Publications

Hoang, DT, Niyato, D, Kim, DI, Huynh, NV & Gong, S 2020, Ambient Backscatter Communication Networks, Cambridge University Press.
View description>>

This is an essential tool for graduate students, researchers, engineers, developers, and entrepreneurs.

Hoang, DT, Niyato, D, Kim, DI, Huynh, NV & Gong, S 2020, Ambient Backscatter Communication Networks, Cambridge University Press.
View description>>

Understand fundamental principles of ambient backscatter technology and their diverse potential applications with this authoritative review.

Ajayan, AR, Al-Doghman, F & Chaczko, Z 2020, 'Tensor Decompositions in Multimodal Big Data: Studying Multiway Behavioral Patterns' in Klempous, R & Nikodem, J (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Cham, Switzerland, pp. 104-118.
View/Download from: Publisher's site
View description>>

Preset day cyber-physical systems (CPS) are the confluence of very large data sets, tight time constraints, and heterogeneous hardware units, ridden with latency and volume constraints, demanding newer analytic perspectives. Their system logistics can be well-defined by the data-streams’ behavioural trends across various modalities, without numerical restrictions, favouring resource-saving over methods of investigating individual component features and operations. The aim of this paper is to demonstrate how behaviour patterns and related anomalies comprehensively define a CPS. Tensor decompositions are hypothesized as the solution in the context of multimodal smart-grid-originated Big Data analysis. Tensorial data representation is demonstrated to capture the complex knowledge encompassed in these data flows. The uniqueness of this approach is highlighted in the modified multiway anomaly patterns models. In addition, higher-order data preparation schemes, design and implementation of tensorial frameworks and experimental-analysis are final outcomes.

Al-Doghman, F, Chaczko, Z & Ajayan, AR 2020, 'Policy-Based Consensus Data Aggregation for the Internet of Things' in Klempous, R & Nikodem, J (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Cham, Switzerland, pp. 119-130.
View/Download from: Publisher's site
View description>>

The new trend of the Internet of Things brings a whole breed of opportunities and applications. Within it, a massive amount of data coming from heterogeneous sources travel in a bidirectional way. Data aggregation is one of the most efficient ways to mitigate Big Data. However, using one type of aggregation within a network at all times is not an optimal option. Various network situations require different aggregation functions at different times. We introduce a policy-based data aggregation framework that can handle this issue by referring to a policy when executing the aggregation strategy. An agreement process is used to reach consensus about the aggregation function that is to be applied on the network (or part of it) at a specific time. Participants are to negotiate the policy terms based on the current network status and the nature of the coming requests. The framework represents a promising scope for fully automated IoT.

Barua, S & Braun, R 2020, 'A Markovian Approach to the Mobility Management for the D2D Communications in 5G Cellular Network System' in Klempous, R & Nikodem, J (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Singapore, pp. 253-265.
View/Download from: Publisher's site
View description>>

Device to Device communication, which is mostly known as D2D communication is considered an integral part of the next generation (5G) telecommunication network system. However, it is still a challenge how to manage the devices while they are in moving condition. Only a few papers highlighted this problem and yet to come up with a reliable solution. We already proposed a unique algorithm of mobility management for the D2D communication system in our previous papers. In this chapter, we will highlight and explain the mobility model mathematically and analytically along with the simulation of the Markovian model. Markov model, which is essentially a simplified approach of describing a system that occupies a discrete state at any point in time. We also make a bridge between our mobility algorithm and Markovian model.

Chaczko, Z, Chiu, C & Yu, Z 2020, 'Assessment of a Multi-agent RFID Platform for Industrial Logistic Operations' in Klempous, R & Nikodem, J (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Cham, Switzerland, pp. 166-182.
View/Download from: Publisher's site
View description>>

The paper examines the experimental implementation of multi-agent, omni-dimensional RFID technology for logistical and industrial operations. The aim of the research is to demonstrate the processes taken to test RFID under real-world industrial scenarios using a multi-agent analytical approach. The results obtained demonstrate the level of accuracy for an omni-directional versus agent-based optimized analytical approach, that is essential in conveying the performance in industrial logistic setups. Further work into the positioning and localization of multidimensional RFID inventory scanning for commercial setups is considered for future investigation.

Chaczko, Z, Sethi, H & Alsawwaf, M 2020, 'An Overview of Multi-layer Security System Using Biometrics' in Klempous, R & Nikodem, J (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Singapore, pp. 219-231.
View/Download from: Publisher's site
View description>>

After years of research on biometrics, still the authentication and verification of a human being is at risk. Many challenges have been faced by the system to authenticate a human on the basis of their biometric trait. The design of this paper is such that it focuses on three main biometric techniques to extract the various features of a human body contour for authentication and verification purposes. This paper presents an approach towards voice recognition patterns, face geometry patterns and gait pattern analysis to increase the accuracy and precision of the system.

Guangsheng Yu, Xu Wang, Kan Yu, Wei Ni, Zhang, JA & Ren Ping Liu 2020, 'Scaling-out blockchains with sharding: an extensive survey' in Blockchains for Network Security: Principles, technologies and applications, Institution of Engineering and Technology, UK, pp. 225-270.
View/Download from: Publisher's site
View description>>

The blockchain technology, featured with its decentralized tamper -resistance based on a peer -to -peer (P2P) network, has been widely applied in financial applications and even further been extended to industrial applications. However, the weak scalability of traditional blockchain technology severely affects the wide adoption due to the well-known trilemma of decentralization-security-scalability in block chains. In regards to this issue, a number of solutions have been proposed, targeting to boost the scalability while preserving the decentralization and security. They range from modifying the on -chain data structure and consensus algorithms to adding the off -chain technologies. Therein, one of the most practical methods to achieve horizontal scalability along with the increasing network size is sharding, by partitioning network into multiple shards so that the overhead of duplicating communication, storage, and computation in each full node can be avoided. This chapter presents a survey focusing on sharding in Blockchains in a systematic and comprehensive way. We provide detailed comparison and evaluation of major sharding mechanisms, along with our insights analyzing the features and restrictions of the existing solutions. The remaining challenges and future research directions are also reviewed.

Saputra, Y, Dinh, H, Nguyen, D, Dutkiewicz, E & Niyato, D 2020, 'Wireless Edge Caching for Mobile Social Networks' in Vu, T, Chatzinotas, S, Baştuğ, E & Quek, TQS (eds), Wireless Edge Caching, Cambridge University Press, UK, pp. 146-172.
View/Download from: Publisher's site
View description>>

Wireless edge caching for mobile social networks (MSNs) has emerged as one of the prospective solutions to provide reliable and low-latency communication services for mobile users on social networking. In this chapter, we first give an overview of MSNs, including their development and challenges. We then discuss mobile edge caching (MEC) paradigms to address emerging issues for MSNs, e.g., service delay, users’ experience, and economic efficiency. In addition to the advantages, the development of MEC networks also places some key challenges’ such as hierarchical architecture of MEC networks, proactive caching, privacy, and security issues. The framework can authenticate MSN users based on public-key cryptography and predict their content demands utilizing a matrix factorization method. Based on the prediction, an optimal content caching policy for an MEC node is presented to minimize the average latency of all MSN users under the MEC nodes’ storage capacity constraints. Furthermore, this framework provides an optimal business model to maximize the revenue for MSN service providers based on the demands of the MSN users and the obtained optimal caching policy.

Shi, J, Braun, R, Yu, Z, Chu, L, Shi, Q & Xu, Z 2020, 'Tracking of Multiple Moving Objects Using Two Dimensional Beamforming Based on Low Cost Crossed Sensor Array' in Rudas, IJ & Szakál, A (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Singapore, pp. 232-242.
View/Download from: Publisher's site
View description>>

Moving objects tracking is very important in intelligent systems. Two dimensional Beamforming technologies have been applied to create images, with additional information such as ranges and radial velocities. Most radars use two dimensional full array to sense the target. To achieve high resolution, with the increasing sensor number, the sampling and computation cost are large. In order to reduce the overall cost, some methods based on sparse arrays have been proposed to create radar images. The sensor number can be reduced without losing much information. In this paper, a two dimensional beamforming method based on low cost crossed array is proposed. One linear phased array transmitting beamforming is used to steer the angle of beams in one dimension while the other orthogonal linear phased array receiving beamforming is used to detect the angle of arrival in the other direction. The sampled one dimensional beamforming image combines with time of steering direction to create a two dimensional spectrum. The proposed algorithm is simulated and the results of the system is analyzed

Xu Wang, Xuan Zha, Guangsheng Yu, Wei Ni & Ren Ping Liu 2020, 'Blockchain for Internet of Things' in Blockchains for Network Security: Principles, technologies and applications, Institution of Engineering and Technology, UK, pp. 87-136.
View/Download from: Publisher's site
View description>>

Internet of Things (IoT) technology is digitizing the physical world by connecting enormous and heterogeneous devices and unleashing great economic benefit. However, data privacy, security and trust issues in current solutions are seriously limiting the adoption of IoT applications. Blockchain, a decentralized and tamperresistant ledger, maintains consistent and immutable blocks of data at different servers and has the potential to tackle the security concerns in IoT applications. Inherent features in IoT, such as the massive IoT devices, heterogeneous IoT networks, limited battery, low computing power and communication bandwidth, make it hard to directly adopt blockchain technology in IoT application. This chapter presents a comprehensive survey on existing blockchain and IoT technologies and emphasizes on the challenges and limitation. Current studies, projects and designs on Blockchain-IoT systems are introduced and compared to illustrate the feasibility of the integration of blockchain and IoT. Blockchain technologies that can potentially address the critical challenges in IoT applications and suit the features of the same are identified with potential adaptations and enhancements elaborated on blockchain data structures, key blockchain technologies and consensus protocols. Future research directions of blockchain are collated for effective adoption in IoT applications.

Zhang, W, Chiu, C, Szymanski, J, Chaczko, Z, Su, S & Zhou, J 2020, 'Internet of Things Is Changing Teaching of Technical Subjects at UTS' in Klempous, R & Nikodem, J (eds), Smart Innovations in Engineering and Technology, Springer International Publishing, Cham, Switzerland, pp. 131-143.
View/Download from: Publisher's site
View description>>

The Internet of Things, involving the connection of intelligent devices (other than standard, already connected products such as computers and smartphones) to the Internet, is in the process of transforming numerous areas of lives. For many years, both academia and industry prepared the background theories and implementations for what is todays IoT, keeping changing and evolving. Students of universities, preparing to be the driving forces of innovation and progress have a very strong interest in IoT and related subjects. Teaching at universities, especially technical ones is strongly affected by the need to give students required background and show the direction for self study. In this paper we will provide a literature review to analyze how IoT inspired changes in teaching some technical subjects at University of Technology Sydney (UTS) by describing, the recent undertaken projects in this area, providing some case studies and drawing conclusions.

Abeywickrama, HV, He, Y, Dutkiewicz, E, Jayawickrama, BA & Mueck, M 2020, 'A Reinforcement Learning Approach for Fair User Coverage Using UAV Mounted Base Stations Under Energy Constraints', IEEE Open Journal of Vehicular Technology, vol. 1, pp. 67-81.
View/Download from: Publisher's site

Afroz, F & Braun, R 2020, 'Energy-efficient MAC protocols for wireless sensor networks: a survey', International Journal of Sensor Networks, vol. 32, no. 3, pp. 150-150.
View/Download from: Publisher's site

Alghamdi, K & Braun, R 2020, 'Software Defined Network (SDN) and OpenFlow Protocol in 5G Network', Communications and Network, vol. 12, no. 01, pp. 28-40.
View/Download from: Publisher's site
View description>>

The world is moving at a high speed in the implementation and innovations of new systems and gadgets. 3G and 4G networks support currently wireless network communications. However, the networks are deemed to be slow and fail to receive signals or data transmission to various regions as a result of solving the problem. This paper will analyze the use of Software Defined Network (SDN) in a 5G (fifth generation) network that can be faster and reliable. Further, in Mobile IP, there exist triangulation problems between the sending and receiving nodes along with latency issues during handoff for the mobile nodes causing huge burden in the network. With Cloud Computing and ecosystem for Virtualization developed for the Core and Radio Networks SDN OpenFlow seems to be a seamless solution for determining signal flow between mobiles. There have been a lot of researches going on for deploying SDN OpenFlow with the 5G Cellular Network. The current paper performs benchmarks as a feasibility need for implementing SDN OpenFlow for 5G Cellular Network. The Handoff mechanism impacts the scalability required for a cellular network and simulation results can be further used to be deployed the 5G Network.

Alsheikh, MA, Hoang, DT, Niyato, D, Leong, D, Wang, P & Han, Z 2020, 'Optimal Pricing of Internet of Things: A Machine Learning Approach', IEEE Journal on Selected Areas in Communications, vol. 38, no. 4, pp. 669-684.
View/Download from: Publisher's site
View description>>

© 1983-2012 IEEE. Internet of things (IoT) produces massive data from devices embedded with sensors. The IoT data allows creating profitable services using machine learning. However, previous research does not address the problem of optimal pricing and bundling of machine learning-based IoT services. In this paper, we define the data value and service quality from a machine learning perspective. We present an IoT market model which consists of data vendors selling data to service providers, and service providers offering IoT services to customers. Then, we introduce optimal pricing schemes for the standalone and bundled selling of IoT services. In standalone service sales, the service provider optimizes the size of bought data and service subscription fee to maximize its profit. For service bundles, the subscription fee and data sizes of the grouped IoT services are optimized to maximize the total profit of cooperative service providers. We show that bundling IoT services maximizes the profit of service providers compared to the standalone selling. For profit sharing of bundled services, we apply the concepts of core and Shapley solutions from cooperative game theory as efficient and fair allocations of payoffs among the cooperative service providers in the bundling coalition.

Alzoubi, YI & Gill, AQ 2020, 'An Empirical Investigation of Geographically Distributed Agile Development: The Agile Enterprise Architecture is a Communication Enabler.', IEEE Access, vol. 8, pp. 80269-80289.
View/Download from: Publisher's site

Amiri, M, Tofigh, F, Shariati, N, Lipman, J & Abolhasan, M 2020, 'Wide-angle metamaterial absorber with highly insensitive absorption for TE and TM modes', Scientific Reports, vol. 10, no. 1.
View/Download from: Publisher's site
View description>>

AbstractBeing incident and polarization angle insensitive are crucial characteristics of metamaterial perfect absorbers due to the variety of incident signals. In the case of incident angles insensitivity, facing transverse electric (TE) and transverse magnetic (TM) waves affect the absorption ratio significantly. In this scientific report, a crescent shape resonator has been introduced that provides over 99% absorption ratio for all polarization angles, as well as 70% and 93% efficiencies for different incident angles up to $$\theta =80^{\circ }$$θ=80∘ for TE and TM polarized waves, respectively. Moreover, the insensitivity for TE and TM modes can be adjusted due to the semi-symmetric structure. By adjusting the structure parameters, the absorption ratio for TE and TM waves at $$\theta =80^{\circ }$$θ=80∘ has been increased to 83% and 97%, respectively. This structure has been designed to operate at 5 GHz spectrum to absorb undesired signals generated due to the growing adoption of Wi-Fi networks. Finally, the proposed absorber has been fabricated in a $$20 \times 20$$20×20 arr...

Ben, X, Gong, C, Zhang, P, Yan, R, Wu, Q & Meng, W 2020, 'Coupled Bilinear Discriminant Projection for Cross-View Gait Recognition', IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 3, pp. 734-747.
View/Download from: Publisher's site
View description>>

© 1991-2012 IEEE. A problem that hinders good performance of general gait recognition systems is that the appearance features of gaits are more affected-prone by views than identities, especially when the walking direction of the probe gait is different from the register gait. This problem cannot be solved by traditional projection learning methods because these methods can learn only one projection matrix, and thus for the same subject, it cannot transfer cross-view gait features into similar ones. This paper presents an innovative method to overcome this problem by aligning gait energy images (GEIs) across views with the coupled bilinear discriminant projection (CBDP). Specifically, the CBDP generates the aligned gait matrix features for two views with two sets of bilinear transformation matrices, so that the original GEIs' spatial structure information can be preserved. By iteratively maximizing the ratio of inter-class distance metric to intra-class distance metric, the CBDP can learn the optimal matrix subspace where the GEIs across views are aligned in both horizontal and vertical coordinates. Therefore, the CBDP is also able to avoid the under-sample problem. We also theoretically prove that the upper and lower bounds of the objective function sequence of the CBDP are both monotonically increasing, so the convergence of the CBDP is demonstrated. In the terms of accuracy, the comparative experiments on the CASIA (B) and OU-ISIR gait databases show that our method is superior to the state-of-the-art cross-view gait recognition methods. More impressively, encouraging performance is obtained by our method even in matching a lateral-view gait with a frontal-view gait.

Beydoun, G, Hoffmann, A, Garcia, RV, Shen, J & Gill, A 2020, 'Towards an assessment framework of reuse: a knowledge-level analysis approach', Complex & Intelligent Systems, vol. 6, no. 1, pp. 87-95.
View/Download from: Publisher's site

Buchlak, QD, Esmaili, N, Leveque, J-C, Bennett, C, Piccardi, M & Farrokhi, F 2020, 'Ethical thinking machines in surgery and the requirement for clinical leadership', The American Journal of Surgery, vol. 220, no. 5, pp. 1372-1374.
View/Download from: Publisher's site

Buchlak, QD, Esmaili, N, Leveque, J-C, Farrokhi, F, Bennett, C, Piccardi, M & Sethi, RK 2020, 'Machine learning applications to clinical decision support in neurosurgery: an artificial intelligence augmented systematic review', Neurosurgical Review, vol. 43, no. 5, pp. 1235-1253.
View/Download from: Publisher's site
View description>>

© 2019, Springer-Verlag GmbH Germany, part of Springer Nature. Machine learning (ML) involves algorithms learning patterns in large, complex datasets to predict and classify. Algorithms include neural networks (NN), logistic regression (LR), and support vector machines (SVM). ML may generate substantial improvements in neurosurgery. This systematic review assessed the current state of neurosurgical ML applications and the performance of algorithms applied. Our systematic search strategy yielded 6866 results, 70 of which met inclusion criteria. Performance statistics analyzed included area under the receiver operating characteristics curve (AUC), accuracy, sensitivity, and specificity. Natural language processing (NLP) was used to model topics across the corpus and to identify keywords within surgical subspecialties. ML applications were heterogeneous. The densest cluster of studies focused on preoperative evaluation, planning, and outcome prediction in spine surgery. The main algorithms applied were NN, LR, and SVM. Input and output features varied widely and were listed to facilitate future research. The accuracy (F(2,19) = 6.56, p < 0.01) and specificity (F(2,16) = 5.57, p < 0.01) of NN, LR, and SVM differed significantly. NN algorithms demonstrated significantly higher accuracy than LR. SVM demonstrated significantly higher specificity than LR. We found no significant difference between NN, LR, and SVM AUC and sensitivity. NLP topic modeling reached maximum coherence at seven topics, which were defined by modeling approach, surgery type, and pathology themes. Keywords captured research foci within surgical domains. ML technology accurately predicts outcomes and facilitates clinical decision-making in neurosurgery. NNs frequently outperformed other algorithms on supervised learning tasks. This study identified gaps in the literature and opportunities for future neurosurgical ML research.

Cao, Y & Veitch, D 2020, 'Toward Trusted Time: Remote Server Vetting and the Misfiring Heart of Internet Timing', IEEE/ACM Transactions on Networking, vol. 28, no. 2, pp. 944-956.
View/Download from: Publisher's site
View description>>

© 1993-2012 IEEE. The core of the Internet's timekeeping system are the Stratum-1 timeservers, those connected to reference hardware, that anchor the server hierarchy. It is essential that these root servers are accurate and reliable, and this it is typically taken as a given. We examine this premise through an examination of 102 prominent Stratum-1 servers, using 3 datasets spanning 6 years, collected in reference testbeds with authoritative timestamping. We describe a methodology capable of rigorously removing congestion related variability, allowing server errors to be unambiguously revealed. We use the data and methodology to assess the health of public network timing, and how it varies over time, by reporting on the type, severity, duration, and prevalence of server errors, and how they relate to protocol level information. We present conclusive evidence that the system has problems. We find that errors are widespread, significant, often endemic, consistent over time, and typically come with no warning at the protocol level. Our results highlight the lack of oversight in the current system, and provides the foundation of a server health monitoring capability, necessary to restore and maintain trust in network timing. We describe three specific applications where our results can have an impact. Our data, detailed results and software are publically available.

Chaczko, Z, Klempous, R, Rozenblit, J, Adegbija, T, Chiu, C, Kluwak, K & Smutnick, C 2020, 'Biomimetic Middleware Design Principles for IoT Infrastructures', Acta Polytechnica Hungarica, vol. 17, no. 5, pp. 135-150.
View/Download from: Publisher's site

Chen, S, Zhou, F, Xu, K, Zhao, P, Yang, Y, Zhu, X & Wang, G 2020, 'A Portable Microwave Interferometry Sensor for Permittivity Detection Based on CCMRC', IEEE Access, vol. 8, pp. 140323-140332.
View/Download from: Publisher's site
View description>>

© 2013 IEEE. A portable microwave complex permittivity sensor based on the interferometry configuration is proposed. A complementary compact microstrip resonant cell (CCMRC) is applied as the sensitive element, which converts the dielectric information of the material under test (MUT) into the phase variations of its transmission coefficient. A miniaturized interferometry platform based on a down-converting mixer further translates the phase change into DC output voltage variation, which can be readily recorded with a direct readout circuit. In this context, expensive and bulky vector network analyzer is no longer needed, thereby leading to a low hardware cost. With comprehensive theoretical analysis, the material permittivity is simply extracted using a specific extrapolation algorithm. As a proof of concept, several different solid material samples with known permittivity values are used to verify the devised detection system.

Chen, S-L, Karmokar, DK, Qin, P-Y, Ziolkowski, RW & Guo, YJ 2020, 'Polarization-Reconfigurable Leaky-Wave Antenna With Continuous Beam Scanning Through Broadside', IEEE Transactions on Antennas and Propagation, vol. 68, no. 1, pp. 121-133.
View/Download from: Publisher's site
View description>>

© 2019 IEEE. A simple single-layer reconfigurable leaky-wave antenna (LWA) is presented that has polarization agility and beam-scanning functionality. This LWA system realizes a scanned beam that can be switched between all of its linear polarization (LP) and circular polarization (CP) states using only one dc biasing source. A slot-loaded substrate-integrated waveguide (SIW)-based LWA is first explored to attain CP performance with continuous beam scanning through broadside. This CP LWA realizes a measured CP performance with a 3 dB gain variance within 2.75-3.35 GHz for scan angles ranging from -28.6° to +31.5°. A row of shorted stubs is then incorporated into the CP LWA to obtain similar LP performance. Finally, by introducing p-i-n diodes into this LP LWA configuration to facilitate reconfigurable connections between the main patch and the shorted stubs, the radiated fields can be switched between all of its CP and LP states. The measured results of all three antennas confirm their simulated performance. It is demonstrated that the main beam of the polarization-reconfigurable LWA can be scanned from -31.5° to +17.1° with gain variations between 9.5 and 12.8 dBic in its CP state and from -34.3° to +20° with them between 7.8 and 11.7 dBi in its LP state.

Chen, Y, An, P, Huang, X, Yang, C, Liu, D & Wu, Q 2020, 'Light Field Compression Using Global Multiplane Representation and Two-Step Prediction', IEEE Signal Processing Letters, vol. 27, pp. 1135-1139.
View/Download from: Publisher's site
View description>>

© 1994-2012 IEEE. Due to its spatio-angular structure, light field image allows for a wealth of post-processing techniques like digital refocusing and depth estimation. In order to compress the data of the two domains, the current proposal intends to embed the disparity-based view synthesis method into the decoder. However, predicting each view separately or in local groups means bringing more computational burden to the decoder and destroying the light field structure. Since disparity contains the relationship between all light rays in the light field, the proposed solution is to predict a disparity-based global representation as the first step. In the second step, all the views can be predicted easily based on this representation. In this letter, we use the recently proposed multiplane as the form of this global representation. The experimental results show the effectiveness of the proposed solution, and the better RD performance compared to other schemes especially under low bitrates.

