, E 1970, 'Distributed Power Allocation Algorithm for General Authorised Access in Spectrum Access System', 2019 IEEE Wireless Communications and Networking Conference (WCNC), 2019 IEEE Wireless Communications and Networking Conference (WCNC), 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.
Hora, JA, Arellano, AC, Zhu, X & Dutkiewicz, E 1970, 'Design of Buck Converter with Dead-time Control and Automatic Power-Down System for WSN Application', 2019 IEEE Wireless Power Transfer Conference (WPTC), 2019 IEEE Wireless Power Transfer Conference (WPTC), IEEE, 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 1970, 'A Highly Linear OTA with 759 µS gm for RF Transceiver Application', 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 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, Mayormita, MA, Rebollos, JRC, Zhu, X & Dutkiewicz, E 1970, 'On-Chip Inductor-Less Indoor Light Energy Harvester with Improved Efficiency for WSN/IoT Device Design', 2019 13th International Conference on Sensing Technology (ICST), 2019 13th International Conference on Sensing Technology (ICST), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. An on-chip inductor-less indoor light energy harvester circuit block for internet-of-things wireless sensor node device design is implemented in 65nm CMOS process technology. The design of the indoor light energy harvester comprises a bootstrapped ring oscillator, a two-phase non-overlapping clock generator, a tapered buffer, a multi-stage differential-drive CMOS rectifier, a charge controlling circuit and a voltage regulator. The system boosts an input of 500 mV from a photovoltaic cell without using a typical boost converter circuit that employs an inductor element. Hence, a simplified on-chip design charges an external 1.3-V rechargeable battery. The use of a multi-stage differential-drive rectifier eliminates the need for expensive on-chip inductors. A charge control circuit is implemented to maintain the battery voltage and avoid overcharging, thus improving battery life. A low-dropout voltage regulator further regulates the battery voltage to produce a stable dc voltage. The chip core design has a total area of 1342μm×1011μm. The output of the harvesting system is a regulated 0.9 V supply with 1.05 mA current at full load.
Hora, JA, Zhu, X & Dutkiewicz, E 1970, '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 IEEE Wireless Power Transfer Conference (WPTC), IEEE, 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 1970, 'Design of High Voltage Output for CMOS Voltage Rectifier for Energy Harvesting Design', 2019 IEEE Wireless Power Transfer Conference (WPTC), 2019 IEEE Wireless Power Transfer Conference (WPTC), IEEE, 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.
Hora, JA, Zhu, X & Dutkiewicz, E 1970, 'Simplified Over- Temperature Protection Circuit Structure for WSN/IoT Device Power Management', 2019 13th International Conference on Sensing Technology (ICST), 2019 13th International Conference on Sensing Technology (ICST), IEEE, Sydney, Australia,.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents a modified structure of the over-temperature protective circuit integrated into the power management converter design for wireless sensor node devices. The design is focused on realizing a simplified circuit structure that is compatible with standard CMOS technology structure and evaluating the over-temperature threshold to assure needed accuracy. HSPICE simulation using Monte Carlo Analysis for the bandgap voltage reference is used to get a better estimation, process variation, and reliability. The target design temperature threshold is obtained at approximately 150 °C, which is the standard chip testing value at worst temperature consideration. Post-layout simulation of the proposed circuit design structure is carried out using 65nm 1P9M CMOS 1.2V/2.5V logic CMOS technology. And it is co-integrated in the power management circuit design for the system-on-chip wireless sensor node device.
Huang, H, Liu, Y, Chen, L, Qin, P-Y & Guo, YJ 1970, 'Synthesis of a Dipole Array with Optimally End-fire Directive Pattern', 2019 IEEE International Conference on Computational Electromagnetics (ICCEM), 2019 IEEE International Conference on Computational Electromagnetics (ICCEM), 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, Zhang, J, Zhang, J, Wu, Q & Xu, J 1970, 'Compare More Nuanced: Pairwise Alignment Bilinear Network for Few-Shot Fine-Grained Learning', 2019 IEEE International Conference on Multimedia and Expo (ICME), 2019 IEEE International Conference on Multimedia and Expo (ICME), 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 1970, 'Towards Terabit Wireless Communications', 2019 Australian Communications Theory Workshop, 2019 Australian Communications Theory Workshop, Sydney, Australia.
Huang, X, Fan, L, Wu, Q, Zhang, J & Yuan, C 1970, '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>>
Many types of 3D acquisition sensors have emerged in recent years and pointcloud has been widely used in many areas. Accurate and fast registration ofcross-source 3D point clouds from different sensors is an emerged researchproblem in computer vision. This problem is extremely challenging becausecross-source point clouds contain a mixture of various variances, such asdensity, partial overlap, large noise and outliers, viewpoint changing. In thispaper, an algorithm is proposed to align cross-source point clouds with bothhigh accuracy and high efficiency. There are two main contributions: firstly,two components, the weak region affinity and pixel-wise refinement, areproposed to maintain the global and local information of 3D point clouds. Then,these two components are integrated into an iterative tensor-based registrationalgorithm to solve the cross-source point cloud registration problem. Weconduct experiments on synthetic cross-source benchmark dataset and realcross-source datasets. Comparison with six state-of-the-art methods, theproposed method obtains both higher efficiency and accuracy.
Huang, X, Zhang, H, Zhang, JA, Guo, YJ, Song, R-L, Xu, X-F, Wang, C-T, Lu, Z & Wu, W 1970, 'Dual Pulse Shaping Transmission with Complementary Nyquist Pulses', 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 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, Y, Wu, Q, Xu, J & Zhong, Y 1970, 'Celebrities-ReID: A Benchmark for Clothes Variation in Long-Term Person Re-Identification', 2019 International Joint Conference on Neural Networks (IJCNN), 2019 International Joint Conference on Neural Networks (IJCNN), 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 1970, 'SBSGAN: Suppression of Inter-Domain Background Shift for Person Re-Identification', 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, Seoul, South Korea, pp. 9526-9535.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Cross-domain person re-identification (re-ID) is challenging due to the bias between training and testing domains. We observe that if backgrounds in the training and testing datasets are very different, it dramatically introduces difficulties to extract robust pedestrian features, and thus compromises the cross-domain person re-ID performance. In this paper, we formulate such problems as a background shift problem. A Suppression of Background Shift Generative Adversarial Network (SBSGAN) is proposed to generate images with suppressed backgrounds. Unlike simply removing backgrounds using binary masks, SBSGAN allows the generator to decide whether pixels should be preserved or suppressed to reduce segmentation errors caused by noisy foreground masks. Additionally, we take ID-related cues, such as vehicles and companions into consideration. With high-quality generated images, a Densely Associated 2-Stream (DA-2S) network is introduced with Inter Stream Densely Connection (ISDC) modules to strengthen the complementarity of the generated data and ID-related cues. The experiments show that the proposed method achieves competitive performance on three re-ID datasets, i.e., Market-1501, DukeMTMC-reID, and CUHK03, under the cross-domain person re-ID scenario.