Cheng, Q, Nguyen, DN, Dutkiewicz, E & Mueck, M 2020, 'Preserving Honest/Dishonest Users’ Operational Privacy with Blind Interference Calculation in Spectrum Sharing System', IEEE Transactions on Mobile Computing, vol. 19, no. 12, pp. 2874-2890.
View/Download from: Publisher's site
View description>>

In this paper, we investigate the operational privacy issue of Incumbent Users (IUs) and honest/dishonest Secondary Users (SUs). For the case of IUs and honest SUs, we propose a privacy-preserving scheme for DSA by leveraging encryption and obfuscation methods (PSEO). To implement PSEO, we introduce an interference calculation scheme that allows users to calculate an interference budget without revealing operational information, referred to as the blind interference calculation scheme (BICS). BICS also reduces the computing overhead of PSEO, compared with FCC's SAS by moving interference budgeting tasks to local users and calculating it in an offline manner. To further save the overhead in calculating the interference map, we introduce a quantization method and optimize the grid sizes of the terrestrial area of interest. Additionally, for the case of IUs and dishonest SUs, we propose a "punishment and forgiveness" (PF) mechanism, which draws support from SUs' reputation scores (RSs) and reputation histories (RHs), to encourage SUs to provide truthful information. Theoretical analysis and extensive simulations show that our proposed PSEO and PF-PSEO schemes can better protect all users' operational privacy under various privacy attacks, yielding higher spectrum utilization with less online overhead, compared with state of the art approaches

Cui, Q, Ni, W, Li, S, Zhao, B, Liu, RP & Zhang, P 2020, 'Learning-Assisted Clustered Access of 5G/B5G Networks to Unlicensed Spectrum', IEEE Wireless Communications, vol. 27, no. 1, pp. 31-37.
View/Download from: Publisher's site

Dandachi, G, De Domenico, A, Hoang, DT & Niyato, D 2020, 'An Artificial Intelligence Framework for Slice Deployment and Orchestration in 5G Networks', IEEE Transactions on Cognitive Communications and Networking, vol. 6, no. 2, pp. 858-871.
View/Download from: Publisher's site
View description>>

© 2015 IEEE. Network slicing is a key enabler to successfully support 5G services with specific requirements and priorities. Due to the diversity of these services, slice deployment and orchestration are essential to guarantee service performance in a cost-effective way. Here, we propose an Artificial Intelligence framework for cross-slice admission and congestion control that simultaneously considers communication, computing, and storage resources to maximize resources utilization and operator revenue. First, we propose a smart feature extraction solution to analyze the characteristics of incoming requests together with the already deployed slices, and then automatically evaluates the request requirements to make appropriate decisions. Second, we design an online algorithm that controls the slice admission based on their priorities, the arrival and departure characteristics, and the available resources. To mitigate system overloading, our framework dynamically adjusts resources allocated to low priority slices, thereby reducing the dropping probability of new slice requests. The proposed algorithm offers outstanding advantages over traditional static approaches by automatically adapting the controller decisions to the system changes. Simulation results show that our framework significantly improves the resource utilization and reduces the slice request dropping probabilities up to 44% as compared to the baseline schemes.

Ding, C, Sun, H-H, Zhu, H & Jay Guo, Y 2020, 'Achieving Wider Bandwidth With Full-Wavelength Dipoles for 5G Base Stations', IEEE Transactions on Antennas and Propagation, vol. 68, no. 2, pp. 1119-1127.
View/Download from: Publisher's site
View description>>

© 1963-2012 IEEE. A new method of designing full-wavelength dipoles (FWDs) is presented. A dual-polarized antenna is built based on FWDs for base station applications as an example. The antenna has four FWDs arranged in a square loop array form. The employed FWDs are bent upward to maintain a small aperture size, so that the realized element still fits in traditional base station antenna (BSA) array. The antenna is first matched across the band from 1.63 to 3.71 GHz, which can cover both the LTE band from 1.7 to 2.7 GHz and the 5G (sub-6 GHz) band from 3.3 to 3.6 GHz simultaneously. Then, band-stop filters are inserted in the feed networks of the antenna to suppress the radiation between 2.7 to 3.3 GHz. The antenna is fabricated and tested. Experimental results validate the simulation results. Comparing with the previously available FWD that has a bandwidth of 32%, the FWD proposed in this article exhibits a much wider bandwidth of 78%. Moreover, this bandwidth is also comparable to and wider than those of the state-of-the-art BSAs based on half-wavelength dipoles (HWDs). The bandwidth enhancement and footprint reduction of the FWD in this article demonstrate a high potential of FWDs to be used in other applications.

Farokhipour, E, Mehrabi, M, Komjani, N & Ding, C 2020, 'A Spoof Surface Plasmon Polaritons (SSPPs) Based Dual-Band-Rejection Filter with Wide Rejection Bandwidth', Sensors, vol. 20, no. 24, pp. 7311-7311.
View/Download from: Publisher's site
View description>>

This paper presents a novel single-layer dual band-rejection-filter based on Spoof Surface Plasmon Polaritons (SSPPs). The filter consists of an SSPP-based transmission line, as well as six coupled circular ring resonators (CCRRs) etched among ground planes of the center corrugated strip. These resonators are excited by electric-field of the SSPP structure. The added ground on both sides of the strip yields tighter electromagnetic fields and improves the filter performance at lower frequencies. By removing flaring ground in comparison to prevalent SSPP-based constructions, the total size of the filter is significantly decreased, and mode conversion efficiency at the transition from co-planar waveguide (CPW) to the SSPP line is increased. The proposed filter possesses tunable rejection bandwidth, wide stop bands, and a variety of different parameters to adjust the forbidden bands and the filter’s cut-off frequency. To demonstrate the filter tunability, the effect of different elements like number (n), width (WR), radius (RR) of CCRRs, and their distance to the SSPP line (yR) are surveyed. Two forbidden bands, located in the X and K bands, are 8.6–11.2 GHz and 20–21.8 GHz. As the proof-of-concept, the proposed filter was fabricated, and a good agreement between the simulation and experiment results was achieved.

Gao, X, Zhang, T, Du, J & Guo, YJ 2020, '340 GHz Double-Sideband Mixer Based on Antenna-Coupled High-Temperature Superconducting Josephson Junction', IEEE Transactions on Terahertz Science and Technology, vol. 10, no. 1, pp. 21-31.
View/Download from: Publisher's site
View description>>

© 2011-2012 IEEE. Wireless communication and sensing are moving from microwave, millimeter-wave into the terahertz (THz) frequency regime to meet the fast growing demand of ultrahigh data-rate communications and super resolution imaging. Faced with severe atmospheric absorption attenuation and the lack of power efficient transmitting source at the higher band, ultrasensitive and cost-effective receiver frontend technology is required for advanced THz wireless systems. To date, the most sensitive heterodyne mixers, the key components of frontend receiver systems, are based on low-temperature superconducting materials that operate at liquid helium (4.2 K) temperature range, requiring expensive and bulky cryogenic cooling systems thus hindering them from commercial applications such as wireless communications and sensing. In this article, we present a 340 GHz double-sideband fundamental mixer based on thin-film antenna-coupled high-temperature superconducting (HTS) Josephson junction that operates at a much higher temperature range attainable with smaller and cheaper cryocoolers. Based on our innovative work in terms of advanced device circuit and on-chip antenna designs, accurate parametric simulation analyses, and Josephson junction parameter optimizations, the reported mixer exhibits a measured noise temperature of 470 and 780 K at operating temperatures of 20 and 40 K respectively at 340 GHz, a performance significantly higher than any HTS THz mixers reported to date.

Gong, S, Lu, X, Hoang, DT, Niyato, D, Shu, L, Kim, DI & Liang, Y-C 2020, 'Toward Smart Wireless Communications via Intelligent Reflecting Surfaces: A Contemporary Survey', IEEE Communications Surveys & Tutorials, vol. 22, no. 4, pp. 2283-2314.
View/Download from: Publisher's site
View description>>

© 1998-2012 IEEE. This paper presents a literature review on recent applications and design aspects of the intelligent reflecting surface (IRS) in the future wireless networks. Conventionally, the network optimization has been limited to transmission control at two endpoints, i.e., end users and network controller. The fading wireless channel is uncontrollable and becomes one of the main limiting factors for performance improvement. The IRS is composed of a large array of scattering elements, which can be individually configured to generate additional phase shifts to the signal reflections. Hence, it can actively control the signal propagation properties in favor of signal reception, and thus realize the notion of a smart radio environment. As such, the IRS's phase control, combined with the conventional transmission control, can potentially bring performance gain compared to wireless networks without IRS. In this survey, we first introduce basic concepts of the IRS and the realizations of its reconfigurability. Then, we focus on applications of the IRS in wireless communications. We overview different performance metrics and analytical approaches to characterize the performance improvement of IRS-assisted wireless networks. To exploit the performance gain, we discuss the joint optimization of the IRS's phase control and the transceivers' transmission control in different network design problems, e.g., rate maximization and power minimization problems. Furthermore, we extend the discussion of IRS-assisted wireless networks to some emerging use cases. Finally, we highlight important practical challenges and future research directions for realizing IRS-assisted wireless networks in beyond 5G communications.

Gong, S, Zou, Y, Hoang, DT, Xu, J, Cheng, W & Niyato, D 2020, 'Capitalizing Backscatter-Aided Hybrid Relay Communications With Wireless Energy Harvesting', IEEE Internet of Things Journal, vol. 7, no. 9, pp. 8709-8721.
View/Download from: Publisher's site

Gong, Y, Zhang, L, Yu, K & Liu, R 2020, 'Exploring Uplink Achievable Rate for HPO MIMO Through Quasi-Monte Carlo and Variance Reduction Techniques', IEEE Access, vol. 8, pp. 75874-75883.
View/Download from: Publisher's site

Grymin, R, Bożejko, W, Chaczko, Z, Pempera, J & Wodecki, M 2020, 'Algorithm for solving the Discrete-Continuous Inspection Problem', Archives of Control Sciences, vol. 30, no. 4, pp. 653-666.
View/Download from: Publisher's site
View description>>

The article introduces an innovative approch for the inspection challenge that represents a generalization of the classical Traveling Salesman Problem. Its priciple idea is to visit continuous areas (circles) in a way, that minimizes travelled distance. In practice, the problem can be defined as an issue of scheduling unmanned aerial vehicle which has discrete-continuous nature. In order to solve this problem the use of local search algorithms is proposed.

Hanawal, MK, Nguyen, DN & Krunz, M 2020, 'Cognitive Networks With In-Band Full-Duplex Radios: Jamming Attacks and Countermeasures', IEEE Transactions on Cognitive Communications and Networking, vol. 6, no. 1, pp. 296-309.
View/Download from: Publisher's site
View description>>

© 2015 IEEE. Although in-band full-duplex (IBFD) radios promise to double the throughput of a wireless link, they are more vulnerable to jamming attacks than their out-of-band full-duplex (OBFD) counterparts. For two communicating OBFD nodes, a jammer needs to attack both the uplink and the downlink channels to completely break the communication link. In contrast, only one common channel needs to be jammed in the case of two IBFD nodes. Even worse, a jammer with self-interference suppression (SIS) capabilities (the underlying technique of IBFD radios) can learn the transmitters' activity while injecting interference, allowing it to react instantly to the transmitter's strategies. In this work, we consider a power-constrained IBFD 'reactive-sweep' jammer that sweeps through the set of channels by jamming a subset of them simultaneously. We model the interactions between the IBFD radios and the jammer as a stochastic constrained zero-sum Markov game in which nodes adopt the frequency hopping (FH) technique as their strategies to counter jamming attacks. Beside the IBFD transmission-reception (TR) mode, we introduce an additional operation mode, called transmission-detection (TD), in which an IBFD radio transmits and leverages its SIS capability to detect jammers. The aim of the TD mode is to make IBFD radios more cognitive to jamming. The nodes' optimal defense strategy that guides them when to hop and which operational mode (TD or TR) to use is then established from the equilibrium of the stochastic Markov game. We prove that this optimal policy has a threshold structure, in which IBFD nodes stay on the same channel up to a certain number of time slots before hopping. Simulation results show that our policy significantly improves the throughput of IBFD nodes under jamming attacks.

He, S, Lyu, X, Ni, W, Tian, H, Liu, RP & Hossain, E 2020, 'Virtual Service Placement for Edge Computing Under Finite Memory and Bandwidth', IEEE Transactions on Communications, vol. 68, no. 12, pp. 7702-7718.
View/Download from: Publisher's site
View description>>

© 1972-2012 IEEE. Edge computing allows an edge server to adaptively place virtual instances to serve different types of data. This article presents a new algorithm which jointly optimizes virtual service placement farsightedly and service data admission instantly to maximize the time-average service throughput of edge computing. The data admission is optimized, adapting to fast-changing data arrivals and wireless channels. The service placement is transformed into a two-dimensional knapsack problem by approximating future arrivals and channels with past observations, and solved over a slow timescale to allow services to be properly installed. Different from existing studies, our algorithm considers practical aspects of edge servers, such as finite memory size and bandwidth. We prove that the algorithm is asymptotically optimal and the optimality loss resulting from the approximation diminishes. Simulations show that our approach can improve the time-average throughput of existing alternatives by 16% for our considered simulation setup. The improvement becomes higher, as the memory size becomes increasingly tight. The number of services to be replaced is reduced without loss of throughput, after being placed farsightedly.

Heidary, A, Radmanesh, H, Bakhshi, A, Samandarpour, S, Rouzbehi, K & Shariati, N 2020, 'Compound ferroresonance overvoltage and fault current limiter for power system protection', IET Energy Systems Integration, vol. 2, no. 4, pp. 325-330.
View/Download from: Publisher's site
View description>>

Power systems are subjected to various types of faults as well as ferroresonance overvoltages. These results in the interruption of the normal operation of the power grid, failure of equipment, electrical fires, etc. To tackle these issues, this study proposes a dual function limiter to control the fault current and ferroresonance phenomenon in power systems. This compound device is a solid-state series transformer-based limiter that includes IGBT switches, capacitors, rectifiers, and a DC reactor. During the grid normal operation, the proposed limiter is not active and therefore is invisible and it operates in the instant of fault inception or ferroresonance overvoltage occurrences. Analytical studies in all operation modes are presented and assessments on the performance of the proposed ferroresonance and fault current limiter (FFCL) are conducted in Matlab. Simulation results confirm the reported analytical studies and FFCL's performance.

Heidary, A, Radmanesh, H, Naghibi, SH, Samandarpour, S, Rouzbehi, K & Shariati, N 2020, 'Distribution system protection by coordinated fault current limiters', IET Energy Systems Integration, vol. 2, no. 1, pp. 59-65.
View/Download from: Publisher's site
View description>>

The protection of distribution networks is one of the most substantial issues, which needs special attention. Using appropriate protective equipment enhances the safety of the power distribution network during the fault conditions. Fault current limiter (FCL) is a kind of modern preserving system being used for protecting power networks and equipment. One of the main concerns of power networks is the voltage restoration of buses during faulty conditions. In this study, a group of coordinated DC reactor type faults current limiters are designed and tested to protect the network and restore its buses voltage within the fault period. To coordinate FCLs and measurement devices during the fault sequences, a wireless communication system and decision-making computer are used. The proposed FCLs coordination strategy is modelled and simulated in MATLAB platform and the results are validated by the developed laboratory test setup.

Hieu, NQ, Hoang, DT, Luong, NC & Niyato, D 2020, 'iRDRC: An Intelligent Real-Time Dual-Functional Radar-Communication System for Automotive Vehicles', IEEE Wireless Communications Letters, vol. 9, no. 12, pp. 2140-2143.
View/Download from: Publisher's site
View description>>

© 2012 IEEE. This letter introduces an intelligent Real-time Dual-functional Radar-Communication (iRDRC) system for autonomous vehicles (AVs). This system enables an AV to perform both radar and data communications functions to maximize bandwidth utilization as well as significantly enhance safety. In particular, the data communications function allows the AV to transmit data, e.g., of current traffic, to edge computing systems and the radar function is used to enhance the reliability and reduce the collision risks of the AV, e.g., under bad weather conditions. The problem of the iRDRC is to decide when to use the communication mode or the radar mode to maximize the data throughput while minimizing the miss detection probability of unexpected events given the uncertainty of surrounding environment. To solve the problem, we develop a deep reinforcement learning algorithm that allows the AV to quickly obtain the optimal policy without requiring any prior information about the environment. Simulation results show that the proposed scheme outperforms baseline schemes in terms of data throughput, miss detection probability, and convergence rate.

Hu, Z, Liu, RP, Ni, W, Wen, X, Lu, Z & Dutkiewicz, E 2020, 'Analysis of Clustered Licensed-Assisted Access in Unlicensed Spectrum', IEEE Transactions on Vehicular Technology, vol. 69, no. 1, pp. 349-360.
View/Download from: Publisher's site
View description>>

© 1967-2012 IEEE. Faced with explosive growth of data traffic and shortage of licensed spectrum, licensed-assisted access (LAA) to unlicensed spectrum has been proposed to boost system capacity. To ensure fair coexistence with WiFi systems, listen-before-talk mechanism has been standardized under LAA framework. However, in densely deployed urban networks, the system performance could severely deteriorate due to high collision probability. In this paper, we propose cooperative LAA (CLAA), where multiple LAA small base stations form a cluster and construct a virtual multiuser multiple-input single-output (MISO)/multiple-input and multiple-output (MIMO) system to transmit data cooperatively. CLAA can effectively reduce the number of contending nodes, thereby alleviating transmission collisions. A closed-form expression for the upper bound sum rate of the cluster is derived. Markov analysis is employed to derive the system collision probability and throughput for WiFi and LTE. Our analytical results point to an adequate cluster sizes, where the highest system throughput can be achieved. Extensive simulations confirm the validity of the proposed approach, and demonstrate that CLAA can increase by up to 27% the overall system throughput, and improve by 30% in fairness.

Huang, L, Yang, Q, Wu, J, Huang, Y, Wu, Q & Xu, J 2020, 'Generated Data With Sparse Regularized Multi-Pseudo Label for Person Re-Identification', IEEE Signal Processing Letters, vol. 27, pp. 391-395.
View/Download from: Publisher's site
View description>>

© 1994-2012 IEEE. Recently, Generative Adversarial Network (GAN) has been adopted to improve person re-identification (person re-ID) performance through data augmentation. However, directly leveraging generated data to train a re-ID model may easily lead to over-fitting issue on these extra data and decrease the generalisability of model to learn true ID-related features from real data. Inspired by the previous approach which assigns multi-pseudo labels on the generated data to reduce the risk of over-fitting, we propose to take sparse regularization into consideration. We attempt to further improve the performance of current re-ID models by using the unlabeled generated data. The proposed Sparse Regularized Multi-Pseudo Label (SRMpL) can effectively prevent the over-fitting issue when some larger weights are assigned to the generated data. Our experiments are carried out on two publicly available person re-ID datasets (e.g., Market-1501 and DukeMTMC-reID). Compared with existing unlabeled generated data re-ID solutions, our approach achieves competitive performance. Two classical re-ID models are used to verify our sparse regularization label on generated data, i.e., an ID-embedding network and a two-stream network.

Huang, Y, Xu, J, Wu, Q, Zhong, Y, Zhang, P & Zhang, Z 2020, 'Beyond Scalar Neuron: Adopting Vector-Neuron Capsules for Long-Term Person Re-Identification', IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 10, pp. 3459-3471.
View/Download from: Publisher's site
View description>>

Current person re-identification (re-ID) works mainly focus on the short-term scenario where a person is less likely to change clothes. However, in the long-term re-ID scenario, a person has a great chance to change clothes. A sophisticated re-ID system should take such changes into account. To facilitate the study of long-term re-ID, this paper introduces a large-scale re-ID dataset called “Celeb-reID” to the community. Unlike previous datasets, the same person can change clothes in the proposed Celeb-reID dataset. Images of Celeb-reID are acquired from the Internet using street snap-shots of celebrities. There is a total of 1,052 IDs with 34,186 images making Celeb-reID being the largest long-term re-ID dataset so far. To tackle the challenge of cloth changes, we propose to use vector-neuron (VN) capsules instead of the traditional scalar neurons (SN) to design our network. Compared with SN, one extra-dimensional information in VN can perceive cloth changes of the same person. We introduce a well-designed ReIDCaps network and integrate capsules to deal with the person re-ID task. Soft Embedding Attention (SEA) and Feature Sparse Representation (FSR) mechanisms are adopted in our network for performance boosting. Experiments are conducted on the proposed long-term re-ID dataset and two common short-term re-ID datasets. Comprehensive analyses are given to demonstrate the challenge exposed in our datasets. Experimental results show that our ReIDCaps can outperform existing state-of-the-art methods by a large margin in the long-term scenario. The new dataset and code will be released to facilitate future researches.

Huynh, NV, Hoang, DT, Nguyen, DN & Dutkiewicz, E 2020, 'DeepFake: Deep Dueling-based Deception Strategy to Defeat Reactive Jammers'.
View description>>

In this paper, we introduce DeepFake, a novel deep reinforcementlearning-based deception strategy to deal with reactive jamming attacks. Inparticular, for a smart and reactive jamming attack, the jammer is able tosense the channel and attack the channel if it detects communications from thelegitimate transmitter. To deal with such attacks, we propose an intelligentdeception strategy which allows the legitimate transmitter to transmit 'fake'signals to attract the jammer. Then, if the jammer attacks the channel, thetransmitter can leverage the strong jamming signals to transmit data by usingambient backscatter communication technology or harvest energy from the strongjamming signals for future use. By doing so, we can not only undermine theattack ability of the jammer, but also utilize jamming signals to improve thesystem performance. To effectively learn from and adapt to the dynamic anduncertainty of jamming attacks, we develop a novel deep reinforcement learningalgorithm using the deep dueling neural network architecture to obtain theoptimal policy with thousand times faster than those of the conventionalreinforcement algorithms. Extensive simulation results reveal that our proposedDeepFake framework is superior to other anti-jamming strategies in terms ofthroughput, packet loss, and learning rate.

Huynh, NV, Nguyen, DN, Hoang, DT & Dutkiewicz, E 2020, 'Optimal Beam Association for High Mobility mmWave Vehicular Networks: Lightweight Parallel Reinforcement Learning Approach'.
View description>>

In intelligent transportation systems (ITS), vehicles are expected to featurewith advanced applications and services which demand ultra-high data rates andlow-latency communications. For that, the millimeter wave (mmWave)communication has been emerging as a very promising solution. However,incorporating the mmWave into ITS is particularly challenging due to the highmobility of vehicles and the inherent sensitivity of mmWave beams to dynamicblockages. This article addresses these problems by developing an optimal beamassociation framework for mmWave vehicular networks under high mobility.Specifically, we use the semi-Markov decision process to capture the dynamicsand uncertainty of the environment. The Q-learning algorithm is then often usedto find the optimal policy. However, Q-learning is notorious for itsslow-convergence. Instead of adopting deep reinforcement learning structures(like most works in the literature), we leverage the fact that there areusually multiple vehicles on the road to speed up the learning process. To thatend, we develop a lightweight yet very effective parallel Q-learning algorithmto quickly obtain the optimal policy by simultaneously learning from variousvehicles. Extensive simulations demonstrate that our proposed solution canincrease the data rate by 47% and reduce the disconnection probability by 29%compared to other solutions.

Jafarizadeh, S, Tofigh, F, Lipman, J & Abolhasan, M 2020, 'Optimizing synchronizability in networks of coupled systems', Automatica, vol. 112, pp. 108711-108711.
View/Download from: Publisher's site
View description>>

© 2019 Elsevier Ltd Of collective behaviors in networks of coupled systems, synchronization is of central importance and an extensively studied area. This is due to the fact that it is essential for the proper functioning of a wide variety of natural and engineered systems. Traditionally, uniform coupling strength has been the default choice and the synchronizability measure has been employed for analysis and enhancement of synchronizability. The main drawback of optimizing the synchronizability measure is that it can reach the Pareto frontier but not necessarily a unique point on the Pareto frontier. Additionally, the shortcoming of uniform coupling strength is that it can reach Pareto frontier in specific topologies including edge-transitive graphs. To achieve a unique optimal answer on the Pareto frontier, this paper takes a different approach and addresses the synchronizability in networks of coupled dynamical systems with nonuniform coupling strength and optimizing the synchronizability via maximizing the minimum distance between the nonzero eigenvalues of the Laplacian and the acceptable boundaries for the stability of the system. Furthermore, two solution methods, namely the concave–convex fractional programming and the Semidefinite Programming (SDP) formulations of the problem have been provided. The proposed solution methods have been compared over different topologies and branches of an arbitrary network, where the SDP based approach has shown to be less restricted and more suitable for a wider range of topologies.

Ju, M, Ding, C & Guo, YJ 2020, 'VROHI: Visibility Recovery for Outdoor Hazy Image in Scattering Media', IEEE Photonics Journal, vol. 12, no. 6, pp. 1-15.
View/Download from: Publisher's site
View description>>

© 2009-2012 IEEE. Additive haze model (AHM), due to its high simplicity, has a potential to increase the efficiency of the restoration procedure of images degraded by scattering media. However, AHM is designed for hazy remote sensing data and is not suitable to be used on outdoor images. In this paper, according to the low-frequency feature (LFC) of haze, AHM is modified via gamma correction technique to make it suitable for modeling outdoor images. Benefitting from the modified AHM (MAHM), a simple yet effective method called VROHI is proposed to enhance the visibility of an outdoor hazy image. In specific, a low complexity LFC extraction method is designed by utilizing characteristic of the discrete cosine transform. Subsequently, by constructing the linear function of unknown parameters and imposing the saturation prior on MAHM, the image dehazing problem can be derived into a global optimization function. Experiments reveal that the proposed VROHI is superior to the other state-of-the-art techniques in terms of both the processing efficiency and recovery quality.

Ju, M, Ding, C, Guo, YJ & Zhang, D 2020, 'IDGCP: Image Dehazing Based on Gamma Correction Prior', IEEE Transactions on Image Processing, vol. 29, pp. 3104-3118.
View/Download from: Publisher's site
View description>>

© 1992-2012 IEEE. This paper introduces a novel and effective image prior, i.e., gamma correction prior (GCP), which leads to an efficient image dehazing method, i.e., IDGCP. A step-by-step procedure of the proposed IDGCP is as follows. First, an input hazy image is preprocessed by the proposed GCP, resulting in a homogeneous virtual transformation of the hazy image. Then, from the original input hazy image and its virtual transformation, the depth ratio is extracted based on atmospheric scattering theory. Finally, a 'global-wise' strategy and a vision indicator are employed to recover the scene albedo, thus restoring the hazy image. Unlike other image dehazing methods, IDGCP is based on the 'global-wise' strategy, and it only needs to determine one unknown constant without any refining process to attain a high-quality restoration, thereby leading to significantly reduced processing time and computation cost. Each step of IDGCP is tested experimentally to validate its robustness. Moreover, a series of experiments are conducted on a number of challenging images with IDGCP and other state-of-the-art technologies, demonstrating the superiority of IDGCP over the others in terms of restoration quality and implementation efficiency.

Karmokar, DK, Chen, S-L, Thalakotuna, D, Qin, P-Y, Bird, TS & Guo, YJ 2020, 'Continuous Backward-to-Forward Scanning 1-D Slot-Array Leaky-Wave Antenna With Improved Gain', IEEE Antennas and Wireless Propagation Letters, vol. 19, no. 1, pp. 89-93.
View/Download from: Publisher's site

Karmokar, DK, Guo, YJ, Chen, S-L & Bird, TS 2020, 'Composite Right/Left-Handed Leaky-Wave Antennas for Wide-Angle Beam Scanning With Flexibly Chosen Frequency Range', IEEE Transactions on Antennas and Propagation, vol. 68, no. 1, pp. 100-110.
View/Download from: Publisher's site
View description>>

© 2019 IEEE. A composite right/left-handed (CRLH) leaky-wave antenna (LWA) can effectively scan the radiation beam from backward-to-forward direction. However, in most cases, a large range of frequency sweep is required to achieve a wide-angle beam scan, which could limit their applications. An in-depth study is conducted on an equivalent circuit model for a CRLH LWA unit cell to find the controlling parameters on the frequency sweeping range. A systematic design guideline is given for a CRLH LWA for a wide-angle beam scan in a flexibly chosen frequency range. It is shown that beam scanning by sweeping frequency in a target range can be achieved by systematically designing the unit cell parameters. To verify our approach, a novel CRLH unit cell is developed and used to design an LWA for a wide-angle beam scan in a narrow frequency range. Finally, the concept is validated through realization of the antenna and its measurement. The measured results show that the antenna prototype can scan its beam from -56° to +51° when frequency sweeps from 5.1 to 6.11 GHz (i.e., 18.02% of fractional bandwidth).

Khan, HU, ARUYA, JOYA & Gill, AQ 2020, 'Web 2.0 Technologies Adoption Barriers for External Contacts and Participation: A Case Study of Federal Establishment of Africa', International Journal of Business Information Systems, vol. 1, no. 1, pp. 1-1.
View/Download from: Publisher's site

Kong, L, Liu, H, Zhu, X, Boon, CC, Li, C, Liu, Z & Yeo, KS 2020, 'Design of a Wideband Variable-Gain Amplifier With Self-Compensated Transistor for Accurate dB-Linear Characteristic in 65 nm CMOS Technology', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 12, pp. 4187-4198.
View/Download from: Publisher's site
View description>>

© 2004-2012 IEEE. A simple yet effective approach for variable-gain amplifier (VGA) design with accurate dB-linear characteristic is presented. In order to extend the bandwidth of the designed VGA with a minimized footprint, an inductorless-based approach is adopted. Moreover, a unique approach that exploits a self-compensated transistor to compensate dB-linear gain error is proposed. Consequently, the overall VGA has an accurate dB-linear inherent characteristic without using any additional exponential generator for gain control. To prove the concept, the designed VGA is fabricated in a standard 65 nm CMOS technology. The measured results show that the voltage gain of the designed VGA can be controlled from -19 dB to 21 dB with a gain error less than 1 dB. Meanwhile, more than 4 GHz of bandwidth can be achieved for the entire gain range. The power consumption of the VGA, excluding the output buffer, is 3.9 mW. The core circuit of this design only occupies an area of 0.012 mm2.

Le, AT, Tran, LC, Huang, X & Guo, YJ 2020, 'Analog Least Mean Square Loop for Self-Interference Cancellation: A Practical Perspective', Sensors, vol. 20, no. 1, pp. 270-270.
View/Download from: Publisher's site
View description>>

Self-interference (SI) is the key issue that prevents in-band full-duplex (IBFD) communications from being practical. Analog multi-tap adaptive filter is an efficient structure to cancel SI since it can capture the nonlinear components and noise in the transmitted signal. Analog least mean square (ALMS) loop is a simple adaptive filter that can be implemented by purely analog means to sufficiently mitigate SI. Comprehensive analyses on the behaviors of the ALMS loop have been published in the literature. This paper proposes a practical structure and presents an implementation of the ALMS loop. By employing off-the-shelf components, a prototype of the ALMS loop including two taps is implemented for an IBFD system operating at the carrier frequency of 2.4 GHz. The prototype is firstly evaluated in a single carrier signaling IBFD system with 20 MHz and 50 MHz bandwidths, respectively. Measured results show that the ALMS loop can provide 39 dB and 33 dB of SI cancellation in the radio frequency domain for the two bandwidths, respectively. Furthermore, the impact of the roll-off factor of the pulse shaping filter on the SI cancellation level provided by the prototype is presented. Finally, the experiment with multicarrier signaling shows that the performance of the ALMS loop is the same as that in the single carrier system. These experimental results validate the theoretical analyses presented in our previous publications on the ALMS loop behaviors.

Le, AT, Tran, LC, Huang, X & Guo, YJ 2020, 'Beam-Based Analog Self-Interference Cancellation in Full-Duplex MIMO Systems', IEEE Transactions on Wireless Communications, vol. 19, no. 4, pp. 2460-2471.
View/Download from: Publisher's site

Le, AT, Tran, LC, Huang, X, Ritz, CH, Dutkiewicz, E, Phung, SL, Bouzerdoum, A & Franklin, DR 2020, 'Unbalanced Hybrid AOA/RSSI Localization for Simplified Wireless Sensor Networks.', Sensors, vol. 20, no. 14, pp. 3838-3838.
View/Download from: Publisher's site
View description>>

Source positioning using hybrid angle-of-arrival (AOA) estimation and received signal strength indicator (RSSI) is attractive because no synchronization is required among unknown nodes and anchors. Conventionally, hybrid AOA/RSSI localization combines the same number of these measurements to estimate the agents’ locations. However, since AOA estimation requires anchors to be equipped with large antenna arrays and complicated signal processing, this conventional combination makes the wireless sensor network (WSN) complicated. This paper proposes an unbalanced integration of the two measurements, called 1AOA/nRSSI, to simplify the WSN. Instead of using many anchors with large antenna arrays, the proposed method only requires one master anchor to provide one AOA estimation, while other anchors are simple single-antenna transceivers. By simply transforming the 1AOA/1RSSI information into two corresponding virtual anchors, the problem of integrating one AOA and N RSSI measurements is solved using the least square and subspace methods. The solutions are then evaluated to characterize the impact of angular and distance measurement errors. Simulation results show that the proposed network achieves the same level of precision as in a fully hybrid nAOA/nRSSI network with a slightly higher number of simple anchors.

Li, C, Xie, H-B, Mengersen, K, Fan, X, Da Xu, RY, Sisson, SA & Van Huffel, S 2020, 'Bayesian Nonnegative Matrix Factorization With Dirichlet Process Mixtures', IEEE Transactions on Signal Processing, vol. 68, pp. 3860-3870.
View/Download from: Publisher's site
View description>>

Nonnegative Matrix Factorization (NMF) is valuable in many applications of blind source separation, signal processing and machine learning. A number of algorithms that can infer nonnegative latent factors have been developed, but most of these assume a specific noise kernel. This is insufficient to deal with complex noise in real scenarios. In this paper, we present a hierarchical Dirichlet process nonnegative matrix factorization (DPNMF) model in which the Gaussian mixture model is used to approximate the complex noise distribution. Moreover, the model is cast in the nonparametric Bayesian framework by using Dirichlet process mixture to infer the necessary number of Gaussian components. We derive a mean-field variational inference algorithm for the proposed nonparametric Bayesian model. We first test the model on synthetic data sets contaminated by Gaussian, sparse and mixed noise. We then apply it to extract muscle synergies from the electromyographic (EMG) signal and to select discriminative features for motor imagery single-trial electroencephalogram (EEG) classification. Experimental results demonstrate that DPNMF performs better in extracting the latent nonnegative factors in comparison with state-of-the-art methods.

Li, M, Xu, RY, Xin, J, Zhang, K & Jing, J 2020, 'Fast non-rigid points registration with cluster correspondences projection', Signal Processing, vol. 170, pp. 107425-107425.
View/Download from: Publisher's site

Li, Y, Li, K, Wang, X & Xu, RYD 2020, 'Exploring temporal consistency for human pose estimation in videos', Pattern Recognition, vol. 103, pp. 107258-107258.
View/Download from: Publisher's site
View description>>

© 2020 In this paper, we introduce a method of exploring temporal information for estimating human poses in videos. The current state-of-the-art methods utilizing temporal information can be categorized into two major branches. The first category is a model-based method that captures the temporal information entirely by using a learnable function such as RNN or 3D convolution. However, these methods are limited in exploring temporal consistency, which is essential for estimating human joint positions in videos. The second category is the posterior enhancement method, where an independent post-processing step (e.g., using optical flow) is applied to enhance the prediction. However, operations such as optical flow estimation can be susceptible to the occlusion and motion blur problems, which will adversely affect the final performance. We propose a novel Temporal Consistency Exploration (TCE) module to address both shortcomings. Compared to previous approaches, the TCE module is more efficient as it captures the temporal consistency at the feature level without having to post-process and calculate extra optical flow. Further, to capture the rich spatial context in video data, we design a multi-scale TCE to explore the time consistency information at multi-scale spatial levels. Finally, a video-based pose estimation network is designed, which is based on the encoder-decoder architecture and extended with the powerful multi-scale TCE module. We comprehensively evaluate the proposed model on two video datasets, Sub-JHMDB and Penn, and our model achieves state-of-the-art performance on both datasets.

Lian, J-W, Ban, Y-L, Zhu, H & Guo, YJ 2020, 'Reduced-Sidelobe Multibeam Array Antenna Based on SIW Rotman Lens', IEEE Antennas and Wireless Propagation Letters, vol. 19, no. 1, pp. 188-192.
View/Download from: Publisher's site
View description>>

© 2002-2011 IEEE. A multibeam array antenna (MAA) fed by a Rotman lens with a reduced sidelobe level (SLL) is designed using a substrate integrate waveguide (SIW) technology. The designed MAA is composed of a Rotman lens and a 12 × 8 slot array, which functions as the beamforming network and the radiation part, respectively. To reduce the SLL in E-plane, dual-port excitations (DPEs) are applied, instead of single-port excitations (SPEs), at each feeding port of the Rotman lens. By using DPEs, a more tapered amplitude distribution can be obtained on the array elements as compared to using SPEs; therefore, the SLL is reduced from about -11 to -18 dB. The SLL in H-plane is controlled by introducing a Chebyshev distribution to the designed eight-element slot array. Based on the designed MAA, a fabricated prototype is measured to test the discrepancy between simulation and experiment.

Lian, J-W, Ban, Y-L, Zhu, H & Guo, YJ 2020, 'Uniplanar Beam-Forming Network Employing Eight-Port Hybrid Couplers and Crossovers for 2-D Multibeam Array Antennas', IEEE Transactions on Microwave Theory and Techniques, vol. 68, no. 11, pp. 4706-4718.
View/Download from: Publisher's site

Lian, J-W, Zhu, H, Ban, Y-L, Karmokar, DK & Guo, YJ 2020, 'Uniplanar High-Gain 2-D Scanning Leaky-Wave Multibeam Array Antenna at Fixed Frequency', IEEE Transactions on Antennas and Propagation, vol. 68, no. 7, pp. 5257-5268.
View/Download from: Publisher's site

Liang, Q, Wu, W, Yang, Y, Zhang, R, Peng, Y & Xu, M 2020, 'Multi-Player Tracking for Multi-View Sports Videos with Improved K-Shortest Path Algorithm', Applied Sciences, vol. 10, no. 3, pp. 864-864.
View/Download from: Publisher's site
View description>>

Sports analysis has recently attracted increasing research efforts in computer vision. Among them, basketball video analysis is very challenging due to severe occlusions and fast motions. As a typical tracking-by-detection method, k-shortest paths (KSP) tracking framework has been well used for multiple-person tracking. While effective and fast, the neglect of the appearance model would easily lead to identity switches, especially when two or more players are intertwined with each other. This paper addresses this problem by taking the appearance features into account based on the KSP framework. Furthermore, we also introduce a similarity measurement method that can fuse multiple appearance features together. In this paper, we select jersey color and jersey number as two example features. Experiments indicate that about 70% of jersey color and 50% of jersey number over a whole sequence would ensure our proposed method preserve the player identity better than the existing KSP tracking method.

Lim, WYB, Luong, NC, Hoang, DT, Jiao, Y, Liang, Y-C, Yang, Q, Niyato, D & Miao, C 2020, 'Federated Learning in Mobile Edge Networks: A Comprehensive Survey', IEEE Communications Surveys & Tutorials, vol. 22, no. 3, pp. 2031-2063.
View/Download from: Publisher's site
View description>>

© 1998-2012 IEEE. In recent years, mobile devices are equipped with increasingly advanced sensing and computing capabilities. Coupled with advancements in Deep Learning (DL), this opens up countless possibilities for meaningful applications, e.g., for medical purposes and in vehicular networks. Traditional cloud-based Machine Learning (ML) approaches require the data to be centralized in a cloud server or data center. However, this results in critical issues related to unacceptable latency and communication inefficiency. To this end, Mobile Edge Computing (MEC) has been proposed to bring intelligence closer to the edge, where data is produced. However, conventional enabling technologies for ML at mobile edge networks still require personal data to be shared with external parties, e.g., edge servers. Recently, in light of increasingly stringent data privacy legislations and growing privacy concerns, the concept of Federated Learning (FL) has been introduced. In FL, end devices use their local data to train an ML model required by the server. The end devices then send the model updates rather than raw data to the server for aggregation. FL can serve as an enabling technology in mobile edge networks since it enables the collaborative training of an ML model and also enables DL for mobile edge network optimization. However, in a large-scale and complex mobile edge network, heterogeneous devices with varying constraints are involved. This raises challenges of communication costs, resource allocation, and privacy and security in the implementation of FL at scale. In this survey, we begin with an introduction to the background and fundamentals of FL. Then, we highlight the aforementioned challenges of FL implementation and review existing solutions. Furthermore, we present the applications of FL for mobile edge network optimization. Finally, we discuss the important challenges and future research directions in FL.

Lin, B, Zhao, L, Suraweera, HA, Luan, TH, Niyato, D & Hoang, DT 2020, 'Guest Editorial Special Issue on Internet of Things for Smart Ocean', IEEE Internet of Things Journal, vol. 7, no. 10, pp. 9675-9677.
View/Download from: Publisher's site

Lin, Z, Lv, T, Ni, W, Zhang, JA & Liu, RP 2020, 'Tensor-Based Multi-Dimensional Wideband Channel Estimation for mmWave Hybrid Cylindrical Arrays', IEEE Transactions on Communications, vol. 68, no. 12, pp. 7608-7622.
View/Download from: Publisher's site
View description>>

© 1972-2012 IEEE. Channel estimation is challenging for hybrid millimeter wave (mmWave) large-scale antenna arrays which are promising in 5G/B5G applications. The challenges are associated with angular resolution losses resulting from hybrid front-ends, beam squinting, and susceptibility to the receiver noises. Based on tensor signal processing, this paper presents a novel multi-dimensional approach to channel parameter estimation with large-scale mmWave hybrid uniform circular cylindrical arrays (UCyAs) which are compact in size and immune to mutual coupling but known to suffer from infinite-dimensional array responses and intractability. We design a new resolution-preserving hybrid beamformer and a low-complexity beam squinting suppression method, and reveal the existence of shift-invariance relations in the tensor models of received array signals at the UCyA. Exploiting these relations, we propose a new tensor-based subspace estimation algorithm to suppress the receiver noises in all dimensions (time, frequency, and space). The algorithm can accurately estimate the channel parameters from both coherent and incoherent signals. Corroborated by the Cramér-Rao lower bound (CRLB), simulation results show that the proposed algorithm is able to achieve substantially higher estimation accuracy than existing matrix-based techniques, with a comparable computational complexity.

Liu, B, Ni, W, Liu, RP & Zhu, H 2020, 'Optimal Selection of Heterogeneous Network Interfaces for High-Speed Rail Communications', IEEE Transactions on Vehicular Technology, vol. 69, no. 12, pp. 15005-15018.
View/Download from: Publisher's site

Liu, D, Huang, Y, Wu, Q, Ma, R & An, P 2020, 'Multi-Angular Epipolar Geometry Based Light Field Angular Reconstruction Network', IEEE Transactions on Computational Imaging, vol. 6, pp. 1507-1522.
View/Download from: Publisher's site

Liu, F, Liu, Y, Han, F, Ban, Y-L & Jay Guo, Y 2020, 'Synthesis of Large Unequally Spaced Planar Arrays Utilizing Differential Evolution With New Encoding Mechanism and Cauchy Mutation', IEEE Transactions on Antennas and Propagation, vol. 68, no. 6, pp. 4406-4416.
View/Download from: Publisher's site
View description>>

© 1963-2012 IEEE. This article presents a differential evolution algorithm with a new encoding mechanism and Cauchy mutation (DE-NEM-CM) for optimizing large unequally spaced planar array layouts with the minimum element spacing constraint. In the new encoding mechanism, each individual represents a certain element position rather than an entire array layout used in traditional stochastic optimization algorithms. Such an encoding mechanism has the following advantages: 1) in each individual updating, the array pattern can be efficiently evaluated by only considering the radiation contribution variation from one element movement, which can greatly reduce the computational time; 2) it naturally facilitates the generated new array layout in population updating to meet the minimum element spacing constraint, and 3) each individual is searched always in 2-D space as the array size increases. These advantages enable it to be very suitable for synthesizing large arrays. Besides, DE serves as a search engine, and Cauchy mutation with chaotic mapping is proposed to enhance the local search while preserving the diversity of the population. A set of experiments for synthesizing different types of unequally spaced planar arrays in both narrow-and broadband applications are conducted. Synthesis results show that the proposed method achieves much lower sidelobe level than some state-of-the-art stochastic optimization methods for all the test cases. Importantly, the proposed method is much more efficient than conventional stochastic optimization algorithm especially for the case of synthesizing large unequally spaced planar array layouts. A array layout optimization with more than 1000 elements can be achieved within acceptable CPU time cost, which has not yet been reported for the existing stochastic optimization methods without resorting to supercomputing facilities.

Liu, Y, Chen, L, Zhu, C, Ban, Y-L & Guo, YJ 2020, 'Efficient and Accurate Frequency-Invariant Beam Pattern Synthesis Utilizing Iterative Spatiotemporal Fourier Transform', IEEE Transactions on Antennas and Propagation, vol. 68, no. 8, pp. 6069-6079.
View/Download from: Publisher's site
View description>>

© 1963-2012 IEEE. An iterative spatiotemporal Fourier transform (STFT) method is presented to efficiently design finite-impulse-response (FIR) filter coefficients for generating a desired frequency-invariant (FI) beam pattern of an antenna array. In this method, by introducing the concepts of normalized temporal angular frequency and spatial angular frequency, the broadband pattern is treated as a spatiotemporal spectral distribution. The relationship between the FIR coefficient distribution and the spatiotemporal spectral distribution can be built as an STFT, and consequently the 2-D fast Fourier transform (2D-FFT) and inverse 2D-FFT (2D-IFFT) can be utilized to efficiently accomplish the transformation between the FIR coefficient distribution and the spatiotemporal spectral distribution. Thus, the proposed synthesis method starts from an initial spatiotemporal spectral distribution and then adopt an iterative modification-and-transformation strategy to successively update the obtained spatiotemporal spectral distribution and the corresponding FIR coefficients. Two kinds of pattern modification techniques including the mainlobe FI modification and broadband sidelobe control are adopted in each iteration. Several examples for synthesizing different FI patterns are conducted. Synthesis results show that the proposed method can obtain much better FI pattern performance in terms of both mainlobe FI property and sidelobe control than the original FT method whilst costing less CPU time than the convex optimization method especially for the case of large FI arrays.