Jauregi Unanue, I, Zare Borzeshi, E, Esmaili, N & Piccardi, M 1970, 'ReWE: Regressing Word Embeddings for Regularization of Neural Machine Translation Systems', Proceedings of the 2019 Conference of the North, Proceedings of the 2019 Conference of the North, Association for Computational Linguistics, Minneapolis, 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 1970, '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>>
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 1970, 'Self-Biased 2.4 GHz CMOS RF-to-DC Converter with 80% Efficiency and −22.04 dBm Sensitivity for Wi-Fi Energy Harvesting', 2019 IEEE International Circuits and Systems Symposium (ICSyS), 2019 IEEE International Circuits and Systems Symposium (ICSyS), IEEE.
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, Bird, TS, Guo, YJ & Esselle, KP 1970, 'A Binary-switch Controlled Periodic Half-width Leaky-wave Antenna for Fixed Frequency Beam Steering near the Endfire Region', 2019 PhotonIcs & Electromagnetics Research Symposium - Spring (PIERS-Spring), 2019 PhotonIcs & Electromagnetics Research Symposium - Spring (PIERS-Spring), IEEE, Rome, Italy, pp. 1799-1803.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. A leaky-wave antenna (LWA) with fixed-frequency beam scanning capabilities is presented in this paper. The main structure is a half-width microstrip LWA (HW-MLWA), and the direction of the main beam at a fixed operating frequency is controlled by using a group of gap capacitors and binary switches. Different binary switching patterns are applied to change the reactance profile of the radiating structure and hence beam scanning at a fixed frequency is achieved in discrete steps. Results from the full-wave simulation demonstrate that the radiating antenna beam can be steered from 38 to 68 in a discrete step at 7 GHz.
Karmokar, DK, Chen, S-L, Qin, P-Y & Guo, YJ 1970, 'Open-Stopband Suppression and Cross-Polarization Reduction of a Substrate Integrated Waveguide Leaky-Wave Antenna', 2019 URSI Asia-Pacific Radio Science Conference (AP-RASC), 2019 URSI Asia-Pacific Radio Science Conference (AP-RASC), IEEE, New Delhi, INDIA.
View/Download from: Publisher's site
View description>>
© 2019 URSI. All rights reserved. Most leaky-wave antennas (LWAs) suffer from significant gain degradation when the main beam points towards broadside. This is because an open stopband (OSB) restricts broadside radiation. In this paper, a method to suppress the OSB of a periodic substrate integrated waveguide (SIW) LWA is discussed. By simultaneously introducing a slot and a partially radiating wall in each unit cell the impedance in the OSB region has been matched and hence a continuous beam scan through broadside is achieved. The developed LWA can scan its main beam from 74° continuously to +40° when the frequency varies from 7.45 to 10.55 GHz, with a broadside gain and a level of cross-polarization for the broadside beam of 10.8 dBi and 21.37 dB, respectively.
Kekirigoda, A, Hui, K-P, Cheng, Q, Lin, Z, Zhang, JA, Nguyen, DN & Huang, X 1970, 'Massive MIMO for Tactical Ad-hoc Networks in RF Contested Environments', MILCOM 2019 - 2019 IEEE Military Communications Conference (MILCOM), MILCOM 2019 - 2019 IEEE Military Communications Conference (MILCOM), IEEE, Norfolk, VA.
View/Download from: Publisher's site
View description>>
Survivability of wireless communications segments in tactical military networks is an enormous challenge in the present and future defence forces, especially as these networks usually operate in radio frequency (RF) contested environments. Therefore, it is necessary to develop techniques to provide effective and efficient communication in RF contested environments. Massive multiple-input-multiple-output (MIMO) techniques use a large number of antennas enabling higher degrees of freedom that can improve communications network's survivability and efficiency compared to conventional MIMO or single antenna systems. This paper presents a novel massive MIMO communications system which enhances the throughput of the network, reduces the bit-error-rate and mitigates the interference from high powered jammers. Simulation results in contested environments verify the effectiveness of this system.
Le, AT, Tran, LC, Huang, X & Guo, YJ 1970, 'Authors', 2019 29th International Telecommunication Networks and Applications Conference (ITNAC), 2019 29th International Telecommunication Networks and Applications Conference (ITNAC), IEEE, Auckland, New Zealand.
View/Download from: Publisher's site
Le, AT, Tran, LC, Huang, X & Guo, YJ 1970, 'Beam-Based Analog Self-Interference Cancellation with Auxiliary Transmit Chains in Full-Duplex MIMO Systems', 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), 2019 IEEE 20th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC), IEEE, Cannes, France, pp. 1-5.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Analog domain cancellation has been considered as the most important step to mitigate self-interference (SI) in full-duplex (FD) radios. However, in FD multiple-input multiple-output (MIMO) systems, this method faces a critical issue of complexity since the number of cancellation circuits increases quadratically with the number of antennas. In this paper, we propose a beam-based radio frequency SI cancellation architecture which uses adaptive filters to significantly reduce the complexity. Data symbols for all the beams are up-converted by auxiliary transmit chains to provide reference signals for all adaptive filters. Hence, the number of cancellation circuits becomes proportional to the number of transmit beams which are much smaller than that of transmit antennas. We then show that the interference suppression ratio in this architecture is neither affected by the number of beams nor transmit or receive antennas. Instead, it is decided by the performance of the adaptive filter. Simulations are conducted to confirm the theoretical analyses.
Li, H, Wang, TQ, Huang, X & Zhang, JA 1970, 'Enhanced AoA Estimation Using Localized Hybrid Dual-Polarized Arrays', 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 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, L, Liu, Z, Zhang, J & Zhou, X 1970, 'Learn Image Object Co-segmentation with Multi-scale Feature Fusion', 2019 IEEE Visual Communications and Image Processing (VCIP), 2019 IEEE Visual Communications and Image Processing (VCIP), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Image object co-segmentation aims to segment common objects in a group of images. This paper proposes a novel neural network, which extracts multi-scale convolutional features at multiple layers via a modified VGG network and fuses them both within and across images as the intra-image and the inter-image features. Then these two kinds of features are further fused at each scale as the multi-scale co-features of common objects, and finally the multi-scale co-features are summed up and upsampled to obtain the co-segmentation results. To simplify the network and reduce the rapidly rising resource cost along with the inputs, the reduced input size, less downsampling and dilation convolution are adopted in the proposed model. Experimental results on the public dataset demonstrate that the proposed model achieves a comparable performance to the state-of-The-Art co-segmentation methods while the computation cost has been effectively reduced.
Li, Q, Wu, Q & Liu, X 1970, 'Multi-scale and Hierarchical Embedding for Polarity Shift Sensitive Sentiment Classification', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Artificial Intelligence and Security, Springer International Publishing, 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 1970, 'Bi-Level Masked Multi-scale CNN-RNN Networks for Short Text Representation', 2019 International Conference on Document Analysis and Recognition (ICDAR), 2019 International Conference on Document Analysis and Recognition (ICDAR), 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 1970, 'Unsupervised User Behavior Representation for Fraud Review Detection with Cold-Start Problem', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Pacific-Asia Conference on Knowledge Discovery and Data Mining, Springer International Publishing, China, pp. 222-236.