Liu, Y, Yang, Y, Han, F, Liu, QH & Guo, YJ 2020, 'Improved Beam-Scannable Ultra-Wideband Sparse Antenna Arrays by Iterative Convex Optimization Based on Raised Power Series Representation', IEEE Transactions on Antennas and Propagation, vol. 68, no. 7, pp. 5696-5701.
View/Download from: Publisher's site

Liu, Y, Zheng, J, Li, M, Luo, Q, Rui, Y & Guo, YJ 2020, 'Synthesizing Beam-Scannable Thinned Massive Antenna Array Utilizing Modified Iterative FFT for Millimeter-Wave Communication', IEEE Antennas and Wireless Propagation Letters, vol. 19, no. 11, pp. 1983-1987.
View/Download from: Publisher's site

Luo, C, Wong, S, Chen, R, Zhu, X, Yang, Y, Lin, J, Tu, Z & Xue, Q 2020, 'Compact on‐chip millimetre wave bandpass filters with meandered grounding resonator in 0.13‐μm (Bi)‐CMOS technology', IET Microwaves, Antennas & Propagation, vol. 14, no. 6, pp. 559-565.
View/Download from: Publisher's site
View description>>

© The Institution of Engineering and Technology 2019. In this study, an ultra-compact meandered grounding resonator is proposed to design two millimetre wave bandpass filters (BPFs) in a standard 0.13-μm silicon-germanium (Bi)-complementary metal oxide semiconductor (CMOS) technology. The fundamental second-order prototype, namely BPF-I, consists of a pair of proposed resonators and a pair of grounded metalinsulator- metal (MIM) capacitors. To better understand the principle of the second-order BPF-I, an equivalent LC-circuit model and theoretical analysis method are presented in this study. Based on BPF-I, the second-order BPF-II is proposed by adding the additional two pairs of MIM capacitors to improve the frequency selectivity, by means of introducing a transmission zero at lower stopband. Finally, both of the two second-order BPFs are fabricated. The measured results show a good agreement with the full-wave simulation results. The insertion loss of the first BPF-I is 1.79 dB at the centre frequency of 46.6 GHz, and the fractional bandwidth is up to 96.5%. The second BPF-II has a centre frequency at 46.8 GHz with a fractional bandwidth of 94.1%. The minimum insertion loss is 2.08 dB and the lower stopband attenuation is up to 42.7 dB. Moreover, the die sizes of the two compact BPFs, excluding the test pads, are only 0.0197 mm2 (0.104 × 0.190 mm2).

Luo, Y, Zhang, JA, Huang, X, Ni, W & Pan, J 2020, 'Multibeam Optimization for Joint Communication and Radio Sensing Using Analog Antenna Arrays', IEEE Transactions on Vehicular Technology, vol. PP, no. 99, pp. 1-1.
View/Download from: Publisher's site
View description>>

Multibeam technology enables the use of two or more subbeams for jointcommunication and radio sensing, to meet different requirements of beamwidthand pointing directions. Generating and optimizing multibeam subject to therequirements is critical and challenging, particularly for systems using analogarrays. This paper develops optimal solutions to a range of multibeam designproblems, where both communication and sensing are considered. We first studythe optimal combination of two pre-generated subbeams, and their beamformingvectors, using a combining phase coefficient. Closed-form optimal solutions arederived to the constrained optimization problems, where the received signalpowers for communication and the beamforming waveforms are alternatively usedas the objective and constraint functions. We also develop global optimizationmethods which directly find optimal solutions for a single beamforming vector.By converting the original intractable complex NP-hard global optimizationproblems to real quadratically constrained quadratic programs, near-optimalsolutions are obtained using semidefinite relaxation techniques. Extensivesimulations validate the effectiveness of the proposed constrained multibeamgeneration and optimization methods.

Lyu, B & Hoang, DT 2020, 'Optimal Time Scheduling in Relay Assisted Batteryless IoT Networks', IEEE Wireless Communications Letters, vol. 9, no. 5, pp. 706-710.
View/Download from: Publisher's site
View description>>

© 2012 IEEE. In this letter, we propose a novel relay transmission scheme in a batteryless IoT network for practical implementation and high energy-efficiency, where communications between a hybrid access point (HAP) and multiple batteryless sensors are assisted by energy-constrained gateways. In the proposed system, while a batteryless sensor backscatters the incident signals from the HAP to transmit data to its gateway, other gateways can simultaneously harvest energy from the HAP. Then, the gateways can use their harvested energy to forward the received signals to the HAP. Under this setup, we formulate the achievable sum-rate maximization problem by optimizing the time allocation between data backscattering, energy harvesting, and data forwarding. Then, an efficient method is proposed to find the optimal solution. Simulation results show that the proposed relay transmission scheme can achieve up to 34% sum-rate gain over two benchmark schemes.

Lyu, B, Hoang, DT & Yang, Z 2020, 'Backscatter Then Forward: A Relaying Scheme for Batteryless IoT Networks', IEEE Wireless Communications Letters, vol. 9, no. 4, pp. 562-566.
View/Download from: Publisher's site
View description>>

IEEE In this paper, we introduce a novel relaying scheme together with a joint energy beamforming (EB) and time allocation optimization to meet requirements about energy efficiency and hardware constraints of batteryless IoT networks. First, we propose an intelligent relaying scheme using RF-powered gateways as relay nodes to deliver information from batteryless IoT devices to a hybrid access point (HAP). The HAP can also transfer energy to the gateways and batteryless devices using EB techniques. The energy from HAP will be then used to supply power for gateways and as a communications means to transmit data for batteryless devices. We then formulate a sum-rate maximization problem by jointly optimizing the EB vectors, time scheduling, and power allocation. Since the optimization problem is non-convex, we exploit EB characteristics for data backscattering and employ variable substitutions and semidefinite relaxation techniques to transform it into a convex one. After that, a low-complexity method is proposed to obtain the optimal solution in a closed-form. Simulation results confirm that the proposed scheme can achieve significant sum-rate gain.

Lyu, B, Hoang, DT, Gong, S, Niyato, D & Kim, DI 2020, 'IRS-Based Wireless Jamming Attacks: When Jammers Can Attack Without Power', IEEE Wireless Communications Letters, vol. 9, no. 10, pp. 1663-1667.
View/Download from: Publisher's site
View description>>

© 2012 IEEE. This letter proposes to use Intelligent Reflecting Surface (IRS) as a green jammer to attack a legitimate communication without using any internal energy to generate jamming signals. In particular, the IRS is used to intelligently reflect the signals from the legitimate transmitter to the legitimate receiver (LR) to guarantee that the received signals from direct and reflecting links can be added destructively, which thus diminishes the Signal-to-Interference-plus-Noise Ratio (SINR) at the LR. To minimize the received signal power at the LR, we consider the joint optimization of magnitudes of reflection coefficients and discrete phase shifts at the IRS. Based on the block coordinate descent, semidefinite relaxation, and Gaussian randomization techniques, the solution can be obtained efficiently. Through simulation results, we show that by using the IRS-based jammer, we can reduce the signal power received at the LR by up to 99%. Interestingly, the performance of the proposed IRS-based jammer is even better than that of the conventional active jamming attacks in some scenarios.

Lyu, X, Ren, C, Ni, W, Tian, H & Liu, RP 2020, 'Cooperative Computing Anytime, Anywhere: Ubiquitous Fog Services', IEEE Wireless Communications, vol. 27, no. 1, pp. 162-169.
View/Download from: Publisher's site

Ma, R, Li, T, Bo, D, Wu, Q & An, P 2020, 'Error sensitivity model based on spatial and temporal features', Multimedia Tools and Applications, vol. 79, no. 43-44, pp. 31913-31930.
View/Download from: Publisher's site
View description>>

© 2020, Springer Science+Business Media, LLC, part of Springer Nature. Packet loss and error propagation induced by it are significant causes of visual impairments in video applications. Most of the existing video quality assessment models are developed at frame or sequence level, which can not accurately describe the impact of packet loss on the local regions in one frame. In this paper, we propose an error sensitivity model to evaluate the impact of a single packet loss. We also make full use of the spatio-temporal correlation of the video and analyze a set of features that directly impact the perceptual quality of videos, based on the specific situation of video packet loss. With the aid of the support vector regression (SVR), these features are used to predict the error sensitivity of the local region. The proposed model is tested on six video sequences. Experimental results show that the proposed model predicts sensitivity of videos to different packet loss cases with certain reasonable accuracy, and provides good generalization ability, which turns out outperform the state-of-art image and video quality assessment methods.

Makhdoom, I, Zhou, I, Abolhasan, M, Lipman, J & Ni, W 2020, 'PrivySharing: A blockchain-based framework for privacy-preserving and secure data sharing in smart cities', Computers & Security, vol. 88, pp. 101653-101653.
View/Download from: Publisher's site
View description>>

© 2019 Elsevier Ltd The ubiquitous use of Internet of Things (IoT) ranges from industrial control systems to e-Health, e-commerce, smart cities, agriculture, 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 privacy-preserving and secure IoT data sharing in a smart city environment. The proposed scheme is distinct from existing strategies on many aspects. The data privacy is preserved by dividing the blockchain network into various channels, where every channel comprises a finite number of authorized organizations and processes a specific type of data such as health, smart car, smart energy or financial details. Moreover, access to users’ data within a channel is controlled by embedding access control rules in the smart contracts. In addition, data within a channel is further isolated and secured by using private data collection and encryption respectively. 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 conforms to some of the significant requirements outlined in the European Union General Data Protection Regulation. We also present a system of reward in the form of a digital token named “PrivyCoin” for users sharing their data with stakeholders/third parties. Lastly, the experimental outcomes advocate that a multi-channel blockchain scales well as compared to a single-channel blockchain system.

Malik, N, Nanda, P, He, X & Liu, RP 2020, 'Vehicular networks with security and trust management solutions: proposed secured message exchange via blockchain technology', Wireless Networks, vol. 26, no. 6, pp. 4207-4226.
View/Download from: Publisher's site
View description>>

© 2020, Springer Science+Business Media, LLC, part of Springer Nature. In vehicular ad hoc networks (VANET), effective trust establishment with authentication is an important requirement. Trust management among communicating vehicles is significant for secure message transmission; however, very less contributions have been made towards evaluating the trustworthiness of the node. This research work intends to introduce a new trust management system in VANET with two major phases: secured message transmission and node trustability prediction. The security assured message passing is carried out by incorporating the privacy preservation model under the data sanitization process. The key used for the sanitization process is optimally tuned by a new hybrid algorithm termed Sea Lion Explored-Whale Optimization Algorithm, which is the combination of Whale Optimization Algorithm and Sea Lion Optimization Algorithm, respectively. The blockchain technology is assisted to handle the key generated by the nodes. Subsequently, the trustability of the node is evaluated under novel specifics “two-level evaluation process” with a rule-based and machine learning-based evaluation process. Finally, the performance of the proposed model is verified and proved over other conventional methods for certain measures.

Nan, Y, Huang, X & Guo, YJ 2020, 'A Millimeter-Wave GCW-SAR Based on Deramp-on-Receive and Piecewise Constant Doppler Imaging', IEEE Transactions on Geoscience and Remote Sensing, vol. 58, no. 1, pp. 680-690.
View/Download from: Publisher's site
View description>>

© 2019 IEEE. A novel generalized continuous-wave synthetic aperture radar (GCW-SAR) based on deramp-on-receive operating in millimeter-wave frequency is proposed in this article. With deramp-on-receive, the receiver sampling rate is drastically reduced, and the downsampled 1-D raw data can be obtained from the received beat signal. Further adopting piecewise constant Doppler (PCD) imaging in the digital domain, a GCW-SAR image can be easily reconstructed by using the existing frequency-modulated continuous-wave (FMCW) radar system. The effects of deramp-on-receive in PCD imaging are analyzed accordingly. The short wavelength of the millimeter-wave carrier used in the proposed GCW-SAR enables high azimuth resolution as well as a short synthetic aperture, which, in turn, significantly reduces the imaging computational complexity. Simulation and experimental results confirm the advantages of the proposed GCW-SAR.

Nan, Y, Huang, X & Guo, YJ 2020, 'Piecewise Constant Doppler Algorithm: Performance Analysis, Further Simplification, and Motion Compensation', IEEE Transactions on Aerospace and Electronic Systems, vol. 56, no. 5, pp. 3613-3631.
View/Download from: Publisher's site
View description>>

IEEE The piecewise constant Doppler (PCD) algorithm is a novel radar imaging process recently proposed for the generalized continuous wave synthetic aperture radar (GCW-SAR). This paper presents a detailed theoretical analysis on the PCD algorithm's performance and proposes a further complexityreduced PCD algorithm with motion compensation (MOCO) suitable for practical applications. Firstly, the difference between conventional SAR imaging and PCD imaging, i.e., the zeroth order versus the first order slant range approximation, is revealed. Exact ambiguity function expressions of the PCD imaging in range and azimuth directions respectively are then derived. An error function of the PCD imaging as compared with the ideal matched filtering method is further defined and shown to be a function of an image quality factor which can be used to quantify the PCD imaging performance. Finally, a faster and more flexible imaging process, called decimated PCD algorithm, is proposed, by which the image azimuth spacing can be easily extended and hence the computational complexity can be significantly reduced. The decimated PCD implementation incorporated with the MOCO is developed for practical GCWSAR applications and its imaging error lower-bounded by the PCD imaging error function is analyzed accordingly. Simulation and experimental results validate the theoretical analysis of the PCD imaging and show that the decimated PCD algorithm can achieve a high imaging quality at low cost.

Nguyen, CT, Saputra, YM, Huynh, NV, Nguyen, N-T, Khoa, TV, Tuan, BM, Nguyen, DN, Hoang, DT, Vu, TX, Dutkiewicz, E, Chatzinotas, S & Ottersten, B 2020, 'A Comprehensive Survey of Enabling and Emerging Technologies for Social Distancing—Part I: Fundamentals and Enabling Technologies', IEEE Access, vol. 8, pp. 153479-153507.
View/Download from: Publisher's site
View description>>

Social distancing plays a pivotal role in preventing the spread of viral diseases illnesses such as COVID-19. By minimizing the close physical contact among people, we can reduce the chances of catching the virus and spreading it across the community. This two-part paper aims to provide a comprehensive survey on how emerging technologies, e.g., wireless and networking, artificial intelligence (AI) can enable, encourage, and even enforce social distancing practice. In this Part I, we provide a comprehensive background of social distancing including basic concepts, measurements, models, and propose various practical social distancing scenarios. We then discuss enabling wireless technologies which are especially effect- in social distancing, e.g., symptom prediction, detection and monitoring quarantined people, and contact tracing. The companion paper Part II surveys other emerging and related technologies, such as machine learning, computer vision, thermal, ultrasound, etc., and discusses open issues and challenges (e.g., privacy-preserving, scheduling, and incentive mechanisms) in implementing social distancing in practice.

Nguyen, CT, Saputra, YM, Huynh, NV, Nguyen, N-T, Khoa, TV, Tuan, BM, Nguyen, DN, Hoang, DT, Vu, TX, Dutkiewicz, E, Chatzinotas, S & Ottersten, B 2020, 'Enabling and Emerging Technologies for Social Distancing: A Comprehensive Survey and Open Problems', IEEE Access, vol. 8, pp. 153479-153507.
View/Download from: Publisher's site
View description>>

Social distancing plays a pivotal role in preventing the spread of viraldiseases illnesses such as COVID-19. By minimizing the close physical contactamong people, we can reduce the chances of catching the virus and spreading itacross the community. This paper aims to provide a comprehensive survey on howemerging technologies, e.g., wireless and networking, artificial intelligence(AI) can enable, encourage, and even enforce social distancing practice. Tothat end, we first provide a comprehensive background of social distancingincluding basic concepts, measurements, models, and propose various practicalsocial distancing scenarios. We then discuss enabling wireless technologieswhich are especially effective and can be widely adopted in practice to keepdistance, encourage, and enforce social distancing in general. After that,other emerging and related technologies such as machine learning, computervision, thermal, ultrasound, etc., are introduced. These technologies open manynew solutions and directions to deal with problems in social distancing, e.g.,symptom prediction, detection and monitoring quarantined people, and contacttracing. Finally, we provide important open issues and challenges (e.g.,privacy-preserving, scheduling, and incentive mechanisms) in implementingsocial distancing in practice. As an example, instead of reacting with ad-hocresponses to COVID-19-like pandemics in the future, smart infrastructures(e.g., next-generation wireless systems like 6G, smart home/building, smartcity, intelligent transportation systems) should incorporate a pandemic mode inits standard architecture/design.

Nguyen, CT, Saputra, YM, Van Huynh, N, Nguyen, N-T, Khoa, TV, Tuan, BM, Nguyen, DN, Hoang, DT, Vu, TX, Dutkiewicz, E, Chatzinotas, S & Ottersten, B 2020, 'A Comprehensive Survey of Enabling and Emerging Technologies for Social Distancing—Part II: Emerging Technologies and Open Issues', IEEE Access, vol. 8, pp. 154209-154236.
View/Download from: Publisher's site
View description>>

This two-part paper aims to provide a comprehensive survey on how emerging technologies, e.g., wireless and networking, artificial intelligence (AI) can enable, encourage, and even enforce social distancing practice. In Part I, an extensive background of social distancing is provided, and enabling wireless technologies are thoroughly surveyed. In this Part II, emerging technologies such as machine learning, computer vision, thermal, ultrasound, etc., are introduced. These technologies open many new solutions and directions to deal with problems in social distancing, e.g., symptom prediction, detection and monitoring quarantined people, and contact tracing. Finally, we discuss open issues and challenges (e.g., privacy-preserving, scheduling, and incentive mechanisms) in implementing social distancing in practice. As an example, instead of reacting with ad-hoc responses to COVID-19-like pandemics in the future, smart infrastructures (e.g., next-generation wireless systems like 6G, smart home/building, smart city, intelligent transportation systems) should incorporate a pandemic mode in their standard architectures/designs

Nguyen, MH, Hà, MH, Nguyen, DN & Tran, TT 2020, 'Solving the k-dominating set problem on very large-scale networks', Computational Social Networks, vol. 7, no. 1.
View/Download from: Publisher's site
View description>>

AbstractThe well-known minimum dominating set problem (MDSP) aims to construct the minimum-size subset of vertices in a graph such that every other vertex has at least one neighbor in the subset. In this article, we study a general version of the problem that extends the neighborhood relationship: two vertices are called neighbors of each other if there exists a path through no more thankedges between them. The problem called “minimumk-dominating set problem” (MkDSP) becomes the classical dominating set problem ifkis 1 and has important applications in monitoring large-scale social networks. We propose an efficient heuristic algorithm that can handle real-world instances with up to 17 million vertices and 33 million edges. This is the first time such large graphs are solved for the minimumk-dominating set problem.

Pham, M, Hoang, DB & Chaczko, Z 2020, 'Congestion-Aware and Energy-Aware Virtual Network Embedding', IEEE/ACM Transactions on Networking, vol. 28, no. 1, pp. 210-223.
View/Download from: Publisher's site

Qin, C, Zhang, JA, Huang, X & Guo, YJ 2020, 'Virtual-Subarray-Based Angle-of-Arrival Estimation in Analog Antenna Arrays', IEEE Wireless Communications Letters, vol. 9, no. 2, pp. 194-197.
View/Download from: Publisher's site
View description>>

© 2012 IEEE. Angle-of-arrival (AoA) estimation is a challenging problem for analog antenna arrays. Typical algorithms are based on beam scanning which can be time-consuming. In this letter, we propose a virtual-subarray based recursive AoA estimation scheme that can get an AoA estimate from every two measurements and recursively improve the performance by updating beamforming weights with soft probability-based information. Simulation results validate the high efficiency of the proposed scheme, demonstrating its superiority for initial AoA estimation in analog arrays.

Qin, C, Zhang, JA, Huang, X, Wu, K & Guo, YJ 2020, 'Fast Angle-of-Arrival Estimation via Virtual Subarrays in Analog Antenna Array', IEEE Transactions on Wireless Communications, vol. 19, no. 10, pp. 6425-6439.
View/Download from: Publisher's site

Rahman, ML, Zhang, JA, Huang, X, Guo, YJ & Heath, RW 2020, 'Framework for a Perceptive Mobile Network Using Joint Communication and Radar Sensing', IEEE Transactions on Aerospace and Electronic Systems, vol. 56, no. 3, pp. 1926-1941.
View/Download from: Publisher's site
View description>>

In this paper, we develop a framework for a novel perceptive mobile/cellular network that integrates radar sensing function into the mobile communication network. We propose a unified system platform that enables downlink and uplink sensing, sharing the same transmitted signals with communications. We aim to tackle the fundamental sensing parameter estimation problem in perceptive mobile networks, by addressing two key challenges associated with sophisticated mobile signals and rich multipath in mobile networks. To extract sensing parameters from orthogonal frequency division multiple access and spatial division multiple access communication signals, we propose two approaches to formulate it to problems that can be solved by compressive sensing techniques. Most sensing algorithms have limits on the number of multipath signals for their inputs. To reduce the multipath signals, as well as removing unwanted clutter signals, we propose a background subtraction method based on simple recursive computation, and provide a closed-form expression for performance characterization. The effectiveness of these methods is validated in simulations.

Rahman, ML, Zhang, JA, Huang, X, Guo, YJ & Lu, Z 2020, 'Joint communication and radar sensing in 5G mobile network by compressive sensing', IET Communications, vol. 14, no. 22, pp. 3977-3988.
View/Download from: Publisher's site

Ren, C, Lyu, X, Ni, W, Tian, H, Song, W & Liu, RP 2020, 'Distributed Online Optimization of Fog Computing for Internet of Things Under Finite Device Buffers', IEEE Internet of Things Journal, vol. 7, no. 6, pp. 5434-5448.
View/Download from: Publisher's site

Saki, M, Abolhasan, M & Lipman, J 2020, 'A Novel Approach for Big Data Classification and Transportation in Rail Networks', IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 3, pp. 1239-1249.
View/Download from: Publisher's site
View description>>

This paper introduces a new framework into future data-driven railway condition monitoring systems (RCM). For this purpose, we have proposed an edge processing unit that includes two main parts: a data classification model that classifies Internet of Things (IoT) data into maintenance-critical data (MCD) and maintenance-non-critical data (MNCD) and a data transmission unit that, based on the class of data, employs appropriate communication methods to transmit data to railway control centers. For the transmission of MNCD, we propose a travel pattern method that employs train stations as points of data offloading so that trains can deliver data as well as passengers at stations. The performance of our proposed solution is successfully validated via three various data sets under different operating conditions.

Saki, M, Abolhasan, M, Lipman, J & Jamalipour, A 2020, 'A Comprehensive Access Point Placement for IoT Data Transmission Through Train-Wayside Communications in Multi-Environment Based Rail Networks', IEEE Transactions on Vehicular Technology, vol. 69, no. 10, pp. 11937-11949.
View/Download from: Publisher's site
View description>>

In this paper, we propose three algorithms for placement of access points (APs) for the purpose of data transportation via train-to-wayside (T2W) communications along a rail network. The first algorithm is proposed to find the minimum number of APs so that the path-loss (PL) does not exceed a desired threshold. Through the second algorithm, the most optimal places for a desired number of APs are determined so that the average PL is minimum. The goal of the third algorithm is to determine the required number and optimal places of APs in a rail network. Furthermore, we propose a model to consider the effects of changes of communication characteristics on the efficiency of the network in different environments. Through such model, the algorithms proposed for placement of APs can be used in different railway scenarios. The proposed algorithms are validated through extensive simulations in Sydney Trains of Australia. The simulation results show that the proposed approach can improve the efficiency of the system at least 21% and up to 165% within 10 different scenarios. We also show that we can approximately transmit over 250 Gigabit data through T2W communications over common WiFi networks.

Saputra, YM, Nguyen, DN, Hoang, DT, Vu, TX, Dutkiewicz, E & Chatzinotas, S 2020, 'Federated Learning Meets Contract Theory: Energy-Efficient Framework for Electric Vehicle Networks', IEEE Transactions on Mobile Computing, pp. 1-1.
View/Download from: Publisher's site
View description>>

In this paper, we propose a novel energy-efficient framework for an electricvehicle (EV) network using a contract theoretic-based economic model tomaximize the profits of charging stations (CSs) and improve the social welfareof the network. Specifically, we first introduce CS-based and CSclustering-based decentralized federated energy learning (DFEL) approacheswhich enable the CSs to train their own energy transactions locally to predictenergy demands. In this way, each CS can exchange its learned model with otherCSs to improve prediction accuracy without revealing actual datasets and reducecommunication overhead among the CSs. Based on the energy demand prediction, wethen design a multi-principal one-agent (MPOA) contract-based method. Inparticular, we formulate the CSs' utility maximization as a non-collaborativeenergy contract problem in which each CS maximizes its utility under commonconstraints from the smart grid provider (SGP) and other CSs' contracts. Then,we prove the existence of an equilibrium contract solution for all the CSs anddevelop an iterative algorithm at the SGP to find the equilibrium. Throughsimulation results using the dataset of CSs' transactions in Dundee city, theUnited Kingdom between 2017 and 2018, we demonstrate that our proposed methodcan achieve the energy demand prediction accuracy improvement up to 24.63% andlessen communication overhead by 96.3% compared with other machine learningalgorithms. Furthermore, our proposed method can outperform non-contract-basedeconomic models by 35% and 36% in terms of the CSs' utilities and socialwelfare of the network, respectively.

Shaukat, K, Luo, S, Varadharajan, V, Hameed, IA & Xu, M 2020, 'A Survey on Machine Learning Techniques for Cyber Security in the Last Decade', IEEE Access, vol. 8, pp. 222310-222354.
View/Download from: Publisher's site
View description>>

Pervasive growth and usage of the Internet and mobile applications have expanded cyberspace. The cyberspace has become more vulnerable to automated and prolonged cyberattacks. Cyber security techniques provide enhancements in security measures to detect and react against cyberattacks. The previously used security systems are no longer sufficient because cybercriminals are smart enough to evade conventional security systems. Conventional security systems lack efficiency in detecting previously unseen and polymorphic security attacks. Machine learning (ML) techniques are playing a vital role in numerous applications of cyber security. However, despite the ongoing success, there are significant challenges in ensuring the trustworthiness of ML systems. There are incentivized malicious adversaries present in the cyberspace that are willing to game and exploit such ML vulnerabilities. This paper aims to provide a comprehensive overview of the challenges that ML techniques face in protecting cyberspace against attacks, by presenting a literature on ML techniques for cyber security including intrusion detection, spam detection, and malware detection on computer networks and mobile networks in the last decade. It also provides brief descriptions of each ML method, frequently used security datasets, essential ML tools, and evaluation metrics to evaluate a classification model. It finally discusses the challenges of using ML techniques in cyber security. This paper provides the latest extensive bibliography and the current trends of ML in cyber security.

Shi, Z, Pan, Q & Xu, M 2020, 'LSTM-Cubic A*-based auxiliary decision support system in air traffic management', Neurocomputing, vol. 391, pp. 167-176.
View/Download from: Publisher's site

Soltanieh, N, Norouzi, Y, Yang, Y & Karmakar, NC 2020, 'A Review of Radio Frequency Fingerprinting Techniques', IEEE Journal of Radio Frequency Identification, vol. 4, no. 3, pp. 222-233.
View/Download from: Publisher's site
View description>>

Radio frequency (RF) fingerprinting techniques have been used as an extra security layer for wireless devices. Unique fingerprints are used to identify wireless devices in order to avoid spoofing or impersonating attacks. These unique features can be extracted from imperfections of analog components during the manufacturing. This paper presents a general review of recent progress on RF fingerprinting techniques. Several studies are investigated for RF fingerprinting using different parts of a signal. The majority of these studies have been focused on the transient part of the signal. For this purpose, the transient signal must be extracted precisely. A number of common techniques of transient extraction are theoretically analyzed in this review. Then, some other approaches using the modulated part of the signal are also discussed. For all these approaches, the applied methodologies, the classification algorithms and a taxonomy of features are described. A comprehensive overview of the methods in RF fingerprinting is presented to demonstrate the state-of-the-art works.

Song, B, Wang, X, Ni, W, Song, Y, Liu, RP, Jiang, G-P & Guo, YJ 2020, 'Reliability Analysis of Large-Scale Adaptive Weighted Networks', IEEE Transactions on Information Forensics and Security, vol. 15, pp. 651-665.
View/Download from: Publisher's site
View description>>

© 2005-2012 IEEE. Disconnecting impaired or suspicious nodes and rewiring to those reliable, adaptive networks have the potential to inhibit cascading failures, such as DDoS attack and computer virus. The weights of disconnected links, indicating the workload of the links, can be transferred or redistributed to newly connected links to maintain network operations. Distinctively different from existing studies focused on adaptive unweighted networks, this paper presents a new mean-field model to analyze the reliability of adaptive weighted networks against cascading failures. By taking mean-field approximation, we develop a new continuous-time Markov model to capture the propagations of cascading failures and the rewiring actions that individual nodes can take to bypass failed neighbors. We analyze the stability of the model to identify the critical conditions, under which the cascading failures can be eventually inhibited or would proliferate. The conditions are evaluated under different link weight distributions and rewiring strategies. Our model reveals that preferentially disconnecting suspicious peers with high weights can effectively inhibit virus and failures.

Srinivas, S, Gill, AQ & Roach, T 2020, 'Analytics-Enabled Adaptive Business Architecture Modeling.', Complex Syst. Informatics Model. Q., vol. 23, no. 23, pp. 23-43.
View/Download from: Publisher's site

Suarez-Rodriguez, C, Haider, N, He, Y & Dutkiewicz, E 2020, 'Network Optimisation in 5G Networks: A Radio Environment Map Approach.', IEEE Trans. Veh. Technol., vol. 69, no. 10, pp. 12043-12057.
View/Download from: Publisher's site

Sun, F, Gómez‐García, R, Zhu, X, Zhu, H, Yang, Y & Tong, X 2020, 'Miniaturised millimetre‐wave BPF with broad stopband suppression in silicon–germanium technology', IET Microwaves, Antennas & Propagation, vol. 14, no. 4, pp. 308-313.
View/Download from: Publisher's site
View description>>

© 2020 Institution of Engineering and Technology. All rights reserved. On-chip passive distributed-element-based bandpass filters (BPFs) usually provide a decent stopband suppression across a limited bandwidth. To solve this drawback without adversely affecting other performance metrics, a simple but effective miniaturised BPF design approach is presented in this work. The proposed integrated BPF topology uses a combination of a coupled-inductor structure with a pair of metal-insulator-metal capacitors in a quasi-lumped-element realisation. To show the operational principles of this BPF approach, a simplified inductor-capacitor-equivalent circuit model is used for its theoretical analysis. From this analytical framework as an initial design guideline, a quasi-millimetre-wave BPF is designed and implemented in a standard 0.13 μmu;mu;m bipolar complementary-metal-oxide semiconductor technology. The measured results show that the developed BPF device has a centre frequency of 28 GHz with a 3 dB fractional bandwidth of 21% and minimum in-band power-insertion-loss level of 3.4 dB. The stopband suppression is higher than 25 dB beyond 45 GHz. The chip size, excluding the pads, is only 0.017 mm2 (0.06 × 0.284 mm2).

Sun, H-H, Ding, C, Zhu, H & Guo, YJ 2020, 'Dual-Polarized Multi-Resonance Antennas With Broad Bandwidths and Compact Sizes for Base Station Applications', IEEE Open Journal of Antennas and Propagation, vol. 1, pp. 11-19.
View/Download from: Publisher's site
View description>>

In this paper, a novel design method for dual-polarized multi-resonance antennas is presented for base station applications. The radiator of the antenna is configured as cross-dipoles with four thin metal strips connected to the adjacent dipole arms. The attached strips create multiple current paths and introduce additional resonant points. As a result, the bandwidth of the antennas is broadened while maintaining a very compact size. Based on this working mechanism, two multi-resonance antennas are designed, fabricated, and tested. The antennas achieve bandwidths of 46.7% and 66.7% respectively, with excellent matching capabilities. The antennas also exhibit high port isolation levels and stable radiation performances. The promising wideband performances with compact physical sizes make the antennas highly suitable for the base station applications.

Sun, H-H, Jones, B, Guo, YJ & Lee, YH 2020, 'Suppression of Cross-Band Scattering in Interleaved Dual-Band Cellular Base-Station Antenna Arrays', IEEE Access, vol. 8, pp. 222486-222495.
View/Download from: Publisher's site

Sun, H-H, Zhu, H, Ding, C, Jones, B & Guo, YJ 2020, 'Scattering Suppression in a 4G and 5G Base Station Antenna Array Using Spiral Chokes', IEEE Antennas and Wireless Propagation Letters, vol. 19, no. 10, pp. 1818-1822.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. This letter presents a novel distributed choking technique, the spiral choke, for scattering suppression in dual-band antenna arrays. The working principle and the scattering suppression capability of the choke are analyzed. The spiral chokes are implemented as low-band radiators in a colocated 4G and 5G dual-band array to suppress cross-band scattering while broadening the bandwidth of the choked element. The experimental results demonstrate that the cross-band scattering in the array is largely eliminated, and the realized dual-band array has very stable radiation performance in both well-matched bands.

Sutton, GJ, Liu, RP & Guo, YJ 2020, 'Coexistence Performance and Limits of Frame-Based Listen-Before-Talk', IEEE Transactions on Mobile Computing, vol. 19, no. 5, pp. 1084-1095.
View/Download from: Publisher's site

Takalkar, MA, Xu, M & Chaczko, Z 2020, 'Manifold feature integration for micro-expression recognition', Multimedia Systems, vol. 26, no. 5, pp. 535-551.
View/Download from: Publisher's site

Tang, Q, Yang, J, Jia, W, He, X, Zhang, Q & Liu, H 2020, 'A GMS-Guided Approach for 2D Feature Correspondence Selection', IEEE Access, vol. 8, pp. 36919-36929.
View/Download from: Publisher's site
View description>>

© 2013 IEEE. Feature correspondence selection, which aims to seek as many true matches (i.e., inliers) as possible from a given putative set while minimizing false matches (i.e., outliers), is crucial to many feature-matching based tasks in computer vision. It remains a challenging problem how to deal with putative sets with low inlier ratios. To address this problem, in this paper, we propose a novel correspondence selection strategy, which is guided by Grid-based Motion Statistics (GMS). We first adopt the GMS to generate a small correspondence set with a high inlier ratio. Then, an accurate geometric model is built using the above correspondence set. Finally, the built geometric model is used to filter the given putative correspondence set to obtain true correspondences. The experimental results on benchmark datasets demonstrate that our proposed approach outperforms the state-of-the-art approaches for putative sets with various inlier ratios, especially for cases with low inlier ratios.

Tofigh, F, Amiri, M, Shariati, N, Lipman, J & Abolhasan, M 2020, 'Crowd Estimation Using Electromagnetic Wave Power-Level Measurements: A Proof of Concept', IEEE Transactions on Vehicular Technology, vol. 69, no. 1, pp. 784-792.
View/Download from: Publisher's site
View description>>

© 1967-2012 IEEE. Current crowd density estimation technologies that leverage IR depth perception, video and image processing or WiFi/BLE-based sniffing and probing have privacy and deployment issues. This paper presents a novel method for non-intrusive crowd density estimation that monitors variation in EM radiation within an environment. The human body's electrical and magnetic characteristics can be correlated with variations in available EM energy. This allows for the determination of the number of people within a room. Simulations conducted using Comsol to analyse and measure electromagnetic energy levels inside a room containing human bodies. Experimental analysis provides validation of the simulation results by showing $\text{0.8}\;\text{dBm}$ drop on the average level of EM energy per person.

Tofigh, F, Amiri, M, Shariati, N, Lipman, J & Abolhasan, M 2020, 'Polarization-Insensitive Metamaterial Absorber for Crowd Estimation Based on Electromagnetic Energy Measurements', IEEE Transactions on Antennas and Propagation, vol. 68, no. 3, pp. 1458-1467.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Noninvasive crowd estimation has remained a challenging issue among researchers. Methods such as image analysis and Wi-Fi/Bluetooth probing can always be used to identify and track people. Lately, authors have introduced a noninvasive method for crowd estimation based on ambient RF energy measurements. In this article, a polarization-insensitive multilayer metamaterial absorber is introduced to measure the variation in the available RF energy levels for crowd estimation purposes. The proposed dual-band absorber is designed to absorb and transfer the maximum of the available Wi-Fi energy to a lumped element to enable proper and accurate measurements. To evaluate the design, the proposed structure is fabricated as an array, and its performance is tested, proving perfect absorption at the desired frequencies, 2.4 and 5 GHz.

Usman, M, Jan, MA, Jolfaei, A, Xu, M, He, X & Chen, J 2020, 'A Distributed and Anonymous Data Collection Framework Based on Multilevel Edge Computing Architecture', IEEE Transactions on Industrial Informatics, vol. 16, no. 9, pp. 6114-6123.
View/Download from: Publisher's site

Van Huynh, N, Nguyen, DN, Thai Hoang, D, Dutkiewicz, E & Mueck, M 2020, 'Ambient Backscatter: A Novel Method to Defend Jamming Attacks for Wireless Networks', IEEE Wireless Communications Letters, vol. 9, no. 2, pp. 175-178.
View/Download from: Publisher's site
View description>>

© 2012 IEEE. This letter introduces a novel idea to defend jamming attacks for wireless communications. In particular, when the jammer attacks the channel, the transmitter can leverage the jamming signals to transmit data by using ambient backscatter technique or harvest energy from the jamming signals to support its operation. To deal with the uncertainty of the jammer, we propose a reinforcement learning-based algorithm that allows the transmitter to obtain the optimal operation policy through real-time interaction processes with the attacker. The simulation results show the effectiveness of ambient backscatter in combating jammers, i.e., it enables the transmitter to transmit data even under the jamming attacks. We observe that the more power the jammer uses to attack the channel, the better performance the network can achieve.

Vu, HD, Nguyen, TV, Nguyen, DN & Nguyen, HT 2020, 'On Design of Protograph LDPC Codes for Large-Scale MIMO Systems', IEEE Access, vol. 8, pp. 46017-46029.
View/Download from: Publisher's site

Vu, L, Nguyen, QU, Nguyen, DN, Hoang, DT & Dutkiewicz, E 2020, 'Deep Transfer Learning for IoT Attack Detection', IEEE Access, vol. 8, pp. 107335-107344.
View/Download from: Publisher's site
View description>>

The digital revolution has substantially changed our lives in which Internet-of-Things (IoT) plays a prominent role. The rapid development of IoT to most corners of life, however, leads to various emerging cybersecurity threats. Therefore, detecting and preventing potential attacks in IoT networks have recently attracted paramount interest from both academia and industry. Among various attack detection approaches, machine learning-based methods, especially deep learning, have demonstrated great potential thanks to their early detecting capability. However, these machine learning techniques only work well when a huge volume of data from IoT devices with label information can be collected. Nevertheless, the labeling process is usually time consuming and expensive, thus, it may not be able to adapt with quick evolving IoT attacks in reality. In this paper, we propose a novel deep transfer learning (DTL) method that allows to learn from data collected from multiple IoT devices in which not all of them are labeled. Specifically, we develop a DTL model based on two AutoEncoders (AEs). The first AE (AE 1 ) is trained on the source datasets (source domains) in the supervised mode using the label information and the second AE (AE 2 ) is trained on the target datasets (target domains) in an unsupervised manner without label information. The transfer learning process attempts to force the latent representation (the bottleneck layer) of AE 2 similarly to the latent representation of AE 1 . After that, the latent representation of AE 2 is used to detect attacks in the incoming samples in the target domain. We carry out intensive experiments on nine recent IoT datasets to evaluate the performance of the proposed model. The experimental results demonstrate that the proposed DTL model significantly improves the accuracy in detecting IoT attacks compared to the baseline deep learning technique and two recent DTL approaches.

Wang, L, Yang, Y, Li, S, Deng, L, Hong, W, Zhang, C, Zhu, J & McGloin, D 2020, 'Terahertz Reconfigurable Metasurface for Dynamic Non-Diffractive Orbital Angular Momentum Beams using Vanadium Dioxide', IEEE Photonics Journal, vol. 12, no. 3, pp. 1-12.
View/Download from: Publisher's site
View description>>

© 2009-2012 IEEE. Orbital angular momentum (OAM) generation based on metasurfaces has attracted tremendous interest due to its potential in capacity enhancement of high-speed wireless communication systems. Reconfigurability is one of the key desired characteristics for the design of future metasurfaces. In this paper, a metasurface taking advantage of vanadium dioxide (VO2) is proposed. The proposed design can generate a non-diffractive OAM beam and achieve the multiple reconfigurability of the topological charge, beam radius, beam deflection angle. The operation frequency can be adjusted by controlling the state of VO2 at terahertz (THz) region. Simulation results demonstrate that the designed metasurface can generate a non-diffractive OAM beam with tunable topological charge and beam radius, propagating along ±x or ±y directions with the controllable deflection angle between 6.74° and 44.77°, ranging from 0.69 THz to 0.79 THz.

Wang, N, Ma, Z, Ding, C, Jia, H, Sui, G & Gao, X 2020, 'Characteristics of Dual‐Gate Graphene Thermoelectric Devices Based on Voltage Regulation', Energy Technology, vol. 8, no. 7, pp. 1901466-1901466.
View/Download from: Publisher's site
View description>>

The bandgap, the carrier concentration, and the polarity in graphene can all be controlled by gate voltage, which provides a new opportunity for the study of the regulation of thermoelectric devices. Herein, a dual‐gate thermoelectric device model for graphene with top‐gate and back‐gate structures is proposed. Based on the influence of gate voltage on carrier concentration and the Fermi level, the relationship between the gate voltage and the channel resistance, the Seebeck coefficient, and the conductivity of dual‐gate graphene, thermoelectric devices are established according to the mechanism of carrier transport. The results demonstrate that the optimal voltage window of the Seebeck coefficient, conductivity, and power factor is obtained independently. Compared with the conventional graphene thermoelectric device without the top‐gate structure, the Seebeck coefficient and the power factor for the proposed dual‐gate structure are increased twofold and tenfold, respectively. Herein, a new approach is provided for high‐performance thermoelectric device designs with accurate regulation.

Wang, Q, Huang, Y, Jia, W, He, X, Blumenstein, M, Lyu, S & Lu, Y 2020, 'FACLSTM: ConvLSTM with focused attention for scene text recognition', Science China Information Sciences, vol. 63, no. 2.
View/Download from: Publisher's site
View description>>

© 2020, Science China Press and Springer-Verlag GmbH Germany, part of Springer Nature. Scene text recognition has recently been widely treated as a sequence-to-sequence prediction problem, where traditional fully-connected-LSTM (FC-LSTM) has played a critical role. Owing to the limitation of FC-LSTM, existing methods have to convert 2-D feature maps into 1-D sequential feature vectors, resulting in severe damages of the valuable spatial and structural information of text images. In this paper, we argue that scene text recognition is essentially a spatiotemporal prediction problem for its 2-D image inputs, and propose a convolution LSTM (ConvLSTM)-based scene text recognizer, namely, FACLSTM, i.e., focused attention ConvLSTM, where the spatial correlation of pixels is fully leveraged when performing sequential prediction with LSTM. Particularly, the attention mechanism is properly incorporated into an efficient ConvLSTM structure via the convolutional operations and additional character center masks are generated to help focus attention on right feature areas. The experimental results on benchmark datasets IIIT5K, SVT and CUTE demonstrate that our proposed FACLSTM performs competitively on the regular, low-resolution and noisy text images, and outperforms the state-of-the-art approaches on the curved text images with large margins.

Wilson, KJ, Alabd, R, Abolhasan, M, Safavi-Naeini, M & Franklin, DR 2020, 'Optimisation of monolithic nanocomposite and transparent ceramic scintillation detectors for positron emission tomography', Scientific Reports, vol. 10, no. 1.
View/Download from: Publisher's site
View description>>

AbstractHigh-resolution arrays of discrete monocrystalline scintillators used for gamma photon coincidence detection in PET are costly and complex to fabricate, and exhibit intrinsically non-uniform sensitivity with respect to emission angle. Nanocomposites and transparent ceramics are two alternative classes of scintillator materials which can be formed into large monolithic structures, and which, when coupled to optical photodetector arrays, may offer a pathway to low cost, high-sensitivity, high-resolution PET. However, due to their high optical attenuation and scattering relative to monocrystalline scintillators, these materials exhibit an inherent trade-off between detection sensitivity and the number of scintillation photons which reach the optical photodetectors. In this work, a method for optimising scintillator thickness to maximise the probability of locating the point of interaction of 511 keV photons in a monolithic scintillator within a specified error bound is proposed and evaluated for five nanocomposite materials (LaBr3:Ce-polystyrene, Gd2O3-polyvinyl toluene, LaF3:Ce-polystyrene, LaF3:Ce-oleic acid and YAG:Ce-polystyrene) and four ceramics (GAGG:Ce, GLuGAG:Ce, GYGAG:Ce and LuAG:Pr). LaF3:Ce-polystyrene and GLuGAG:Ce were the best-performing nanocomposite and ceramic materials, respectively, with maximum sensitivities of 48.8% and 67.8% for 5 mm localisation accuracy with scintillator thicknesses of 42.6 mm and 27.5 mm, respectively.

Wong, S-W, Lin, J-Y, YangYang, Zhu, H, Chen, R-S, Zhu, L & He, Y 2020, 'Cavity Balanced and Unbalanced Diplexer Based on Triple-Mode Resonator', IEEE Transactions on Industrial Electronics, vol. 67, no. 6, pp. 4969-4979.
View/Download from: Publisher's site
View description>>

© 1982-2012 IEEE. In this paper, a series of designs for cavity balanced and unbalanced diplexer are proposed. The balanced and unbalanced designs can be categorized into four groups - unbalanced-to-unbalanced, unbalanced-to-balanced, balanced-to-unbalanced (B2U), and balanced-to-balanced. First, two approaches to achieve out-of-phase characteristics of three fundamental modes, namely TE011, TE101, and TM110 in a single triple-mode resonator, are proposed for balun filter designs. Second, four types of unbalanced and balanced diplexers are presented by adopting these three fundamental modes, of which the Butterworth response applies with specific external quality and coupling coefficient. To the authors' best knowledge, full-metal cavity balun diplexer and balanced diplexer are not reported in the open literature. For proof of concept, the design of a B2U diplexer is fabricated and measured. Good matching between simulated and measured results shows the accuracy of the proposed design and methodology, which would be attractive in the high-power radio frequency (RF) front-end systems.

Wu, K, Ni, W, Andrew Zhang, J, Liu, RP & Jay Guo, Y 2020, 'Refinement of Optimal Interpolation Factor for DFT Interpolated Frequency Estimator', IEEE Communications Letters, vol. 24, no. 4, pp. 782-786.
View/Download from: Publisher's site
View description>>

© 1997-2012 IEEE. Frequency estimation is a fundamental problem in many areas. The previously proposed q-shift estimator (QSE), which interpolates the discrete Fourier transform (DFT) coefficients by a factor of q, enables the estimation accuracy to approach the Cramér-Rao lower bound (CRLB). However, it becomes less effective when the number of samples is small. In this letter, we provide an in-depth analysis to unveil the impact of q on the convergence of QSE, and derive the bounds of a refined region of q that ensures the convergence of QSE to the CRLB even with a small number of samples. Simulations validate our analysis, showing that the refined interpolation factor is able to reduce the estimation mean squared error of QSE by up to 13.14 dB when the sample number is as small as 8.

Wu, K, Ni, W, Zhang, JA, Liu, RP & Guo, J 2020, 'Secrecy Rate Analysis for Millimeter-Wave Lens Antenna Array Transmission', IEEE Communications Letters, vol. 24, no. 2, pp. 272-276.
View/Download from: Publisher's site
View description>>

© 1997-2012 IEEE. Physical layer security is vital to millimeter-wave communications enabled by large-scale arrays, particularly the energy-efficient lens antenna arrays (LAAs). However, the broad application of LAAs can be hindered by the lack of a proper understanding of the secrecy performance. This letter derives an asymptotic closed-form expression for the secrecy rate of LAA, despite the critical challenges including the coupling of unknown lens beam responses. With the new secrecy rate analysis, the optimal power assignment for the legitimate transmission is achieved, leading to the maximization of LAA secrecy. This power assignment is unprecedentedly studied in LAA due to the previous absence of an analytical secrecy rate. Simulations validate the accuracy of the analysis over wide ranges of system parameters.