View/Download from: Publisher's site
View description>>
© Springer Nature Switzerland AG 2019. Detecting fraud review is becoming extremely important in order to provide reliable information in cyberspace, in which, however, handling cold-start problem is a critical and urgent challenge since the case of cold-start fraud review rarely provides sufficient information for further assessing its authenticity. Existing work on detecting cold-start cases relies on the limited contents of the review posted by the user and a traditional classifier to make the decision. However, simply modeling review is not reliable since reviews can be easily manipulated. Also, it is hard to obtain high-quality labeled data for training the classifier. In this paper, we tackle cold-start problems by (1) using a user’s behavior representation rather than review contents to measure authenticity, which further (2) consider user social relations with other existing users when posting reviews. The method is completely (3) unsupervised. Comprehensive experiments on Yelp data sets demonstrate our method significantly outperforms the state-of-the-art methods.
Li, Z, Zhang, J, Wu, Q, Gong, Y, Yi, J & Kirsch, C 1970, 'Sample Adaptive Multiple Kernel Learning for Failure Prediction of Railway Points', Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD '19: The 25th ACM SIGKDD Conference on Knowledge Discovery and 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 algorit...
Li, Z, Zou, Y, Wang, G & Zhang, J 1970, 'Scale-Informed Density Estimation for Dense Crowd Counting', 2019 IEEE Visual Communications and Image Processing (VCIP), 2019 IEEE Visual Communications and Image Processing (VCIP), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Dense crowd counting (DCC) remains challenging due to the scale variation and occlusion. Several deep learning based DCC methods have achieved the state-of-Arts on public datasets. However, experimental results show that the scale variation is still the main factor to hinder the DCC performance. In this work, we propose a scale-informed dense crowd counting method focusing on handling the negative effect caused by scale variation. More specifically, we propose a method to obtain the scale information of the patch from its GT density maps via estimating the mean value of the Gaussian kernel width and then a scale-classifier is deigned and trained accordingly. Moreover, with the estimated scale information, two sub-nets are dedicatedly deigned to learn the density maps for large-scale head patch and small-scale patch separately. Experimental results validate the performance of our proposed method which achieves the best performance on three dense crowd datasets.
Liao, Q, Wang, D, Holewa, H & Xu, M 1970, 'Squeezed Bilinear Pooling for Fine-Grained Visual Categorization', 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), IEEE, South Korea.
View/Download from: Publisher's site
Lin, J-Y, Wong, S-W, Yang, Y & Zhu, L 1970, 'Cavity Balanced-to-Unbalanced Magic-T with Filtering Response', 2019 IEEE MTT-S International Microwave Symposium (IMS), 2019 IEEE/MTT-S International Microwave Symposium - IMS 2019, 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, Z, Lv, T, Zhang, JA & Liu, RP 1970, '3D Wideband mmWave Localization for 5G Massive MIMO Systems', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Waikoloa, HI, USA, pp. 1-6.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper proposes a novel 3D localization method for wideband mmWave massive MIMO systems. A high dimensional linear interpolation (HDLI)-based preprocessing is first proposed to transform the frequency-associated dynamical array response vectors into the common counterparts at the reference frequency. Through this method, the received data in all frequency bands can be processed jointly, and thus the high temporal resolution provided by wideband mmWave systems can be fully exploited for position estimation. To reduce the computational complexity in the process of the parameter estimation, we then present a wideband beamspace (WBS)-based parameter estimation algorithm to estimate the angle and delay in the low-dimensional beamspace. By exploiting the quasi- optical propagation at the mmWave frequencies, a novel positioning scheme is also designed to determine the 3D location of the target. According to our analysis and simulation results, the proposed method is capable of achieving significantly reduced computational complexity, while maintaining high localization accuracy.
Liu, J, Chaczko, Z, Braun, R & Gudzbeler, G 1970, 'Collaborative RFID Agent Simulation in Dynamic Environment', 2019 18th International Conference on Information Technology Based Higher Education and Training (ITHET), 2019 18th International Conference on Information Technology Based Higher Education and Training (ITHET), 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, Liu, F, Qin, P-Y & Guo, YJ 1970, 'Recent development in nonuniformly spaced array synthesis methods', 2019 IEEE International Symposium on Phased Array System & Technology (PAST), 2019 IEEE International Symposium on Phased Array System & Technology (PAST), IEEE, Waltham, MA, USA,.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Synthesis of sparse arrays with reduced number of elements are significant for some applications where the available space, weight and the cost of the antenna system is very limited. In recent years, a variety of advanced techniques have been presented to deal with sparse array synthesis problems in either narrow and wideband cases. This paper presents a brief review of these techniques, and gives rough comparative study on some of sparse array synthesis methods.
Luo, Y, Zhang, JA, Huang, S, Pan, J & Huang, X 1970, 'Quantization with Combined Codebook for Hybrid Array using Two-Phase-Shifter Structure', ICC 2019 - 2019 IEEE International Conference on Communications (ICC), ICC 2019 - 2019 IEEE International Conference on Communications (ICC), 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.
Luo, Y, Zhang, JA, Ni, W, Pan, J & Huang, X 1970, 'Constrained Multibeam Optimization for Joint Communication and Radio Sensing', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Hawaii, USA.
View/Download from: Publisher's site
View description>>
Multibeam technology has recently been proposed for joint communication and radio sensing (JCAS) in millimeter wave systems using analog antenna arrays. Generation of the multibeam satisfying both communication and sensing requirements is yet to be developed. In this paper, we develop closed-form solutions for optimizing the coefficient that combines communication and sensing subbeams to generate a multibeam. Our solutions maximize the received signal power for communication, in the cases (1) without constraint on sensing beamforming (BF) waveform, (2) with minimum BF gain constraints on discrete sensing directions, and (3) with a minimum total power constraint on a range of sensing directions. Simulation results are provided and validate the effectiveness of the proposed solutions.