Wu, K, Zhang, JA, Huang, X, Guo, YJ & Jr, RWH 2020, 'Waveform Design and Accurate Channel Estimation for Frequency-Hopping MIMO Radar-Based Communications', IEEE Transactions on Communications, vol. PP, no. 99, pp. 1-1.
View/Download from: Publisher's site
View description>>

Frequency-hopping (FH) MIMO radar-based dual-function radar communication(FH-MIMO DFRC) enables communication symbol rate to exceed radar pulserepetition frequency, which requires accurate estimations of timing offset andchannel parameters. The estimations, however, are challenging due to unknown,fast-changing hopping frequencies and the multiplicative coupling betweentiming offset and channel parameters. In this paper, we develop accuratemethods for a single-antenna communication receiver to estimate timing offsetand channel for FH-MIMO DFRC. First, we design a novel FH-MIMO radar waveform,which enables a communication receiver to estimate the hopping frequencysequence (HFS) used by radar, instead of acquiring it from radar. Importantly,the novel waveform incurs no degradation to radar ranging performance. Then,via capturing distinct HFS features, we develop two estimators for timingoffset and derive mean squared error lower bound of each estimator. Using thebounds, we design an HFS that renders both estimators applicable. Furthermore,we develop an accurate channel estimation method, reusing the single hop fortiming offset estimation. Validated by simulations, the accurate channelestimates attained by the proposed methods enable the communication performanceof DFRC to approach that achieved based on perfect timing and ideal knowledgeof channel.

Wu, L, Xu, M, Qian, S & Cui, J 2020, 'Image to Modern Chinese Poetry Creation via a Constrained Topic-aware Model', ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 16, no. 2, pp. 1-21.
View/Download from: Publisher's site
View description>>

Artificial creativity has attracted increasing research attention in the field of multimedia and artificial intelligence. Despite the promising work on poetry/painting/music generation, creating modern Chinese poetry from images, which can significantly enrich the functionality of photo-sharing platforms, has rarely been explored. Moreover, existing generation models cannot tackle three challenges in this task: (1) Maintaining semantic consistency between images and poems; (2) preventing topic drift in the generation; (3) avoidance of certain words appearing frequently. These three points are even common challenges in other sequence generation tasks. In this article, we propose a Constrained Topic-aware Model (CTAM) to create modern Chinese poetries from images regarding the challenges above. Without image-poetry paired dataset, we construct a visual semantic vector to embed visual contents via image captions. For the topic-drift problem, we propose a topic-aware poetry generation model. Additionally, we design an Anti-frequency Decoding (AFD) scheme to constrain high-frequency characters in the generation. Experimental results show that our model achieves promising performance and is effective in poetry’s readability and semantic consistency.

Wu, L, Xu, M, Wang, J & Perry, S 2020, 'Recall What You See Continually Using GridLSTM in Image Captioning', IEEE Transactions on Multimedia, vol. 22, no. 3, pp. 808-818.
View/Download from: Publisher's site
View description>>

The goal of image captioning is to automatically describe an image with a sentence, and the task has attracted research attention from both the computer vision and natural-language processing research communities. The existing encoder–decoder model and its variants, which are the most popular models for image captioning, use the image features in three ways: first, they inject the encoded image features into the decoder only once at the initial step, which does not enable the rich image content to be explored sufficiently while gradually generating a text caption; second, they concatenate the encoded image features with text as extra inputs at every step, which introduces unnecessary noise; and, third, they using an attention mechanism, which increases the computational complexity due to the introduction of extra neural nets to identify the attention regions. Different from the existing methods, in this paper, we propose a novel network, Recall Network, for generating captions that are consistent with the images. The recall network selectively involves the visual features by using a GridLSTM and, thus, is able to recall image contents while generating each word. By importing the visual information as the latent memory along the depth dimension LSTM, the decoder is able to admit the visual features dynamically through the inherent LSTM structure without adding any extra neural nets or parameters. The Recall Network efficiently prevents the decoder from deviating from the original image content. To verify the efficiency of our model, we conducted exhaustive experiments on full and dense image captioning. The experimental results clearly demonstrate that our recall network outperforms the conventional encoder–decoder model by a large margin and that it performs comparably to the state-of-the-art methods.

Wu, W, Xu, M, Liang, Q, Mei, L & Peng, Y 2020, 'Multi‐camera 3D ball tracking framework for sports video', IET Image Processing, vol. 14, no. 15, pp. 3751-3761.
View/Download from: Publisher's site

Xi, Y, Zheng, J, He, X, Jia, W, Li, H, Xie, Y, Feng, M & Li, X 2020, 'Beyond context: Exploring semantic similarity for small object detection in crowded scenes', Pattern Recognition Letters, vol. 137, pp. 53-60.
View/Download from: Publisher's site
View description>>

© 2019 Small object detection in crowded scene aims to find those tiny targets with very limited resolution from crowded scenes. Due to very little information available on tiny objects, it is often not suitable to detect them merely based on the information presented inside their bounding boxes, resulting low accuracy. In this paper, we propose to exploit the semantic similarity among all predicted objects’ candidates to boost the performance of detectors when handling tiny objects. For this purpose, we construct a pairwise constraint to depict such semantic similarity and propose a new framework based on Discriminative Learning and Graph-Cut techniques. Experiments conducted on three widely used benchmark datasets demonstrate the improvement over the state-of-the-art approaches gained by applying this idea.

Xi, Y, Zheng, J, Jia, W, He, X, Li, H, Ren, Z & Lam, K-M 2020, 'See Clearly in the Distance: Representation Learning GAN for Low Resolution Object Recognition', IEEE Access, vol. 8, pp. 53203-53214.
View/Download from: Publisher's site
View description>>

© 2013 IEEE. Identifying tiny objects with extremely low resolution is generally considered a very challenging task even for human vision, due to limited information presented inside the object areas. There have been very limited attempts in recent years to deal with low-resolution recognition. The existing solutions rely on either generating super-resolution images or learning multi-scale features. However, their performance improvement becomes very limited, especially when the resolution becomes very low. In this paper, we propose a Representation Learning Generative Adversarial Network (RL-GAN) to generate super image representation that is optimized for recognition. Our solution deals with the classical vision task of object recognition in the distance. We evaluate our idea on the challenging task of low-resolution object recognition. Comparison of experimental results conducted on public and our newly created WIDER-SHIP datasets demonstrate the effectiveness of our RL-GAN, which improves the classification results significantly, with 10-15% gain on average, compared with benchmark solutions.

Xie, H & Veitch, D 2020, 'Nested saturation control of multiple vector integrators and its application to motion control of UAVs', International Journal of Robust and Nonlinear Control, vol. 30, no. 1, pp. 246-265.
View/Download from: Publisher's site
View description>>

SummaryThis paper presents two nested input saturation control schemes for a special class of multiple vector integrators with bounded additive disturbances. The considered systems originate from the motion control of rotary‐wing unmanned aerial vehicles (UAVs). The first scheme is based on a feedforward form, which requires state transformation and can be applied to stabilize arbitrary order vector integrator systems. The second scheme is constructed with original state variables using a new approach, applied here to double vector integrator systems. The capability of handling external disturbance using the two schemes is also analyzed. The two schemes are applied to design motion controllers for rotary‐wing UAVs and simulation results are provided to show the performances of two controllers.

Xu, J-X, Zhang, XY, Li, H-Y, Yang, Y, Dutkiewicz, E & Xue, Q 2020, 'Ultracompact Multichannel Bandpass Filter Based on Trimode Dielectric-Loaded Cavities', IEEE Transactions on Microwave Theory and Techniques, vol. 68, no. 5, pp. 1668-1677.
View/Download from: Publisher's site
View description>>

© 1963-2012 IEEE. In this article, a method for designing multichannel bandpass filters (BPFs) based on trimode dielectric-loaded cavities is presented. A large number of BPFs are integrated as one multichannel BPF with multiple inputs and multiple outputs, resulting in an ultracompact size. The cubic trimode dielectric-loaded cavities are utilized with the TE101, TE011, and TM110 modes resonating at the same frequency and orthogonal to each other. Feeding probes and coupling probes are properly arranged where the three modes in one cavity can be excited for different BPF channels without interference with each other. Consequently, excellent isolation among multiple channels can be obtained. The multichannel BPF is designed based on a 3-D structure, which can be easily extended to higher filter orders with a larger number of channels to satisfy different requirements in wireless systems. For demonstration, a 12-channel BPF is fabricated and measured, which exhibits good filtering responses of each channel and high isolation among channels. Significant size reduction is achieved compared to conventional multiple single-channel filters, which is potential in high-integration base station applications.

Xu, K-D, Zhu, X, Yang, Y & Chen, Q 2020, 'A Broadband On-Chip Bandpass Filter Using Shunt Dual-Layer Meander-Line Resonators', IEEE Electron Device Letters, vol. 41, no. 11, pp. 1617-1620.
View/Download from: Publisher's site

Yan, J, Liu, H, Zhu, X, Men, K & Yeo, KS 2020, 'Ka-Band Marchand Balun with Edge- and Broadside-Coupled Hybrid Configuration', Electronics, vol. 9, no. 7, pp. 1116-1116.
View/Download from: Publisher's site
View description>>

This article presents a novel Ka-band Marchand balun implemented in 0.13-μm SiGe bipolar complementary metal–oxide–semiconductor (BiCMOS) process. By combining both edge- and broadside-coupled structures, the new hybrid balun is able to increase the coupling and minimize the balun insertion loss. As compared with conventional edge-coupled or broadside-coupled structures, the proposed balun achieves the lowest insertion loss of 1.02 dB across a wide 1-dB bandwidth from 29.0 GHz to 46.0 GHz, with a core size of 270 μm × 280 μm.

Yang, Y, Hou, ZJ, Zhu, X, Che, W & Xue, Q 2020, 'A Millimeter-Wave Reconfigurable On-Chip Coupler With Tunable Power-Dividing Ratios in 0.13-$\mu$ m BiCMOS Technology', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 5, pp. 1516-1526.
View/Download from: Publisher's site
View description>>

This paper presents a millimeter-wave (mm-wave) on-chip coupler with tunable power dividing ratios and constant phase response. Composed by two coupled lines, two capacitors and two series-connected varactors, the proposed tunable coupler offers broadband frequency responses for 5G applications. Theoretical analysis for wideband operation is provided. For demonstration, a millimeter-wave tunable coupler is implemented in a standard 0.13-\mu \text{m} SiGe (Bi) CMOS technology and measured through an on-wafer probing system. From 24 to 38 GHz, the proposed tunable coupler shows a power-dividing ratio ranged from 0 to 6.5 dB, while maintaining an in-band return loss of better than 10 dB and output isolation of 20 dB, simultaneously. The phase imbalance is better than ±2.5° with a measured insertion loss of 1.3 dB across the entire tuning range. To the authors' best knowledge, this is the first time that an on-chip coupler with tunable power-dividing ratios is reported operating at mm-wave bands for, particularly, 5G applications.

Yao, Y, Shen, F, Xie, G, Liu, L, Zhu, F, Zhang, J & Shen, HT 2020, 'Exploiting Web Images for Multi-Output Classification: From Category to Subcategories', IEEE Transactions on Neural Networks and Learning Systems, vol. 31, no. 7, pp. 1-13.
View/Download from: Publisher's site

Yao, Y, Zhang, J, Shen, F, Liu, L, Zhu, F, Zhang, D & Shen, HT 2020, 'Towards Automatic Construction of Diverse, High-Quality Image Datasets', IEEE Transactions on Knowledge and Data Engineering, vol. 32, no. 6, pp. 1199-1211.
View/Download from: Publisher's site
View description>>

© 1989-2012 IEEE. The availability of labeled image datasets has been shown critical for high-level image understanding, which continuously drives the progress of feature designing and models developing. However, constructing labeled image datasets is laborious and monotonous. To eliminate manual annotation, in this work, we propose a novel image dataset construction framework by employing multiple textual queries. We aim at collecting diverse and accurate images for given queries from the Web. Specifically, we formulate noisy textual queries removing and noisy images filtering as a multi-view and multi-instance learning problem separately. Our proposed approach not only improves the accuracy but also enhances the diversity of the selected images. To verify the effectiveness of our proposed approach, we construct an image dataset with 100 categories. The experiments show significant performance gains by using the generated data of our approach on several tasks, such as image classification, cross-dataset generalization, and object detection. The proposed method also consistently outperforms existing weakly supervised and web-supervised approaches.

Yu, G, Wang, X, Yu, K, Ni, W, Zhang, JA & Liu, RP 2020, 'Survey: Sharding in Blockchains', IEEE Access, vol. 8, pp. 14155-14181.
View/Download from: Publisher's site
View description>>

© 2013 IEEE. The Blockchain technology, featured with its decentralized tamper-resistance based on a Peer-to-Peer network, has been widely applied in financial applications, and even further been extended to industrial applications. However, the weak scalability of traditional Blockchain technology severely affects the wide adoption due to the well-known trillema of decentralization-security-scalability in Blockchains. In regards to this issue, a number of solutions have been proposed, targeting to boost the scalability while preserving the decentralization and security. They range from modifying the on-chain data structure and consensus algorithms to adding the off-chain technologies. Therein, one of the most practical methods to achieve horizontal scalability along with the increasing network size is sharding, by partitioning network into multiple shards so that the overhead of duplicating communication, storage, and computation in each full node can be avoided. This paper presents a survey focusing on sharding in Blockchains in a systematic and comprehensive way. We provide detailed comparison and quantitative evaluation of major sharding mechanisms, along with our insights analyzing the features and restrictions of the existing solutions. We also provide theoretical upper-bound of the throughput for each considered sharding mechanism. The remaining challenges and future research directions are also reviewed.

Yu, G, Zha, X, Wang, X, Ni, W, Yu, K, Yu, P, Zhang, JA, Liu, RP & Guo, YJ 2020, 'Enabling Attribute Revocation for Fine-Grained Access Control in Blockchain-IoT Systems', IEEE Transactions on Engineering Management, vol. 67, no. 4, pp. 1213-1230.
View/Download from: Publisher's site
View description>>

© 1988-2012 IEEE. The attribute-based encryption (ABE) has drawn a lot of attention for fine-grained access control in blockchains, especially in blockchain-enabled tampering-resistant Internet-of-Things (IoT) systems. However, its adoption has been severely hindered by the incompatibility between the immutability of typical blockchains and the attribute updates/revocations of ABE. In this article, we propose a new blockchain-based IoT system, which is compatible with the ABE technique, and fine-grained access control is implemented with the attribute update enabled by integrating Chameleon Hash algorithms into the blockchains. We design and implement a new verification scheme over a multilayer blockchain architecture to guarantee the tamper resistance against malicious and abusive tampering. The system can provide an update-oriented access control, where historical on-chain data can only be accessible to new members and inaccessible to the revoked members. This is distinctively different from existing solutions, which are threatened by data leakage toward the revoked members. We also provide analysis and simulations showing that our system outperforms other solutions in terms of overhead, searching complexity, security, and compatibility.

Yu, G, Zha, X, Wang, X, Ni, W, Yu, K, Zhang, JA & Liu, RP 2020, 'A Unified Analytical model for proof-of-X schemes', Computers & Security, vol. 96, pp. 101934-101934.
View/Download from: Publisher's site
View description>>

© 2020 Nakamoto protocol, practically solving the Byzantine Generals Problem, can support a variety of proof-based consensus engines, referred to as Proof-of-X (PoX) in permissionless Blockchains. However, there has been to date in lack of a general approach for each miner to evaluate its steady-state profit against the competitors. This paper presents a Markov model which captures explicitly the weighted resource distribution of PoX schemes in large-scale networks and unifies the analysis of different PoX schemes. The new model leads to the development of three new unified metrics for the evaluation, namely, Resource Sensitivity, System Convergence, and Resource Fairness, accounting for security, stability, and fairness, respectively. The generality and applicability of our model are validated by simulation results, revealing that among typically non-Fairness-oriented PoX schemes (such as Proof-of-Work (PoW) and Proof-of-Stake (PoS)), the strongly restricted coinage-based PoS with a Pareto-distributed resource can offer the best performance on Resource Sensitivity, while Proof-of-Publication (PoP) with normal-distributed resource performs the best on System Convergence. Our simulations also reveal the important role of carefully designed Resource Fairness parameter in balancing Resource Sensitivity and System Convergence and improving the performance compared with other non-Fairness-oriented PoX schemes.

Yu, H, Zhang, T & Jia, W 2020, 'Shared subspace least squares multi-label linear discriminant analysis', Applied Intelligence, vol. 50, no. 3, pp. 939-950.
View/Download from: Publisher's site
View description>>

© 2019, Springer Science+Business Media, LLC, part of Springer Nature. Multi-label linear discriminant analysis (MLDA) has been explored for multi-label dimension reduction. However, MLDA involves dense matrices eigen-decomposition which is known to be computationally expensive for large-scale problems. In this paper, we show that the formulation of MLDA can be equivalently casted as a least squares problem so as to significantly reduce the computation burden and scale to the data collections with higher dimension. Further, it is also found that appealing regularization techniques can be incorporated into the least-squares model to boost generalization accuracy. Experimental results on several popular multi-label benchmarks not only verify the established equivalence relationship, but also demonstrate the effectiveness and efficiency of our proposed algorithms.

Yuan, C, Tao, X, Ni, W, Li, N, Jamalipour, A & Liu, RP 2020, 'Joint Power Allocation and Beamforming for Overlaid Secrecy Transmissions in MIMO-OFDM Channels', IEEE Transactions on Vehicular Technology, vol. 69, no. 9, pp. 10019-10032.
View/Download from: Publisher's site

Yuan, W, Wu, N, Zhang, A, Huang, X, Li, Y & Hanzo, L 2020, 'Iterative Receiver Design for FTN Signaling Aided Sparse Code Multiple Access', IEEE Transactions on Wireless Communications, vol. 19, no. 2, pp. 915-928.
View/Download from: Publisher's site
View description>>

© 2002-2012 IEEE. The sparse code multiple access (SCMA) is a promising candidate for bandwidth-efficient next generation wireless communications, since it can support more users than the number of resource elements. On the same note, faster-than-Nyquist (FTN) signaling can also be used to improve the spectral efficiency. Hence in this paper, we consider a combined uplink FTN-SCMA system in which the data symbols corresponding to a user are further packed using FTN signaling. As a result, a higher spectral efficiency is achieved at the cost of introducing intentional inter-symbol interference (ISI). To perform joint channel estimation and detection, we design a low complexity iterative receiver based on the factor graph framework. In addition, to reduce the signaling overhead and transmission latency of our SCMA system, we intrinsically amalgamate it with grant-free scheme. Consequently, the active and inactive users should be distinguished. To address this problem, we extend the aforementioned receiver and develop a new algorithm for jointly estimating the channel state information, detecting the user activity and for performs data detection. In order to further reduce the complexity, an energy minimization based approximation is employed for restricting the user state to Gaussian. Finally, a hybrid message passing algorithm is conceived. Our Simulation results show that the FTN-SCMA system relying on the proposed receiver design has a higher throughput than conventional SCMA scheme at a negligible performance loss.

Yuan, X, Feng, Z, Ni, W, Liu, RP, Zhang, JA & Xu, W 2020, 'Secrecy Performance of Terrestrial Radio Links Under Collaborative Aerial Eavesdropping', IEEE Transactions on Information Forensics and Security, vol. 15, pp. 604-619.
View/Download from: Publisher's site
View description>>

© 2005-2012 IEEE. Motivated to understand the increasingly severe threat of unmanned aerial vehicles (UAVs) to the confidentiality of terrestrial radio links, this paper analyzes the ergodic and ϵ-outage secrecy capacities of the links in the presence of multiple cooperative aerial eavesdroppers flying autonomously in three-dimensional (3D) spaces and exploiting selection combining (SC) or maximal ratio combining (MRC). The 'cut-off' density of the eavesdroppers under which the secrecy capacities vanish is identified. By decoupling the analysis of the random trajectories from the random channel fading, closed-form approximations with almost sure convergence to the secrecy capacities are devised. The analysis is extended to study the impact of the oscillator phase noises and finite memories of the aerial eavesdroppers on the secrecy performance of the ground link. Validated by simulations, the cut-off density only depends on the range of the link in the case of SC eavesdropping, while it depends on the flight region of the eavesdroppers in the case of MRC eavesdropping.

Yuan, X, Feng, Z, Ni, W, Wei, Z, Liu, RP & Xu, C 2020, 'Connectivity of UAV Swarms in 3D Spherical Spaces Under (Un)Intentional Ground Interference', IEEE Transactions on Vehicular Technology, vol. 69, no. 8, pp. 8792-8804.
View/Download from: Publisher's site
View description>>

© 1967-2012 IEEE. This paper analyzes the wireless connectivity of an Unmanned Aerial Vehicle (UAV) swarm in the presence of (un)intentional external interference from the ground. Different from existing studies, the swarm UAVs fly independently around a given three-dimensional (3D) location and are all within a 3D spherical space. Closed-form bounds are delivered for the average outage probability of a UAV from its nearest neighbor in the swarm, and the density of the swarm, which allows the swarm to operate uninterruptedly in the presence of the interference. Our analysis involves closed-form approximations of the instantaneous outage probability of a UAV from its nearest neighbor by using the first-order Marcum Q-function and the zero-Th order modified Bessel function of the first kind. The analysis also involves applying Jensen's inequality to the instantaneous outage probability to bound the average outage probability and the density of the swarm. Corroborated by simulations, our analysis is accurate, and useful to evaluate the impact of external interference on the connectivity of UAV swarms. Interesting insights are shed on the connectivity and coverage of the UAV swarm.

Zeng, J, Lv, T, Lin, Z, Liu, RP, Mei, J, Ni, W & Guo, YJ 2020, 'Achieving Ultrareliable and Low-Latency Communications in IoT by FD-SCMA', IEEE Internet of Things Journal, vol. 7, no. 1, pp. 363-378.
View/Download from: Publisher's site
View description>>

To enable ultrareliable and low-latency communications (URLLCs) in the Internet of Things (IoT), a sparse-code multiple-access (SCMA)-enhanced full-duplex (FD) scheme (FD-SCMA) is proposed in this article. FD-SCMA can support short-packet transmissions of several SCMA users in the uplink (UL) and downlink (DL) simultaneously by an FD next generation node B (gNB). First, the gNB and UL users can generate and superpose signals according to the preconfigured SCMA codebooks, and simultaneously transmit the signals via occupied subcarriers in a joint SCMA pattern. The receivers at the gNB and DL users can demodulate and decode the signals with multiuser detection (MUD). With the imperfect self-interference suppression (SIS) of FD considered, the effective signal-to-noise ratio (SNR) of FD-SCMA at the gNB and DL users is formulated. The error probability of FD-SCMA in the UL and DL is also derived under a given transmission latency constraint of short-packet transmissions. In the stationary flat-fading channel, it is proved that FD-SCMA can achieve better reliability than the existing FD and SCMA schemes. In the time-invariant frequency-selective fading channel, the upper bounds for error probability of the UL and DL users in FD-SCMA are derived, respectively. Through the theoretical calculation and Monte Carlo simulation, it is verified that the superiority of FD-SCMA in supporting ultrareliable and low-latency short-packet transmissions in IoT.

Zeng, J, Lv, T, Liu, RP, Su, X, Guo, YJ & Beaulieu, NC 2020, 'Enabling Ultrareliable and Low-Latency Communications Under Shadow Fading by Massive MU-MIMO', IEEE Internet of Things Journal, vol. 7, no. 1, pp. 234-246.
View/Download from: Publisher's site
View description>>

© 2014 IEEE. It is challenging to satisfy the critical requirements of ultrareliable and low-latency communications (URLLCs) in the Internet of Things (IoT) under severe channel fading. The emerging massive multiuser multiple-input-multiple-output (MU-MIMO) concept is applied in IoT networks under shadow fading, enabling URLLC with pilot-assisted channel estimation (PACE) and zero-forcing (ZF) detection. Assuming users are uniformly and randomly deployed under log-normal shadow fading, the probability density function (pdf) of postprocessing signal-to-noise ratios (SNRs) is derived for the uplink (UL) of massive MU-MIMO with perfect channel state information (CSI) and imperfect CSI obtained by PACE. Then, finite blocklength (FBL) information theory is utilized to derive the error probability of accessing users with a given latency, thereby evaluating the reliability of massive MU-MIMO for short-packet transmissions. Further, the length of pilots to minimize the error probability can be decided by the golden section search method (GSSM), which can converge rapidly. Numerical results verify that massive MU-MIMO can support a large number of UL URLLC users even when users are randomly deployed under shadow fading.

Zhang, H & Xu, M 2020, 'Improving the generalization performance of deep networks by dual pattern learning with adversarial adaptation', Knowledge-Based Systems, vol. 200, pp. 106016-106016.
View/Download from: Publisher's site

Zhang, H, Huang, X, Zhang, JA & Guo, YJ 2020, 'Dual Pulse Shaping Transmission and Equalization for High-Speed Wideband Wireless Communication Systems', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 7, pp. 2372-2382.
View/Download from: Publisher's site
View description>>

© 2004-2012 IEEE. Analog-to-digital and digital-to-analog conversion devices for signals with very large bandwidth are not always available due to technical or cost issues. This limits the realization of very high data-rate digital communication systems. In this paper, we propose a dual pulse shaping (DPS) transmission scheme, which can achieve full Nyquist rate transmission with only a half of the sampling rate for each of the two data streams. Two classes of ideal complementary Nyquist pulses are formulated assuming raised-cosine (RC) pulse shaping. The condition for cross-symbol interference (CSI) free transmission is derived and validated for the proposed pulses. Structures of the DPS transmitter and receiver are described and low-complexity equalization techniques tailored to DPS are proposed. With DPS, a millimeter wave system with commercially available and affordable data conversion devices is exemplified for achieving high-speed low-cost wireless communications. Simulation results with two sets of practical dual spectral shaping pulses are provided. The results verify the effectiveness of the proposed scheme, with comparison to the benchmark conventional Nyquist pulse shaping system.

Zhang, JA, Rahman, ML, Wu, K, Huang, X, Guo, YJ, Chen, S & Yuan, J 2020, 'Enabling Joint Communication and Radar Sensing in Mobile Networks -- A Survey'.
View description>>

Mobile network is evolving from a communication-only network towards one withjoint communication and radar/radio sensing (JCAS) capabilities, that we callperceptive mobile network (PMN). In PMNs, JCAS integrates sensing intocommunications, sharing a majority of system modules and the same transmittedsignals. The PMN is expected to provide a ubiquitous radio sensing platform andenable a vast number of novel smart applications, whilst providingnon-compromised communications. In this paper, we present a broad picture ofthe motivation, methodologies, challenges, and research opportunities ofrealizing PMN, by providing a comprehensive survey for systems and technologiesdeveloped mainly in the last ten years. Beginning by reviewing the work oncoexisting communication and radar systems, we highlight their limits onaddressing the interference problem, and then introduce the JCAS technology. Wethen set up JCAS in the mobile network context and envisage its potentialapplications. We continue to provide a brief review of three types of JCASsystems, with particular attention to their differences in design philosophy.We then introduce a framework of PMN, including the system platform andinfrastructure, three types of sensing operations, and signals usable forsensing. Subsequently, we discuss required system modifications to enablesensing on current communication-only infrastructure. Within the context ofPMN, we review stimulating research problems and potential solutions, organizedunder nine topics: performance bounds, waveform optimization, antenna arraydesign, clutter suppression, sensing parameter estimation, resolution ofsensing ambiguity, pattern analysis, networked sensing under cellular topology,and sensing-assisted communications. We conclude the paper by listing key openresearch problems for the aforementioned topics and sharing some lessons thatwe have learned.

Zhang, P, Xu, J, Wu, Q, Huang, Y & Zhang, J 2020, 'Top-Push Constrained Modality-Adaptive Dictionary Learning for Cross-Modality Person Re-Identification', IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 12, pp. 4554-4566.
View/Download from: Publisher's site

Zhang, R, Wu, L, Yang, Y, Wu, W, Chen, Y & Xu, M 2020, 'Multi-camera multi-player tracking with deep player identification in sports video', Pattern Recognition, vol. 102, pp. 107260-107260.
View/Download from: Publisher's site
View description>>

© 2020 Identity switches caused by inter-object interactions remain a critical problem for multi-player tracking in real-world sports video analysis. Existing approaches utilizing the appearance model is difficult to associate detections and preserve identities due to the similar appearance of players in the same team. Instead of the appearance model, we propose a distinguishable deep representation for player identity in this paper. A robust multi-player tracker incorporating with deep player identification is further developed to produce identity-coherent trajectories. The framework consists of three parts: (1) the core component, a Deep Player Identification (DeepPlayer) model that provides an adequate discriminative feature through the coarse-to-fine jersey number recognition and the pose-guided partial feature embedding; (2) an Individual Probability Occupancy Map (IPOM) model for players 3D localization with ID; and (3) a K-Shortest Path with ID (KSP-ID) model that links nodes in the flow graph by a proposed player ID correlation coefficient. With the distinguishable identity, the performance of tracking is improved. Experiment results illustrate that our framework handles the identity switches effectively, and outperforms state-of-the-art trackers on the sports video benchmarks.

Zhang, Z, Cheng, Z, Li, H, Ke, H & Guo, YJ 2020, 'A Broadband Doherty Power Amplifier With Hybrid Class-EFJ Mode', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 12, pp. 4270-4280.
View/Download from: Publisher's site
View description>>

This paper proposes a method that employs novel hybrid continuous class-EFJ power amplifiers (PAs) as carrier PA to design a broadband high-efficiency Doherty power amplifier (DPA). Bandwidth characteristic of the proposed DPA is analyzed in detail. By proper selection of related parameter values, up to 78% fabrication bandwidth can be obtained. Post-harmonic tuning network is applied to improve the bandwidth and enhance the efficiency. Then, a closed design process is presented to design broadband DPA based on derived theories. For validation, a broadband DPA operating in 1.2-2.8 GHz is designed and fabricated. Measurements illustrate that the DPA can deliver saturated output power between 43.7 dBm and 44.1 dBm in 1.2-2.8 GHz, and the saturated drain efficiency from 60.5% to 74.2 % is achieved. Moreover, drain efficiency is 48.1%-57.6% at the 6 dB power back-off. Compared with conventional DPAs, the proposed DPA exhibits superior performance of bandwidth characteristics and power back-off efficiency over a wide bandwidth.

Zhao, M, Zhang, C, Zhang, J, Porikli, F, Ni, B & Zhang, W 2020, 'Scale-Aware Crowd Counting via Depth-Embedded Convolutional Neural Networks', IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 10, pp. 3651-3662.
View/Download from: Publisher's site
View description>>

© 1991-2012 IEEE. Scale variation of pedestrians in a crowd image presents a significant challenge for vision-based people counting systems. Such variations are mainly caused by perspective-related distortions due to the camera pose relative to the ground plane. Following the density-based counting paradigm, we postulate that generating density values adaptive to object scales plays a critical role in the accuracy of the final counting results. Motivated by this, we distill the underlying information from depth cues to obtain scale-aware representations that can respond to object scales considering the fact that the scale is inversely proportional to the object depth. Specifically, we propose a depth embedding module as add-ons into existing networks. This module exploits essential depth cues to spatially re-calibrate the magnitude of the original features. In this way, the objects, although in the same class, will attain distinct representations according to their scales, which directly benefits the estimation of scale-aware density values. We conduct a comprehensive analysis of the effects of the depth embedding module and validate that exploiting depth cues to perceive object scale variations in convolutional neural networks improves crowd counting performances. Our experiments demonstrate the effectiveness of the proposed approach on four popular benchmark datasets.

Zhe, T, Huang, L, Wu, Q, Zhang, J, Pei, C & Li, L 2020, 'Inter-Vehicle Distance Estimation Method Based on Monocular Vision Using 3D Detection', IEEE Transactions on Vehicular Technology, vol. 69, no. 5, pp. 4907-4919.
View/Download from: Publisher's site
View description>>

© 1967-2012 IEEE. Most autonomous vehicles build their perception systems on expensive sensors, such as LIDAR, RADAR, and high-precision Global Positioning System (GPS). However, cameras can provide richer sensing at a considerably lower cost, this makes them a more appealing alternative. A driving assistance system (DAS) based on monocular vision has gradually become a research hotspot, and inter-vehicle distance estimation based on monocular vision is an important technology in DAS. There are still constrains in the existing methods for estimating the inter-vehicle distance based on monocular vision, such as low accuracy when distance is larger, unstable accuracy for different types vehicles, and significantly poor performance on distance estimation for severely occluded vehicles. To improve the accuracy and robustness of ranging results, this study proposes a monocular vision end-to-end inter-vehicle distance estimation method based on 3D detection. The actual area of the rare view of the vehicle and the corresponding projection area in the image are obtained by 3D detection method. An area-distance geometric model is then established on the basis of the camera projection principle to recover distance. Our method shows its potential in complex traffic scenarios by testing the test set data provided on the real-world computer vision benchmark, KITTI. The experimental results have superior performance than the existing published methods. Moreover, the accuracy of occluded vehicle ranging results can reach approximately 98%, while the accuracy deviation between vehicles with different visual angles is less than 2%.

Zhou, I, Lipman, J, Abolhasan, M, Shariati, N & Lamb, DW 2020, 'Frost Monitoring Cyber–Physical System: A Survey on Prediction and Active Protection Methods', IEEE Internet of Things Journal, vol. 7, no. 7, pp. 6514-6527.
View/Download from: Publisher's site

Zhu, H, Qin, P-Y & Guo, YJ 2020, 'Single-Ended-to-Balanced Power Divider With Extended Common-Mode Suppression and Its Application to Differential $2\times4$ Butler Matrices', IEEE Transactions on Microwave Theory and Techniques, vol. 68, no. 4, pp. 1510-1519.
View/Download from: Publisher's site

Zhu, H, Zhu, X, Yang, Y & Sun, Y 2020, 'Design of Miniaturized On-Chip Bandpass Filters Using Inverting-Coupled Inductors in (Bi)-CMOS Technology', IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 2, pp. 647-657.
View/Download from: Publisher's site
View description>>

© 2004-2012 IEEE. In this work, a new type of miniaturized on-chip resonator using coupled-inductor structure is presented. The impact on resonances of the structure due to the use of non- inverting- and inverting-coupled configuration is extensively investigated. It has been found that using the inverting-coupled structure, a stronger resonance can be generated, which is ideally suitable for device miniaturization. To fully understand the working mechanism of the resonator and use it effectively for bandpass filter (BPF) design, simplified LC equivalent-circuit models and detailed theoretical analysis are provided. To further demonstrate the proposed concept is useful in practice, not only a 1st-order BPF, but also another two 2nd-order BPFs are designed and fabricated in a standard 0.13-μm (Bi)-CMOS technology. All of them are designed to have a centre frequency around 15 GHz. Their physical dimensions are 0.13 × 0.25 mm2, 0.26 × 0.25 mm2, 0.24 × 0.22 mm2, respectively. Good agreements between simulation and measurement have been obtained, which verify that the presented design approach is suitable for miniaturized on-chip passive design.

Zhu, J, Yang, Y, McGloin, D, Rajasekharan Unnithan, R, Li, S, Liao, S & Xue, Q 2020, '3-D Printed Planar Dielectric Linear-to-Circular Polarization Conversion and Beam-Shaping Lenses Using Coding Polarizer', IEEE Transactions on Antennas and Propagation, vol. 68, no. 6, pp. 4332-4343.
View/Download from: Publisher's site
View description>>

© 1963-2012 IEEE. This article presents a new linear-to-circular polarization conversion coding unit, on which two new kinds of beam-shaping lenses are proposed. First, under periodic boundary conditions, a linear-to-circular polarization conversion coding unit is introduced, which introduces the necessary phase delay by adjusting its geometrical parameters. The phase delay ranges from 0° to 360° and is discretized into 3 bit coding units corresponding to specific delays. Second, by properly arranging the coding units, a high-gain circularly polarized (CP) lens is proposed. The lens achieves linear-to-circular polarization conversion and beam collimation in the transmission mode simultaneously with a planar configuration, which is different from counterparts that place a lens atop of a polarizer. Furthermore, the coding units are used to form Wollaston-prism-like and Rochon-prism-like planar CP beam-shaping lenses, which split the beams with different polarizations into right-and left-handed components. These beams can be controlled independently. Prototypes working at 30 GHz band are designed, fabricated, and measured to verify the idea.

Zou, Y, Gong, S, Xu, J, Cheng, W, Hoang, DT & Niyato, D 2020, 'Wireless Powered Intelligent Reflecting Surfaces for Enhancing Wireless Communications', IEEE Transactions on Vehicular Technology, vol. 69, no. 10, pp. 12369-12373.
View/Download from: Publisher's site
View description>>

© 1967-2012 IEEE. Recently, the intelligent reflecting surface (IRS) has become a promising technology for energy-, and spectrum-efficient communications by reconfiguring the radio environment. In this paper, we consider multiple-input single-output (MISO) transmissions from a multi-antenna access point (AP) to a receiver, assisted by a practical IRS with a power budget constraint. The IRS can work in energy harvesting, and signal reflecting phases. It firstly harvests RF energy from the AP's signal beamforming, and then uses it to sustain its operations in the signal reflecting phase. We aim to characterize the maximum capacity by optimizing the AP's transmit beamforming, the IRS's time allocation in two operational phases, and the IRS's passive beamforming to enhance the information rate. To solve the non-convex maximization problem, we exploit its structural properties, and decompose it into two sub-problems in two phases. The IRS's phase shift optimization in the reflecting phase follows a conventional passive beamforming problem to maximize the received signal power. In the energy harvesting phase, the IRS's time allocation, and the AP's transmit beamforming are jointly optimized using monotonic optimization. Simulation results verify the effectiveness of the proposed algorithm.

Zuo, Y, Fang, Y, Yang, Y, Shang, X & Wu, Q 2020, 'Depth Map Enhancement by Revisiting Multi-Scale Intensity Guidance Within Coarse-to-Fine Stages', IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 12, pp. 4676-4687.
View/Download from: Publisher's site
View description>>

IEEE Being different from the most methods of guided depth map enhancement based on deep convolutional neural network which focus on increasing the depth of networks, this paper is to improve the effectiveness of intensity guidance when the network goes deep. Overall, the proposed network upsamples the low-resolution depth maps from coarse to fine. Within each refinement stage of certain-scale depth features, the current-scale and all coarse-scales of the guidance features are revisited by dense connection. Therefore, the multi-scale guidance is efficiently maintained as the propagation of features. Furthermore, the proposed network maintains the intensity features in the high-resolution domain from which the multi-scale guidance is directly extracted. This design further improves the quality of intensity guidance. In addition, the shallow depth features upsampled via transposed convolution layer are directly transferred to the final depth features for reconstruction, which is called global residual learning in feature domain. Similarly, the global residual learning in pixel domain learns the difference between the depth ground truth and the coarsely upsampled depth map. Also, the local residual learning is to maintain the low frequency within each refinement stage and progressively recover the high frequency. The proposed method is tested for noise-free and noisy cases which compares against 16 state-of-the-art methods. Our experimental results show the improved performances based on the qualitative and quantitative evaluations.

Zuo, Y, Wu, Q, Fang, Y, An, P, Huang, L & Chen, Z 2020, 'Multi-Scale Frequency Reconstruction for Guided Depth Map Super-Resolution via Deep Residual Network', IEEE Transactions on Circuits and Systems for Video Technology, vol. 30, no. 2, pp. 297-306.
View/Download from: Publisher's site
View description>>

© 1991-2012 IEEE. The depth maps obtained by the consumer-level sensors are always noisy in the low-resolution (LR) domain. Existing methods for the guided depth super-resolution, which are based on the pre-defined local and global models, perform well in general cases (e.g., joint bilateral filter and Markov random field). However, such model-based methods may fail to describe the potential relationship between RGB-D image pairs. To solve this problem, this paper proposes a data-driven approach based on the deep convolutional neural network with global and local residual learning. It progressively upsamples the LR depth map guided by the high-resolution intensity image in multiple scales. A global residual learning is adopted to learn the difference between the ground truth and the coarsely upsampled depth map, and the local residual learning is introduced in each scale-dependent reconstruction sub-network. This scheme can restore the depth structure from coarse to fine via multi-scale frequency synthesis. In addition, batch normalization layers are used to improve the performance of depth map denoising. Our method is evaluated in noise-free and noisy cases. A comprehensive comparison against 17 state-of-the-art methods is carried out. The experimental results show that the proposed method has faster convergence speed as well as improved performances based on the qualitative and quantitative evaluations.

Alam, SL & Gill, AQ 1970, 'A social engagement framework for the government ecosystem: Insights from australian government facebook pages', International Conference on Information Systems, ICIS 2020 - Making Digital Inclusive: Blending the Local and the Global, International Conference on Information Systems, AISEL, India.
View description>>

Government agencies are using social media in an ad-hoc manner for bi-directional broadcast style communication, rather than systematic and deep engagement through open participation for co-creating value. However, capabilities and practices of participation for value creation is less understood for an increasingly networked government ecosystem. This calls for the need of a structured social engagement framework for government agencies. Thus, based on an empirical analysis of over 68 federal government Facebook pages, this paper presents insights on online engagement and levels of maturity among Australian federal government Facebook pages. Informed through engagement research and social architecture lens, we propose an empirically bounded government Facebook engagement framework (GFEF) that has implications and recommendations for agency benchmarking and social engagement capability building.

Almeida, R, Cunha, I, Teixeira, R, Veitch, D & Diot, C 1970, 'Classification of Load Balancing in the Internet', IEEE INFOCOM 2020 - IEEE Conference on Computer Communications, IEEE INFOCOM 2020 - IEEE Conference on Computer Communications, IEEE, Toronto, ON, Canada.
View/Download from: Publisher's site
View description>>

Recent advances in programmable data planes, software-defined networking, and the adoption of IPv6 support novel, more complex load balancing strategies. We introduce the Multipath Classification Algorithm (MCA), a probing algorithm that extends traceroute to identify and classify load balancing in Internet routes. MCA extends existing formalism and techniques to consider that load balancers may use arbitrary combinations of bits in the packet header for load balancing. We propose optimizations to reduce probing cost that are applicable to MCA and existing load balancing measurement techniques. Through large-scale measurement campaigns, we characterize and study the evolution of load balancing on the IPv4 and IPv6 Internet with multiple transport protocols. Our results show that load balancing is more prevalent and that load balancing strategies are more mature than previous characterizations have found.

Alsawwaf, M, Chaczko, Z & Kulbacki, M 1970, 'In Your Face: Person Identification Through Ratios of Distances Between Facial Features', Intelligent Information and Database Systems, Asian Conference on Intelligent Information and Database Systems, Springer International Publishing, Phuket, Thailand, pp. 527-536.
View/Download from: Publisher's site
View description>>

© 2020, Springer Nature Switzerland AG. These days identification of a person is an integral part of many computer-based solutions. It is a key characteristic for access control, customized services, and a proof of identity. Over the last couple of decades, many new techniques were introduced for how to identify human faces. The purpose of this paper is to introduce yet another innovative approach for face recognition. The human face consists of multiple features that when considered together produces a unique signature that identifies a single person. Building upon this premise, we are studying the identification of faces by producing ratios from the distances between the different features on the face and their locations in an explainable algorithm with the possibility of future inclusion of multiple spectrum and 3D images for data processing and analysis.

Amiri, M, Tofigh, F, Shariati, N, Lipman, J & Abolhasan, M 1970, 'Ultra Wideband Dual Polarization Metamaterial Absorber for 5G frequency spectrum', 2020 14th European Conference on Antennas and Propagation (EuCAP), 2020 14th European Conference on Antennas and Propagation (EuCAP), IEEE, Copenhagen, Denmark.
View/Download from: Publisher's site
View description>>

Implementing 5G technology contributes to improve communication quality and facilitate several interesting applications in daily life such as Internet of things. Despite outstanding features of 5G, the amount of ambient electromagnetic waves will be increased significantly in the environment, which may be undesired. Ultra-wideband metamaterial perfect absorber is a promising solution to collect these undesired signals. Using lumped elements in absorber structure to increase the absorption bandwidth leads to design and fabrication process complexity. In this paper, a low profile polarization angle selective metamaterial absorber has been designed to absorb signals in the frequency range of 21.79 GHz to 53.23 GHz with more than 90% efficiency. The relative absorption bandwidth of the final structure is 83.81%. Moreover, the final structure is reasonably insensitive facing different incident angle up to 40 degree.

Ansari, M, Zhu, H, Shariati, N & Guo, YJ 1970, 'Mm-wave Multi-Beam Antenna Array Based on Miniaturized Butler Matrix for 5G Applications', 2020 IEEE International Symposium on Antennas and Propagation and North American Radio Science Meeting, 2020 IEEE International Symposium on Antennas and Propagation and North American Radio Science Meeting, IEEE.
View/Download from: Publisher's site

Anwar, MJ & Gill, AQ 1970, 'Developing an Integrated ISO 27701 and GDPR based Information Privacy Compliance Requirements Model', ACIS 2020 Proceedings - 31st Australasian Conference on Information Systems, Australasian Conference on Information Systems 2020, Wellington, New Zealand.
View description>>

The protection of information assets requires interdisciplinary approach and cross-functional capabilities. In recent times, information security and privacy compliance continue to be a complicated task due to increasing regulatory restrictions, changing legislations and public awareness. The newly published information security and privacy standard ISO/IEC 27701:2019 provides support for organisations looking to put in place systems to support compliance with global data privacy requirements. However, there is little known about how does this standard map to other regulatory requirements in different jurisdictions specifically the globally relevant General Data Protection Regulation (GDPR). Hence, this research aims to answer an important research question: whether and how the ISO/IEC 27701:2019 framework represents an opportunity for the GDPR compliance? This research provides a review and mapping of ISO/IEC 27701:2019 and GDPR by using an integrated requirement engineering model as a kernel theory. The results of this research will assist organisations contemplating to meet their compliance needs. It will also help academics and practitioners interested in integrating the ISO/IEC 27701:2019 and GDPR for developing relevant compliance frameworks and tools.

Bawden, R, Di Nunzio, GM, Grozea, C, Unanue, IJ, Yepes, AJ, Mah, N, Martinez, D, Névéol, A, Neves, M, Oronoz, M, de Viñaspre, OP, Piccardi, M, Roller, R, Siu, A, Thomas, P, Vezzani, F, Navarro, MV, Wiemann, D & Yeganova, L 1970, 'Findings of the WMT 2020 Biomedical Translation Shared Task: Basque, Italian and Russian as New Additional Languages', 5th Conference on Machine Translation, WMT 2020 - Proceedings, Fifth Conference in Machine Translation (WMT 2020), The Association for Computational Linguistics, Online, pp. 660-687.
View description>>

Machine translation of scientific abstracts and terminologies has the potential to support health professionals and biomedical researchers in some of their activities. In the fifth edition of the WMT Biomedical Task, we addressed a total of eight language pairs. Five language pairs were previously addressed in past editions of the shared task, namely, English/German, English/French, English/Spanish, English/Portuguese, and English/Chinese. Three additional languages pairs were also introduced this year: English/Russian, English/Italian, and English/Basque. The task addressed the evaluation of both scientific abstracts (all language pairs) and terminologies (English/Basque only). We received submissions from a total of 20 teams. For recurring language pairs, we observed an improvement in the translations in terms of automatic scores and qualitative evaluations, compared to previous years.

Betti, F, Ramponi, G & Piccardi, M 1970, 'Controlled Text Generation with Adversarial Learning', INLG 2020 - 13th International Conference on Natural Language Generation, Proceedings, 13th International Conference on Natural Language Generation (INLG 2020), The Association for Computational Linguistics, Dublin, Ireland, pp. 29-34.
View description>>

In recent years, generative adversarial networks (GANs) have started to attain promising results also in natural language generation. However, the existing models have paid limited attention to the semantic coherence of the generated sentences. For this reason, in this paper we propose a novel network - the Controlled TExt generation Relational Memory GAN (CTERM-GAN) - that uses an external input to influence the coherence of sentence generation. The network is composed of three main components: a generator based on a Relational Memory conditioned on the external input; a syntactic discriminator which learns to discriminate between real and generated sentences; and a semantic discriminator which assesses the coherence with the external conditioning. Our experiments on six probing datasets have showed that the model has been able to achieve interesting results, retaining or improving the syntactic quality of the generated sentences while significantly improving their semantic coherence with the given input.