Makhdoom, I, Zhou, I, Abolhasan, M, Lipman, J & Ni, W 1970, '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, 16th International Conference on Security and Cryptography, SCITEPRESS - Science and Technology Publications, 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, R 1970, 'Trust and Reputation in Vehicular Networks: A Smart Contract-Based Approach', 2019 18th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/13th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE), 2019 18th IEEE International Conference On Trust, Security And Privacy In Computing And Communications/13th IEEE International Conference On Big Data Science And Engineering (TrustCom/BigDataSE), IEEE, Rotorua, New Zealand, pp. 34-41.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Appending digital signatures and certificates to messages guarantee data integrity and ensure non-repudiation, but do not identify greedy authenticated nodes. Trust evolves if some reputable and trusted node verifies the node, data and evaluates the trustworthiness of the node using an accurate metric. But, even if the verifying party is a trusted centralized party, there is opacity and obscurity in computed reputation rating. The trusted party maps it with the node's identity, but how is it evaluated and what inputs derive the reputation rating remains hidden, thus concealment of transparency leads to privacy. Besides, the malevolent nodes might collude together for defamatory actions against reliable nodes, and eventually bad mouth these nodes or praise malicious nodes collaboratively. Thus, we cannot always assume the fairness of the nodes as the rating they give to any node might not be a fair one. In this paper, we propose a smart contract-based approach to update and query the reputation of nodes, stored and maintained by IPFS distributed storage. The use case particularly deals with an emergency scenario, dealing against colluding attacks. Our scheme is implemented using MATLAB simulation. The results show how smart contracts are capable of accurately identifying trustworthy nodes and record the reputation of a node transparently and immutably.
Minh, HN, Minh, HH, Dinh, TH, Nguyen, DN, Dutkiewicz, E & The, TT 1970, 'An Efficient Algorithm for the k-Dominating Set Problem on Very Large-Scale Networks', COMPUTATIONAL DATA AND SOCIAL NETWORKS, 8th International Conference on Computational Data and Social Networks (CSoNet), SPRINGER INTERNATIONAL PUBLISHING AG, Ho Chi Minh City, VIETNAM, pp. 74-76.
View/Download from: Publisher's site
Nan, Y, Huang, X & Guo, YJ 1970, 'A Fast Piecewise Constant Doppler Algorithm for Generalized Continuous Wave Synthetic Aperture Radar', 2019 International Radar Conference (RADAR), 2019 International Radar Conference (RADAR), IEEE, Toulon, France.
View/Download from: Publisher's site
View description>>
The generalized continuous wave synthetic aperture radar (GCW-SAR) adopts one-dimensional data recording without the slow time dimension and hence offers many advantages compared with the conventional SAR system. In this paper, a fast piecewise constant Doppler algorithm is proposed based on further zero-th order approximation on top of the linear approximation of the slant range, leading to a flexible azimuth imaging spacing. Significant reduction of the complexity can be achieved by extending the azimuth imaging spacing and downsampling the received signal in digital domain. Simulation results validate the advantages of the proposed algorithm.
Nguyen, HV, Nguyen, V-D, Dobre, OA, Nguyen, DN, Dutkiewicz, E & Shin, O-S 1970, 'A Novel Spectral-Efficient Resource Allocation Approach for NOMA-Based Full-Duplex Systems', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Hawaii.
View/Download from: Publisher's site
View description>>
This paper investigates the coexistence of non- orthogonal multiple access (NOMA) and full-duplex (FD), where the NOMA successive interference cancellation technique is applied simultaneously to both uplink (UL) and downlink (DL) transmissions in the same time-frequency resource block. Specifically, we jointly optimize the user association (UA) and power control to maximize the overall sum rate, subject to user-specific quality-of-service and total transmit power constraints. To be spectrally-efficient, we introduce the tensor model to optimize the UL users' decoding order and the DL users' clustering, which results in a mixed-integer non- convex problem. For solving this problem, we first relax the binary variables to be continuous, and then propose a low-complexity design based on the combination of the inner convex approximation framework and the penalty method. Numerical results show that the proposed algorithm significantly outperforms the conventional FD-based schemes, FD-NOMA and its half-duplex counterpart with random UA.
Nguyen, MH, Hà, MH, Hoang, DT, Nguyen, DN, Dutkiewicz, E & Tran, TT 1970, '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 International Publishing, 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, N-T, Van Huynh, N, Hoang, DT, Nguyen, DN, Nguyen, N-H, Nguyen, Q-T & Dutkiewicz, E 1970, '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
View description>>
© 2019 IEEE. In this paper, we propose a novel approach to jointly address energy management and network throughput maximization problems for heterogeneous IoT low-power wireless communication networks. In particular, we consider a low-power communication network in which the IoT devices can harvest energy from a dedicated RF energy source to support their transmissions or backscatter the signals of the RF energy source to transmit information to the gateway. Different IoT devices may have dissimilar hardware configurations, and thus they may have various communications types and energy requirements. In addition, the RF energy source may have a limited energy supply source which needs to be minimized. Thus, to maximize the network throughput, we need to jointly optimize energy usage and operation time for the IoT devices under different energy demands and communication constraints. However, this optimization problem is non-convex due to the strong relation between energy supplied by the RF energy source and the IoT communication time, and thus obtaining the optimal solution is intractable. To address this problem, we study the relation between energy supply and communication time, and then transform the non-convex optimization problem to an equivalent convex-optimization problem which can achieve the optimal solution. Through simulation results, we show that our solution can achieve greater network throughputs (up to five times) than those of other conventional methods, e.g., TDMA. In addition, the simulation results also reveal some important information in controlling energy supply and managing low-power IoT devices in heterogeneous wireless communication networks.
Pham, TT, Takalkar, MA, Xu, M, Hoang, DT, Truong, HA, Dutkiewicz, E & Perry, S 1970, '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 International Publishing, 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.
Poostchi, H, Borzeshi, EZ & Piccardi, M 1970, 'BILSTM-CRF for Persian named-entity recognition armanpersonercorpus: The first entity-annotated Persian dataset', LREC 2018 - 11th International Conference on Language Resources and Evaluation, Language Resources and Evaluation Conference, European Language Resources Association (ELRA, Miyazaki, Japan, pp. 4427-4431.
View description>>
Named-entity recognition (NER) can still be regarded as work in progress for a number of Asian languages due to the scarcity of annotated corpora. For this reason, with this paper we publicly release an entity-annotated Persian dataset and we present a performing approach for Persian NER based on a deep learning architecture. In addition to the entity-annotated dataset, we release a number of word embeddings (including GloVe, skip-gram, CBOW and Hellinger PCA) trained on a sizable collation of Persian text. The combination of the deep learning architecture (a BiLSTM-CRF) and the pre-trained word embeddings has allowed us to achieve a 77.45% CoNLL F1 score, a result that is more than 12 percentage points higher than the best previous result and interesting in absolute terms.
Poostchimohammadabadi, H & Piccardi, M 1970, '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 usedto 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.
Qin, C, Zhang, JA, Huang, X & Guo, YJ 1970, 'Angle-of-Arrival Acquisition and Tracking via Virtual Subarrays in an Analog Array', 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 2019 IEEE 90th Vehicular Technology Conference (VTC2019-Fall), 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 1970, 'Mininet Topology: Mirror of the Optical Switch Fabric', 2019 29th International Telecommunication Networks and Applications Conference (ITNAC), 2019 29th International Telecommunication Networks and Applications Conference (ITNAC), IEEE, New Zealand, pp. 1-6.
View/Download from: Publisher's site
View description>>
Software Defined Networks (SDN) is a new approach to change the conventional networking and is being researched in the various networking domains. To test and prototype SDN based concepts, a lightweight and closer to reality option is Mininet emulator. Mininet emulates SDN behavior by creating a virtual network of elements using Network Namespaces on a single Linux kernel machine. In this work, we have developed a Mininet topology that emulates the structure of a WDM Switch. The topology mirrors the paths that can be used by the wavelengths in a WDM switch fabric. The SDN controller can find a path for communication between hosts through this network. We simulated our Mininet topology, which mirrors an architecture for three wavelengths. The Ping command results show that only a set of hosts can be reached out by a particular host; which is the requirement of a WDM switch, and this verifies that Mininet topology is mapping a WDM switch.
Rahman, ML, Cui, P-F, Zhang, JA, Huang, X, Guo, YJ & Lu, Z 1970, 'Joint Communication and Radar Sensing in 5G Mobile Network by Compressive Sensing', 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 2019 19th International Symposium on Communications and Information Technologies (ISCIT), IEEE, 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 1970, 'A Big Sensor Data Offloading Scheme in Rail Networks', 2019 IEEE 89th Vehicular Technology Conference (VTC2019-Spring), 2019 IEEE 89th Vehicular Technology Conference (VTC2019-Spring), 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 1970, '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 International Publishing, 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.
Saputra, YM, Hoang, DT, Nguyen, DN & Dutkiewicz, E 1970, 'JOCAR: A Jointly Optimal Caching and Routing Framework for Cooperative Edge Caching Networks', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Hawaii.
View/Download from: Publisher's site
View description>>
We propose a jointly optimal caching and routing framework (JOCAR) for a cooperative mobile edge caching network. This novel network architecture enables mobile edge servers/nodes (MENs) to collaborate in not only caching but also routing contents to users, in order to simultaneously minimize the total content-access delay for all mobile users and reduce the traffic on the backhaul network. To that end, we first formulate an access- delay minimization problem by jointly optimizing the content caching and routing decisions while accounting for various network configurations. Solving this problem requires us to deal with a nested dual optimization due to the strong mutual dependence between content caching and routing decisions. To tackle it, we first transform the nested dual problem to an equivalent mixed-integer nonlinear programming (MINLP) problem. Then, we design a branch-and-bound based algorithm with the interior-point method to find the near-optimal policy for the MINLP problem. Extensive simulations show that JOCAR can reduce the total average delay and increase the cache hit rate for the whole network by more than 40% and by four times, respectively, compared with other conventional policies.
Saputra, YM, Hoang, DT, Nguyen, DN, Dutkiewicz, E, Mueck, MD & Srikanteswara, S 1970, 'Energy Demand Prediction with Federated Learning for Electric Vehicle Networks', 2019 IEEE Global Communications Conference (GLOBECOM), IEEE Global Communications Conference, IEEE, Hawaii.
View/Download from: Publisher's site
View description>>
In this paper, we propose novel approaches using state-of-the-art machinelearning techniques, aiming at predicting energy demand for electric vehicle(EV) networks. These methods can learn and find the correlation of complexhidden features to improve the prediction accuracy. First, we propose an energydemand learning (EDL)-based prediction solution in which a charging stationprovider (CSP) gathers information from all charging stations (CSs) and thenperforms the EDL algorithm to predict the energy demand for the consideredarea. However, this approach requires frequent data sharing between the CSs andthe CSP, thereby driving communication overhead and privacy issues for the EVsand CSs. To address this problem, we propose a federated energy demand learning(FEDL) approach which allows the CSs sharing their information withoutrevealing real datasets. Specifically, the CSs only need to send their trainedmodels to the CSP for processing. In this case, we can significantly reduce thecommunication overhead and effectively protect data privacy for the EV users.To further improve the effectiveness of the FEDL, we then introduce a novelclustering-based EDL approach for EV networks by grouping the CSs into clustersbefore applying the EDL algorithms. Through experimental results, we show thatour proposed approaches can improve the accuracy of energy demand prediction upto 24.63% and decrease communication overhead by 83.4% compared with otherbaseline machine learning algorithms.
Seifollahi, S, Piccardi, M, Borzeshi, EZ & Kruger, B 1970, 'Taxonomy-Augmented Features for Document Clustering', Communications in Computer and Information Science, Australasian Conference on Data Mining, Springer Singapore, 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.
Shi, Z, Zhang, JA, Xu, R & Cheng, Q 1970, 'Deep Learning Networks for Human Activity Recognition with CSI Correlation Feature Extraction', ICC 2019 - 2019 IEEE International Conference on Communications (ICC), ICC 2019 - 2019 IEEE International Conference on Communications (ICC), 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.
Song, LZ, Qin, PY & Guo, YJ 1970, '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>>
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.
Suarez-Rodriguez, C, He, Y, Jayawickrama, BA & Dutkiewicz, E 1970, 'Low-Overhead Handover-Skipping Technique for 5G Networks.', WCNC, IEEE, pp. 1-6.
View/Download from: Publisher's site
View description>>
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 1970, 'Design of Ultra-Wideband On-Chip Millimter-Wave Bandpass Filter in 0.13-μm (Bi)-CMOS Technology', 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 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 1970, 'On-Chip Millimeter-Wave Bandpass Filter Design Using Multi-Layer Modified-Ground-Ring Structure', 2019 IEEE MTT-S International Microwave Symposium (IMS), 2019 IEEE/MTT-S International Microwave Symposium - IMS 2019, IEEE, 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.
Sun, H-H, Ding, C, Zhu, H, Jones, B & Guo, YJ 1970, 'Cross-Band Scattering Suppression for MultiBand Base Station Antenna Arrays', 2019 8th Asia-Pacific Conference on Antennas and Propagation (APCAP), 2019 8th Asia-Pacific Conference on Antennas and Propagation (APCAP), IEEE, pp. 579-580.
View/Download from: Publisher's site
View description>>
This paper presents a dual-band dual-polarized interleaved base station antenna array unit based on a filtering antenna for cross-band de-scattering. The array is configured as two columns of antenna arrays operating at a higher band from 1.71 GHz to 2.30 GHz interleaved with one column of antenna array operating at a lower band from 0.80 GHz to 0.96 GHz. By inserting low-pass high-stop filters into the low-band dipole arms, a filtering antenna that can efficiently suppress the radiation at the higher band is achieved. On one hand, the obtained filtering antenna has a slightly reduced gain and narrower bandwidth, which is attributed to the filters. On the other hand, the obtained filtering antenna working at the low band has minimum negative effect on the high band antenna performance.
Takalkar, MA, Zhang, H & Xu, M 1970, 'Improving Micro-expression Recognition Accuracy Using Twofold Feature Extraction', MultiMedia Modeling (LNCS), International Conference on Multimedia Modeling, Springer International Publishing, 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.
Tao, Q, Luo, X, Wang, H & Xu, R 1970, 'Enhancing Relation Extraction Using Syntactic Indicators and Sentential Contexts', 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI), 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI), IEEE, USA, pp. 1574-1580.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. State-of-the-art methods for relation extraction consider the sentential context by modeling the entire sentence. However, syntactic indicators, certain phrases or words like prepositions that are more informative than other words and may be beneficial for identifying semantic relations. Other approaches using fixed text triggers capture such information but ignore the lexical diversity. To leverage both syntactic indicators and sentential contexts, we propose an indicator-aware approach for relation extraction. Firstly, we extract syntactic indicators under the guidance of syntactic knowledge. Then we construct a neural network to incorporate both syntactic indicators and the entire sentences into better relation representations. By this way, the proposed model alleviates the impact of noisy information from entire sentences and breaks the limit of text triggers. Experiments on the SemEval-2010 Task 8 benchmark dataset show that our model significantly outperforms the state-of-the-art methods.
Tello, AMD & Abolhasan, M 1970, 'SDN Controllers Scalability and Performance Study', 2019 13th International Conference on Signal Processing and Communication Systems (ICSPCS), 2019 13th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, Malaysia.
View/Download from: Publisher's site
View description>>
Software Defined Networks (SDN) is a networking approach that decouples the intelligent control plane from networking devices and establishes a separate entity called ”controller” that rule the behaviour of the data plane on physical networking devices. Due to the rapid evolution and growth of SDN controllers in the market, this paper aims to present an extensive study on performance and scalability of different open source SDN controllers available in the existing literature. This work covers previous studies and expands them with updated information and official benchmarking methodologies. The study provides a framework based on the standards recommended by the IETF (Internet Engineering Task Force) and it will serve as a guideline to the SDN community to benchmark different SDN controllers.
Trede, F, Braun, R & Brookes, W 1970, 'Studio-based learning in a first year engineering curriculum: Exploring students' learning experiences and reflections using the rich picture method.', ITHET, International Conference on Information Technology Based Higher Education and Training, IEEE, Magdeburg, Germany, pp. 1-5.
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 1970, 'ISCIT 2019 Message', 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 2019 19th International Symposium on Communications and Information Technologies (ISCIT), IEEE, p. XXI.
View/Download from: Publisher's site
Tuyen Le, A, Tran, LC, Huang, X & Guo, J 1970, 'Analog Least Mean Square Loops for Self-Interference Cancellation in In-Band Full-Duplex Systems', 2019 13th International Conference on Signal Processing and Communication Systems (ICSPCS), 2019 13th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, pp. 1-1.
View/Download from: Publisher's site
Unanue, IJ, Arratibel, LG, Borzeshi, EZ & Piccardi, M 1970, 'English-Basque statistical and neural machine translation', LREC 2018 - 11th International Conference on Language Resources and Evaluation, Language Resources and Evaluation Conference, European Language Resource Association, Miyazaki, Japan, pp. 880-885.
View description>>
Neural Machine Translation (NMT) has attracted increasing attention in the recent years. However, it tends to require very large training corpora which could prove problematic for languages with low resources. For this reason, Statistical Machine Translation (SMT) continues to be a popular approach for low-resource language pairs. In this work, we address English-Basque translation and compare the performance of three contemporary statistical and neural machine translation systems: OpenNMT, Moses SMT and Google Translate. For evaluation, we employ an open-domain and an IT-domain corpora from the WMT16 resources for machine translation. In addition, we release a small dataset (Berriak) of 500 highly-accurate English-Basque translations of complex sentences useful for a thorough testing of the translation systems.
Unanue, IJ, Borzeshi, EZ, Esmaili, N & Piccardi, M 1970, '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, pp. 430-436.
View description>>
Regularization of neural machine translation is still a significant problem,especially in low-resource settings. To mollify this problem, we proposeregressing word embeddings (ReWE) as a new regularization technique in a systemthat is jointly trained to predict the next word in the translation(categorical value) and its word embedding (continuous value). Such a jointtraining allows the proposed system to learn the distributional propertiesrepresented by the word embeddings, empirically improving the generalization tounseen sentences. Experiments over three translation datasets have showed aconsistent improvement over a strong baseline, ranging between 0.91 and 2.54BLEU points, and also a marked improvement over a state-of-the-art system.
Van Huynh, N, Hoang, DT, Nguyen, DN & Dutkiewicz, E 1970, '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
View description>>
© 2019 IEEE. Practical and efficient network slicing often faces real-time dynamics of network resources and uncertain customer demands. This work provides an optimal and fast resource slicing solution under such dynamics by leveraging the latest advances in deep learning. Specifically, we first introduce a novel system model which allows the network provider to effectively allocate its combinatorial resources, i.e., spectrum, computing, and storage, to various classes of users. To allocate resources to users while taking into account the dynamic demands of users and resources constraints of the network provider, we employ a semi-Markov decision process framework. To obtain the optimal resource allocation policy for the network provider without requiring environment parameters, e.g., uncertain service time and resource demands, a Q-learning algorithm is adopted. Although this algorithm can maximize the revenue of the network provider, its convergence to the optimal policy is particularly slow, especially for problems with large state/action spaces. To overcome this challenge, we propose a novel approach using an advanced deep Q-learning technique, called deep dueling that can achieve the optimal policy at few thousand times faster than that of the conventional Q-learning algorithm. Simulation results show that our proposed framework can improve the long-term average return of the network provider up to 40% compared with other current approaches.
Vu, L, Cao, VL, Nguyen, QU, Nguyen, DN, Hoang, DT & Dutkiewicz, E 1970, '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
View description>>
© 2019 IEEE. We develop a novel deep learning model, Multi-distributed Variational AutoEncoder (MVAE), for the network intrusion detection. To make the traffic more distinguishable, MVAE introduces the label information of data samples into the Kullback-Leibler (KL) term of the loss function of Variational AutoEncoder (VAE). This label information allows MVAEs to force/partition network data samples into different classes with different regions in the latent feature space. As a result, the network traffic samples are more distinguishable in the new representation space (i.e., the latent feature space of MVAE), thereby improving the accuracy in detecting intrusions. To evaluate the efficiency of the proposed solution, we carry out intensive experiments on two popular network intrusion datasets, i.e., NSL-KDD and UNSW-NB15 under four conventional classifiers including Gaussian Naive Bayes (GNB), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The experimental results demonstrate that our proposed approach can significantly improve the accuracy of intrusion detection algorithms up to 24.6% compared to the original one (using area under the curve metric).
Vu, TT, Nguyen, DN, Hoang, DT & Dutkiewicz, E 1970, 'QoS-Aware Fog Computing Resource Allocation Using Feasibility-Finding Benders Decomposition', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Hawaii.
View/Download from: Publisher's site
View description>>
We investigate a joint offloading and resource allocation under a multi-layer cooperative fog and cloud computing architecture, aiming to minimize the total energy consumption of mobile devices while meeting users' QoS requirements, e.g., delay, security, and application compatibility. Due to the mutual coupling amongst offloading decision and resource allocation variables, the resulting optimization is a mixed integer non- linear programming problem that is NP-hard. Such problem often requires exponential time to find the optimal solution. In this work, we propose a distributed approach, namely feasibility-finding Benders decomposition (FFBD), that decomposes the original problem into a master problem for the offloading decision and subproblems for resource allocation. These (simpler) subproblems can be solved in parallel at fog nodes, thereby reducing both the complexity and the computational time. The numerical results show that the FFBD always returns the optimal solution of the problem with significantly less computation time (e.g., in comparing with the branch-and-bound method).
Wang, Q, Jia, W, He, X, Lu, Y, Blumenstein, M, Huang, Y & Lyu, S 1970, 'DeepText: Detecting Text from the Wild with Multi-ASPP-Assembled DeepLab', 2019 International Conference on Document Analysis and Recognition (ICDAR), 2019 International Conference on Document Analysis and Recognition (ICDAR), IEEE, Sydney, Australia, pp. 208-213.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this paper, we address the issue of scene text detection in the way of direct regression and successfully adapt an effective semantic segmentation model, DeepLab v3+ [1], for this application. In order to handle texts with arbitrary orientations and sizes and improve the recall of small texts, we propose to extract features of multiple scales by inserting multiple Atrous Spatial Pyramid Pooling (ASPP) layers to the DeepLab after the feature maps with different resolutions. Then, we set multiple auxiliary IoU losses at the decoding stage and make auxiliary connections from the intermediate encoding layers to the decoder to assist network training and enhance the discrimination ability of lower encoding layers. Experiments conducted on the benchmark scene text dataset ICDAR2015 demonstrate the superior performance of our proposed network, named as DeepText, over the state-of-the-art approaches.
Wang, Q, Jia, W, He, X, Lu, Y, Blumenstein, M, Huang, Y & Lyu, S 1970, '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 1970, 'A High-Performance Hybrid Blockchain System for Traceable IoT Applications', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Network and System Security, Springer International Publishing, 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.
Wilson, KJ, Alabd, R, Abolhasan, M, Franklin, DR & Safavi-Naeini, M 1970, 'Localisation of the Lines of Response in a Continuous Cylindrical Shell PET Scanner.', EMBC, IEEE Engineering in Medicine and Biology Conference, IEEE, Berlin, Germany, pp. 4844-4850.
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.
Wu, J, Yao, L, Huang, Y, Xu, J, Wu, Q & Huang, L 1970, 'Improving Person Re-Identification Performance Using Body Mask Via Cross-Learning Strategy', 2019 IEEE Visual Communications and Image Processing (VCIP), 2019 IEEE Visual Communications and Image Processing (VCIP), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. The task of person re-identification (re-id) is to find the same pedestrian across non-overlapping cameras. Normally, the performance of person re-id can be affected by background clutters. However, existing segmentation algorithms are hard to obtain perfect foreground person images. To effectively leverage the body (foreground) cue, and in the meantime pay attention to discriminative information in the background (e.g., companion or vehicle), we propose to use a cross-learning strategy to take both foreground and other discriminative information into account. In addition, since currently existing foreground segmentation result always involves noise, we use Label Smoothing Regularization (LSR) to strengthen the generalization capability during our learning process. In experiments, we pick up two state-of-The-Art person re-id methods to verify the effectiveness of our proposed cross-learning strategy. Our experiments are carried out on two publicly available person re-id datasets. Obvious performance improvements can be observed on both datasets.
Xie, H-B, Li, C, Xu, RYD & Mengersen, K 1970, '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 International Publishing, 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.
Xie, Y, Xu, Z, Gong, S, Xu, J, Hoang, DT & Niyato, D 1970, 'Backscatter-Assisted Hybrid Relaying Strategy for Wireless Powered IoT Communications', 2019 IEEE Global Communications Conference (GLOBECOM), GLOBECOM 2019 - 2019 IEEE Global Communications Conference, IEEE, Waikoloa, HI, USA, pp. 1-6.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this work, we consider multiple energy harvesting relays to assist information transmission from a hybrid access point (HAP) to a distant receiver. The multi-antenna HAP also beamforms RF power to the relays by using a power-splitting protocol. We aim to maximize the throughput by jointly optimizing the HAP's beamforming strategy as well as individual relays' energy harvesting and collaborative beamforming strategies. With dense user devices, the throughput maximization takes account of the direct links from the HAP to the receiver as they are short and contribute considerably to the overall throughput. Moreover, we introduce the concept of hybrid relaying communications which allows the energy harvesting relays to switch between two radio modes. In particular, the relays can operate either in RF communications or backscatter communications, depending on their channel conditions and energy status. This results in a non-convex and combinatorial throughput maximization problem. With the fixed relay mode, we can find a feasible lower performance bound via convex approximation, which further motivates our algorithm design to update the relay mode in an iterative manner. Simulation results verify that the proposed hybrid relaying strategy can achieve significant performance improvement compared to the conventional relaying strategy with all relays operating in the RF communications mode.
Xu, Y, Xu, D, Hong, X, Ouyang, W, Ji, R, Xu, M & Zhao, G 1970, 'Structured Modeling of Joint Deep Feature and Prediction Refinement for Salient Object Detection', 2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, Seoul, Korea.
View/Download from: Publisher's site
View description>>
Recent saliency models extensively explore to incorporate multi-scale contextual information from Convolutional Neural Networks (CNNs). Besides direct fusion strategies, many approaches introduce message-passing to enhance CNN features or predictions. However, the messages are mainly transmitted in two ways, by feature-to-feature passing, and by prediction-to-prediction passing. In this paper, we add message-passing between features and predictions and propose a deep unified CRF saliency model . We design a novel cascade CRFs architecture with CNN to jointly refine deep features and predictions at each scale and progressively compute a final refined saliency map. We formulate the CRF graphical model that involves message-passing of feature-feature, feature-prediction, and prediction-prediction, from the coarse scale to the finer scale, to update the features and the corresponding predictions. Also, we formulate the mean-field updates for joint end-to-end model training with CNN through back propagation. The proposed deep unified CRF saliency model is evaluated over six datasets and shows highly competitive performance among the state of the arts.
Yang, T, Ding, C, Ziolkowski, RW & Guo, YJ 1970, '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) and Australian Conference on Optics, Lasers, and Spectroscopy (ACOLS) 2019, SPIE, Melbourne, AUSTRALIA.
View/Download from: Publisher's site
View description>>
© COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only. To overcome the crosstalk happening between two degenerately fundamental modes of a fiber in Terahertz (THz) regime, a novel photonic crystal fiber (PCF) that yields a wide range of single-polarization-single-mode (SPSM) propagation with large loss differences (LDs) is designed. The method used to realize this SPSM PCF is to deposit an epsilon-near-zero (ENZ) material in four selected air holes in the cladding, which ends up with four ENZ rings. These ENZ rings introduce significant LDs between the wanted (X-polarized) and unwanted (Y-polarized and high order) modes. Extensive simulation results demonstrate that the LDs between the wanted and unwanted modes vary with the thickness of ENZ rings. With a very short length (4 cm) of the proposed PCF, pure SPSM propagation, i.e., the unwanted modes are 20 dB lower than the wanted mode, can be achieved from 1 to 1.2 THz.
Yang, Y & Zhu, X 1970, 'Overview of Millimeter-Wave On-Chip and Off-Chip Bandpass Filter Designs', 2019 International Conference on Microwave and Millimeter Wave Technology (ICMMT), 2019 International Conference on Microwave and Millimeter Wave Technology (ICMMT), IEEE, Guangzhou, China, pp. 1-2.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. In this paper, a brief review of recent advancement in on-chip and off-chip millimeter-wave bandpass filter designs are presented. First, the general background about the two solutions are provided, with a general discussion about their pros and cons. Second, the representative state-of-the-art works are presented. This short paper provides the researchers, in the relative fields, with some guidelines when deciding bandpass filters for a system-on-chip (SoC) solution.
Yang, Y, Hou, Z, Zhu, X, Che, W & Xue, Q 1970, 'A Millimeter-Wave Reconfigurable On-Chip Coupler with Tunable Power-Dividing Ratios', 2019 IEEE MTT-S International Wireless Symposium (IWS), 2019 IEEE MTT-S International Wireless Symposium (IWS), IEEE.
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, Zhu, X, Che, W & Xue, Q 1970, 'A Millimeter-Wave On-Chip Bandpass Filter with All-Pole Characteristics', 2019 IEEE MTT-S International Wireless Symposium (IWS), 2019 IEEE MTT-S International Wireless Symposium (IWS), IEEE.
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.
Ye, P, Wang, Y, Xia, Y, An, P & Zhang, J 1970, 'Enhanced Saliency Prediction via Free Energy Principle', Digital TV and Multimedia Communication, International Forum on Digital TV and Wireless Multimedia Communications, Springer Singapore, 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.
Zhang, H, Huang, X & Zhang, JA 1970, '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 (ICCC), 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, Guo, YJ, Song, R, Wang, C, Wu, W, Xu, X & Lu, Z 1970, 'A High-Speed Low-Cost Millimeter Wave System with Dual Pulse Shaping Transmission and Symbol Rate Equalization Techniques', 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 2019 IEEE International Symposium on Circuits and Systems (ISCAS), IEEE, 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 1970, 'A 30 Gbps Low-Complexity and Real-Time Digital Modem for Wireless Communications at 0.325 THz', 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 2019 19th International Symposium on Communications and Information Technologies (ISCIT), IEEE, 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 1970, '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, P, Wu, Q & Xu, J 1970, 'VN-GAN: Identity-preserved Variation Normalizing GAN for Gait Recognition', 2019 International Joint Conference on Neural Networks (IJCNN), 2019 International Joint Conference on Neural Networks (IJCNN), 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 1970, 'VT-GAN: View Transformation GAN for Gait Recognition Across Views', 2019 International Joint Conference on Neural Networks (IJCNN), 2019 International Joint Conference on Neural Networks (IJCNN), 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, X, Liu, J, Li, Y, Cui, Q, Tao, X & Liu, RP 1970, 'Blockchain Based Secure Package Delivery via Ridesharing', 2019 11th International Conference on Wireless Communications and Signal Processing (WCSP), 2019 11th International Conference on Wireless Communications and Signal Processing (WCSP), IEEE.
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, Z, Wang, Y, Wu, Q & Chen, F 1970, 'Visual Relationship Attention for Image Captioning', 2019 International Joint Conference on Neural Networks (IJCNN), 2019 International Joint Conference on Neural Networks (IJCNN), IEEE, Budapest, HUNGARY.
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.
Zhao, M, Zhang, J, Zhang, C & Zhang, W 1970, '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, Long Beach, CA, pp. 12728-12737.
View/Download from: Publisher's site
Zhao, M, Zhang, J, Zhang, C & Zhang, W 1970, 'Towards Locally Consistent Object Counting with Constrained Multi-stage Convolutional Neural Networks', COMPUTER VISION - ACCV 2018, PT VI, Asian Conference on Computer Vision, Springer International Publishing, 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.
Zhou, I, Makhdoom, I, Abolhasan, M, Lipman, J & Shariati, N 1970, 'A Blockchain-based File-sharing System for Academic Paper Review', 2019 13th International Conference on Signal Processing and Communication Systems (ICSPCS), 2019 13th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, Australia.
View/Download from: Publisher's site
Zhou, K, Luo, X, Wang, H & Xu, R 1970, 'Multi-task Learning for Relation Extraction', 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI), 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI), IEEE, USA, pp. 1480-1487.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. Distantly supervised relation extraction leverages knowledge bases to label training data automatically. However, distant supervision may introduce incorrect labels, which harm the performance. Many efforts have been devoted to tackling this problem, but most of them treat relation extraction as a simple classification task. As a result, they ignore useful information that comes from related tasks, i.e., dependency parsing and entity type classification. In this paper, we first propose a novel Multi-Task learning framework for Relation Extraction (MTRE). We employ dependency parsing and entity type classification as auxiliary tasks and relation extraction as the target task. We learn these tasks simultaneously from training instances to take advantage of inductive transfer between auxiliary tasks and the target task. Then we construct a hierarchical neural network, which incorporates dependency and entity representations from auxiliary tasks into a more robust relation representation against the noisy labels. The experimental results demonstrate that our model improves the predictive performance substantially over single-task learning baselines.
Zhu, H & Guo, YJ 1970, 'Design of Out-of-phase Filtering Power Dividers Using Flexible Coupling Schemes', 2019 IEEE Asia-Pacific Microwave Conference (APMC), 2019 IEEE Asia-Pacific Microwave Conference (APMC), IEEE, Singapore, Singapore, pp. 1023-1025.
View/Download from: Publisher's site
View description>>
© 2019 IEEE. This paper presents based on flexible coupling schemes using multiple quarter-wavelength resonators. Microstrip-to-slotline transitions were are used to provide external coupling to the filters. A dual-mode and tri-mode filter are designed. For the dual-mode design, a pair of quarter-wavelength resonators is used with a gap between them providing a suitable coupling coefficient. For the tri-band design, an additional resonator is coupled to the previous quarter-wavelength resonators, leading to a three-pole response within the passband. A microstrip line with a grounded resistor is added at the middle point of two output ports, leading to excellent output matching and isolation between output ports. Two design examples with dual-mode and tri-mode responses are realized with the verification using full-wave simulations.
Zhu, H, Lin, J-Y & Guo, YJ 1970, 'Wideband Filtering Out-of-Phase Power Dividers Using Slotline Resonators and Microstrip-to-Slotline Transitions', 2019 IEEE MTT-S International Microwave Symposium (IMS), 2019 IEEE/MTT-S International Microwave Symposium - IMS 2019, 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 1970, '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>>
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 1970, 'Design of Miniaturized On-Chip Bandpass Filters using Inverting-Coupled Structure for Millimter-Wave Applications', 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 2019 IEEE International Symposium on Circuits and Systems (ISCAS), 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.