Chaczko, Z, Kulbacki, M, Gudzbeler, G, Alsawwaf, M, Thai-Chyzhykau, I & Wajs-Chaczko, P 1970, 'Exploration of Explainable AI in Context of Human-Machine Interface for the Assistive Driving System', Intelligent Information and Database Systems, Asian Conference on Intelligent Information and Database Systems, Springer International Publishing, Phuket, Thailand, pp. 507-516.
View/Download from: Publisher's site
View description>>

This paper presents the application and issues related to explainable AI in context of a driving assistive system. One of the key functions of the assistive system is to signal potential risks or hazards to the driver in order to allow for prompt actions and timely attention to possible problems occurring on the road. The decision making of an AI component needs to be explainable in order to minimise the time it takes for a driver to decide on whether any action is necessary to avoid the risk of collision or crash. In the explored cases, the autonomous system does not act as a “replacement” for the human driver, instead, its role is to assist the driver to respond to challenging driving situations, possibly difficult manoeuvres or complex road scenarios. The proposed solution validates the XAI approach for the design of a safety and security system that is able to identify and highlight potential risk in autonomous vehicles.

Chemalamarri, VD, Braun, R & Abolhasan, M 1970, 'Constraint-Based Rerouting mechanism to address Congestion in Software Defined Networks', 2020 30th International Telecommunication Networks and Applications Conference (ITNAC), 2020 30th International Telecommunication Networks and Applications Conference (ITNAC), IEEE, Melbourne, VIC, Australia, pp. 1-6.
View/Download from: Publisher's site
View description>>

In this paper, we propose a traffic rerouting mechanism to address congestion in Software-Defined networks. We employ back-tracking and constraint propagation techniques to find alternate paths to reroute multiple active flows simultaneously. Cost function is based on standard deviation of link-loads. We then compare traffic distribution and link utilisation with and without rerouting active flows. We measure and compare network performance using parameters such as total rate of transfer, jitter, and packet loss with that of Shortest Path First with no rerouting. Our proposed solution produces lower jitter, packet drops, and higher transfer rate. We finally conclude the paper by making observations and discussing the scope of the future work.

Chen, S-L, Ziolkowski, RW, Guo, YJ & Liu, Y 1970, 'Single-Feed, Highly-Directive, Higher-Order-Mode Cavity Antenna and Its Beam Tilting Realization', 2020 IEEE Asia-Pacific Microwave Conference (APMC), 2020 IEEE Asia-Pacific Microwave Conference (APMC 2020), IEEE, Hong Kong, Hong Kong, pp. 10-12.
View/Download from: Publisher's site
View description>>

Fast-speed, mast-capacity, and low-cost communications are highly desired for future wireless systems. Single-feed overmoded slot-based rectangular cavity antennas are developed to meet this demand. A TE(10)(11)(0) mode is excited in the cavity with a rectangular waveguide. A total of 110 slots appropriately etched in its top surface yields a system that radiates its beam into the broadside direction with a gain of 26.6 dBi. An engineered phased patch surface is then introduced tofacilitate tilted-beam pattern for high-order-mode cavity antennas. The realized cavity antenna augmented with an appropriately-shaped phased patch surface attained a tilted beam at 30° with respect to the broadside direction. An antenna prototype was fabricated, and measured results agree well with the simulated ones.

Chen, T, Zhang, J, Xie, G-S, Yao, Y, Huang, X & Tang, Z 1970, 'Classification Constrained Discriminator For Domain Adaptive Semantic Segmentation', 2020 IEEE International Conference on Multimedia and Expo (ICME), 2020 IEEE International Conference on Multimedia and Expo (ICME), IEEE, London, United Kingdom, pp. 1-6.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Unsupervised domain adaptation for semantic segmentation aims to transfer knowledge from label-rich synthetic datasets to real-world images without any annotation. The traditional adversarial learning methods for domain adaptation learn to extract domain-invariant feature representations by aligning the feature distributions of both domains. However, these methods suffer from an imbalance in adversarial training and feature distortion. In this work, we propose a classification constrained discriminator to alleviate these problems. Specifically, we first propose to balance the adversarial training by eliminating any pooling layers or strided convolutions in the discriminator. Then, we propose to constrain the discriminator with an auxiliary classification loss to help the feature generator extract the domain-invariant features that are useful for segmentation rather than just ambiguous features to fool the domain discriminator. Extensive experiments demonstrate the superiority of our proposed approach. The source code and models have been made available at https://github.com/NUSTMachine-Intelligence-Laboratory/ccd.

Cheng, Q, Lin, Z, Zhang, JA, Nguyen, D, Huang, X, Kekirigoda, A & Hui, K-P 1970, 'Multi-user MIMO with Jamming Suppression for Spectrum-Efficient Tactical Communications', 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, Adelaide, Australia, pp. 1-6.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Being spectrum-efficient and robust to adversarial interference caused by jammers are critical to tactical wireless systems. Leveraging multiple-input multiple-output (MIMO) techniques, this paper investigates the realization of spectrum-efficient multi-user MIMO communications in the presence of high-power jammers. Unlike most existing work that only exploits the MIMO degree of freedom to nullify the jamming signal, we also aim to improve the spectral efficiency of the system with the MIMO spatial multiplexing capability. To that end, we first design a combiner at the receiver spanning the null space of the jamming channels, which can completely remove the jamming signals and optimize the communication reception. We further propose two methods for the design of precoders at the transmitter to mitigate multi-user interference. Simulation results are presented to verify the effectiveness of the proposed schemes in radio-frequency contested environments.

Cheng, Q, Shi, Z, Nguyen, DN & Dutkiewicz, E 1970, 'Non-cooperative OFDM Spectrum Sensing Using Deep Learning', 2020 International Conference on Computing, Networking and Communications (ICNC), 2020 International Conference on Computing, Networking and Communications (ICNC), IEEE, Hawaii, pp. 704-708.
View/Download from: Publisher's site
View description>>

Although spectrum sensing, a key technique in dynamic spectrum access, has been widely investigated, conventional methods suffer from carrier frequency offset (CFO), timing delay and noise uncertainty, which can significantly degrade the sensing performance. In this paper, we aim to tackle those challenging issues by developing a stacked autoencoder based spectrum sensing approach (SAE-SS). The SAE architecture is employed to effectively learn useful and hidden information from the original received signals. Compared to the existing sensing methods, our approach is more robust to CFO, noise uncertainty and timing delay. Unlike the traditional feature-based detection approaches, the proposed framework does not require the prior knowledge or specific features of incumbent users (IUs). Moreover, in comparison with machine learning based sensing approaches, our solution does not need any external feature extraction algorithms to extract specific features (that is essential for ML-based ones). Through extensive experimental results, our proposed method is demonstrated to achieve notably higher sensing accuracy, e.g., 29% reduced probability of miss detection, than that of state-of-the-art approaches.

Cotton, D, Traish, J & Chaczko, Z 1970, 'Coevolutionary Deep Reinforcement Learning', 2020 IEEE Symposium Series on Computational Intelligence (SSCI), 2020 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Australia, pp. 2600-2607.
View/Download from: Publisher's site
View description>>

The ability to learn without instruction is a powerful enabler for learning systems. A mechanism for this, selfplay, allows reinforcement learning to develop high performing policies without large datasets or expert knowledge. Despite these benefits, self-play is known to be less sample efficient and suffer unstable learning dynamics. This is in part due to a nonstationary learning problem where an agent's actions influence their opponents and as a consequence the training data they receive. In this paper we demonstrate that competitive pressures can be utilised to improve self-play. This paper leverages coevolution, an evolutionary inspired process in which individuals are compelled to innovate and adapt, to optimise the training of a population of reinforcement learning agents. We demonstrate that our algorithm improves the final performance of a Rainbow DQN trained in the game Connect Four, achieving a 15% higher win percentage over the next leading self-play algorithm. Furthermore, our algorithm exhibits more stable training with less variation in evaluation performance.

Ding, C, Sun, H, Zhu, H & Guo, YJ 1970, 'Achieving Wider Impedance Bandwidth Using FullWavelength Dipoles', 2020 14th European Conference on Antennas and Propagation (EuCAP), 2020 14th European Conference on Antennas and Propagation (EuCAP), IEEE, Copenhagen, Denmark.
View/Download from: Publisher's site
View description>>

© 2020 EurAAP. This paper investigates the use of full-wavelength dipoles (FWD) to achieve wider bandwidth than halfwavelength dipoles (HWD). Two dual-polarized antennas are built based on FWDs for base station applications as examples. The first antenna is an isolated cross-dipole employing two FWDs with simple configuration. It is able to cover the lower band for cellular communication from 698 to 960 MHz. The second antenna has four FWDs arranged in a square loop array form and tightly coupled with each other. The employed full-wavelength dipoles are bent upward to maintain a small aperture size, so that the realized element still fits in traditional base station antenna (BSA) array. The antenna can be matched across the band from 1.65 to 3.7 GHz, which can cover both the 3G/4G band from 1.7 to 2.7 GHz and the 5G (sub-6 GHz) band from 3.3 to 3.6 GHz simultaneously. By comparing the attained antennas with comparable antennas based on HWDs, it demonstrates a fact that, when fed properly, FWDs exhibit wider bandwidth than HWDs, and the available methods to improve the bandwidth of HWDs can also be used on FWDs.

Du, A, Pang, S, Huang, X, Zhang, J & Wu, Q 1970, 'Exploring Long-Short-Term Context For Point Cloud Semantic Segmentation', 2020 IEEE International Conference on Image Processing (ICIP), 2020 IEEE International Conference on Image Processing (ICIP), IEEE, pp. 2755-2759.
View/Download from: Publisher's site

Gill, AQ, Beydoun, G, Niazi, M & Khan, HU 1970, 'Adaptive Architecture and Principles for Securing the IoT Systems.', IMIS, International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing, Springer, Lodz, Poland, pp. 173-182.
View/Download from: Publisher's site
View description>>

© Springer Nature Switzerland AG 2021. There is an increasing interest in IoT-enabled smart digital systems. However, it is important to address their security concerns. This paper aims to address this need and proposes an adaptive architecture driven approach to securing IoT systems. The paper proposes IoT security principles and a foundational adaptive architecture framework. These two combined provide a guide to design and embed the security across various layers of an IoT system. This will ensure that the important aspects of the IoT security are not accidentally missed, and thus provides a holistic end to end adaptive architecture driven approach for IoT security. This paper covers the interaction, human, digital technology, physical facility and environment architecture layers and principles related to IoT security as opposed to focusing only on the IoT devices. Thus, it demonstrates and concludes that the IoT security is much more than IoT device, network and perimeter security.

Gong, Y, Li, Z, Zhang, J, Liu, W & Yi, J 1970, 'Potential Passenger Flow Prediction: A Novel Study for Urban Transportation Development', Proceedings of the AAAI Conference on Artificial Intelligence, Thirty-Fourth AAAI Conference on Artificial Intelligence (AAAI-20), Association for the Advancement of Artificial Intelligence (AAAI), New York USA, pp. 4020-4027.
View/Download from: Publisher's site
View description>>

Recently, practical applications for passenger flow prediction have brought many benefits to urban transportation development. With the development of urbanization, a real-world demand from transportation managers is to construct a new metro station in one city area that never planned before. Authorities are interested in the picture of the future volume of commuters before constructing a new station, and estimate how would it affect other areas. In this paper, this specific problem is termed as potential passenger flow (PPF) prediction, which is a novel and important study connected with urban computing and intelligent transportation systems. For example, an accurate PPF predictor can provide invaluable knowledge to designers, such as the advice of station scales and influences on other areas, etc. To address this problem, we propose a multi-view localized correlation learning method. The core idea of our strategy is to learn the passenger flow correlations between the target areas and their localized areas with adaptive-weight. To improve the prediction accuracy, other domain knowledge is involved via a multi-view learning process. We conduct intensive experiments to evaluate the effectiveness of our method with real-world official transportation datasets. The results demonstrate that our method can achieve excellent performance compared with other available baselines. Besides, our method can provide an effective solution to the cold-start problem in the recommender system as well, which proved by its outperformed experimental results.

Gong, Y, Li, Z, Zhang, J, Liu, W, Chen, B & Dong, X 1970, 'A Spatial Missing Value Imputation Method for Multi-view Urban Statistical Data', Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, Twenty-Ninth International Joint Conference on Artificial Intelligence and Seventeenth Pacific Rim International Conference on Artificial Intelligence {IJCAI-PRICAI-20}, International Joint Conferences on Artificial Intelligence Organization, pp. 1310-1316.
View/Download from: Publisher's site
View description>>

Large volumes of urban statistical data with multiple views imply rich knowledge about the development degree of cities. These data present crucial statistics which play an irreplaceable role in the regional analysis and urban computing. In reality, however, the statistical data divided into fine-grained regions usually suffer from missing data problems. Those missing values hide the useful information that may result in a distorted data analysis. Thus, in this paper, we propose a spatial missing data imputation method for multi-view urban statistical data. To address this problem, we exploit an improved spatial multi-kernel clustering method to guide the imputation process cooperating with an adaptive-weight non-negative matrix factorization strategy. Intensive experiments are conducted with other state-of-the-art approaches on six real-world urban statistical datasets. The results not only show the superiority of our method against other comparative methods on different datasets, but also represent a strong generalizability of our model.

Huang, C, Jiang, S, Li, Y, Zhang, Z, Traish, J, Deng, C, Ferguson, S & Da Xu, RY 1970, 'End-to-end Dynamic Matching Network for Multi-view Multi-person 3D Pose Estimation', Computer Vision – ECCV 2020, European Conference on Computer Vision, Springer International Publishing, Glasgow, UK, pp. 477-493.
View/Download from: Publisher's site
View description>>

As an important computer vision task, 3d human pose estimation in a multi-camera, multi-person setting has received widespread attention and many interesting applications have been derived from it. Traditional approaches use a 3d pictorial structure model to handle this task. However, these models suffer from high computation costs and result in low accuracy in joint detection. Recently, especially since the introduction of Deep Neural Networks, one popular approach is to build a pipeline that involves three separate steps: (1) 2d skeleton detection in each camera view, (2) identification of matched 2d skeletons and (3) estimation of the 3d poses. Many existing works operate by feeding the 2d images and camera parameters through the three modules in a cascade fashion. However, all three operations can be highly correlated. For example, the 3d generation results may affect the results of detection in step 1, as does the matching algorithm in step 2. To address this phenomenon, we propose a novel end-to-end training scheme that brings the three separate modules into a single model. However, one outstanding problem of doing so is that the matching algorithm in step 2 appears to disjoint the pipeline. Therefore, we take our inspiration from the recent success in Capsule Networks, in which its Dynamic Routing step is also disjointed, but plays a crucial role in deciding how gradients are flowed from the upper to the lower layers. Similarly, a dynamic matching module in our work also decides the paths in which gradients flow from step 3 to step 1. Furthermore, as a large number of cameras are present, the existing matching algorithm either fails to deliver a robust performance or can be very inefficient. Thus, we additionally propose a novel matching algorithm that can match 2d poses from multiple views efficiently. The algorithm is robust and able to deal with situations of incomplete and false 2d detection as well.

Huang, W, Xu, RYD, Du, W, Zeng, Y & Zhao, Y 1970, 'Mean field theory for deep dropout networks: Digging up gradient backpropagation deeply', Frontiers in Artificial Intelligence and Applications, pp. 1215-1222.
View/Download from: Publisher's site
View description>>

In recent years, the mean field theory has been applied to the study of neural networks and has achieved a great deal of success. The theory has been applied to various neural network structures, including CNNs, RNNs, Residual networks, and Batch normalization. Inevitably, recent work has also covered the use of dropout. The mean field theory shows that the existence of depth scales that limit the maximum depth of signal propagation and gradient backpropagation. However, the gradient backpropagation is derived under the gradient independence assumption that weights used during feed forward are drawn independently from the ones used in backpropagation. This is not how neural networks are trained in a real setting. Instead, the same weights used in a feed-forward step needs to be carried over to its corresponding backpropagation. Using this realistic condition, we perform theoretical computation on linear dropout networks and a series of experiments on dropout networks with different activation functions. Our empirical results show an interesting phenomenon that the length gradients can backpropagate for a single input and a pair of inputs are governed by the same depth scale. Besides, we study the relationship between variance and mean of statistical metrics of the gradient and shown an emergence of universality. Finally, we investigate the maximum trainable length for deep dropout networks through a series of experiments using MNIST and CIFAR10 and provide a more precise empirical formula that describes the trainable length than original work.

Huang, X, Mei, G & Zhang, J 1970, 'Feature-metric Registration: A Fast Semi-supervised Approach for Robust Point Cloud Registration without Correspondences', Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE, pp. 11363-11371.
View/Download from: Publisher's site
View description>>

We present a fast feature-metric point cloud registration framework, whichenforces the optimisation of registration by minimising a feature-metricprojection error without correspondences. The advantage of the feature-metricprojection error is robust to noise, outliers and density difference incontrast to the geometric projection error. Besides, minimising thefeature-metric projection error does not need to search the correspondences sothat the optimisation speed is fast. The principle behind the proposed methodis that the feature difference is smallest if point clouds are aligned verywell. We train the proposed method in a semi-supervised or unsupervisedapproach, which requires limited or no registration label data. Experimentsdemonstrate our method obtains higher accuracy and robustness than thestate-of-the-art methods. Besides, experimental results show that the proposedmethod can handle significant noise and density difference, and solve bothsame-source and cross-source point cloud registration.

Huynh, NV, Hoang, DT, Nguyen, DN, Dutkiewicz, E & Mueck, M 1970, 'Defeating Smart and Reactive Jammers with Unlimited Power', 2020 IEEE Wireless Communications and Networking Conference (WCNC), 2020 IEEE Wireless Communications and Networking Conference (WCNC), IEEE, Seoul.
View/Download from: Publisher's site

Huynh, NV, Nguyen, DN, Hoang, DT, Dutkiewicz, E, Mueck, M & Srikanteswara, S 1970, 'Defeating Jamming Attacks with Ambient Backscatter Communications', 2020 International Conference on Computing, Networking and Communications (ICNC), 2020 International Conference on Computing, Networking and Communications (ICNC), IEEE, Hawaii.
View/Download from: Publisher's site

Jauregi Unanue, I, Esmaili, N, Haffari, G & Piccardi, M 1970, 'Leveraging Discourse Rewards for Document-Level Neural Machine Translation', Proceedings of the 28th International Conference on Computational Linguistics, Proceedings of the 28th International Conference on Computational Linguistics, International Committee on Computational Linguistics, Barcelona, Spain, pp. 4467-4482.
View/Download from: Publisher's site

Keshavarz, R, Miyanaga, Y, Yamamoto, M, Hikage, T & Shariati, N 1970, 'Metamaterial-Inspired Quad-Band Notch Filter for LTE Band Receivers and WPT Applications', 2020 XXXIIIrd General Assembly and Scientific Symposium of the International Union of Radio Science, 2020 XXXIIIrd General Assembly and Scientific Symposium of the International Union of Radio Science (URSI GASS), IEEE.
View/Download from: Publisher's site

Khoa, TV, Saputra, YM, Hoang, DT, Trung, NL, Nguyen, D, Ha, NV & Dutkiewicz, E 1970, 'Collaborative Learning Model for Cyberattack Detection Systems in IoT Industry 4.0', 2020 IEEE Wireless Communications and Networking Conference (WCNC), 2020 IEEE Wireless Communications and Networking Conference (WCNC), IEEE, Seoul.
View/Download from: Publisher's site

Kieu, T-B, Pham, SB, Phan, X-H & Piccardi, M 1970, 'A Submodular Approach for Reference Recommendation', Communications in Computer and Information Science, International Conference of the Pacific Association for Computational Linguistics, Springer Singapore, Hanoi, Vietnam, pp. 3-14.
View/Download from: Publisher's site
View description>>

© 2020, Springer Nature Singapore Pte Ltd. Choosing appropriate references for a given topic is an important, yet challenging task. The pool of potential candidates is typically very large, in the order of tens of thousands, and growing by the day. For this reason, this paper proposes an approach for automatically providing a reference list for a given manuscript. The approach is based on an original submodular inference function which balances relevance, coverage and diversity in the reference list. Experiments are carried out using an ACL corpus as a source for the references and evaluated by MAP, MRR and precision-recall. The results show the remarkable comparative performance of the proposed approach.

Le, AT, Tran, LC, Huang, X, Ritz, C, Dutkiewicz, E, Bouzerdoum, A & Franklin, DR 1970, 'Hybrid TOA/AOA Localization with 1D Angle Estimation in UAV-assisted WSN.', ICSPCS, 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, pp. 1-6.
View/Download from: Publisher's site
View description>>

Unmanned aerial vehicles (UAVs) are considered as a great solution for a flexible and rapid deployment of wireless sensor networks (WSN) in emergency scenarios. Hybrid time-of-arrival (TOA) and angle-of-arrival (AOA) localization is widely used to estimate agents’ positions in WSN. Conventional TOA/AOA localization methods normally require both elevation and azimuth AOA estimations to estimate agents’ positions, leading to complicated L-shape antenna arrays and power-thirsty two-dimensional signal processing at the agents. We propose a hybrid TOA/1AOA localization approach which only requires elevation AOA estimations to combine with TOA measurements. A weighted least square algorithm is proposed to solve the non-linear problem. The performance of the proposed method is compared with that of the conventional approach under various scenarios. Simulation results show that, by adjusting different parameters such as transmit power, signal bandwidth, and the number of anchors, the proposed method outperforms the conventional counterpart while significantly reduces the complexity of the agents.

Li, Y, Li, K, Jiang, S, Zhang, Z, Huang, C & Da Xu, RY 1970, 'Geometry-driven self-supervised method for 3D human pose estimation', AAAI 2020 - 34th AAAI Conference on Artificial Intelligence, pp. 11442-11449.
View description>>

The neural network based approach for 3D human pose estimation from monocular images has attracted growing interest. However, annotating 3D poses is a labor-intensive and expensive process. In this paper, we propose a novel self-supervised approach to avoid the need of manual annotations. Different from existing weakly/self-supervised methods that require extra unpaired 3D ground-truth data to alleviate the depth ambiguity problem, our method trains the network only relying on geometric knowledge without any additional 3D pose annotations. The proposed method follows the two-stage pipeline: 2D pose estimation and 2D-to-3D pose lifting. We design the transform re-projection loss that is an effective way to explore multi-view consistency for training the 2D-to-3D lifting network. Besides, we adopt the confidences of 2D joints to integrate losses from different views to alleviate the influence of noises caused by the self-occlusion problem. Finally, we design a two-branch training architecture, which helps to preserve the scale information of re-projected 2D poses during training, resulting in accurate 3D pose predictions. We demonstrate the effectiveness of our method on two popular 3D human pose datasets, Human3.6M and MPI-INF-3DHP. The results show that our method significantly outperforms recent weakly/self-supervised approaches.

Li, Z, Zhang, J, Gong, Y, Yao, Y & Wu, Q 1970, 'Field-wise learning for multi-field categorical data', Advances in Neural Information Processing Systems, Conference on Neural Information Processing Systems, On-line.
View description>>

We propose a new method for learning with multi-field categorical data. Multi-field categorical data are usually collected over many heterogeneous groups. These groups can reflect in the categories under a field. The existing methods try to learn a universal model that fits all data, which is challenging and inevitably results in learning a complex model. In contrast, we propose a field-wise learning method leveraging the natural structure of data to learn simple yet efficient one-to-one field-focused models with appropriate constraints. In doing this, the models can be fitted to each category and thus can better capture the underlying differences in data. We present a model that utilizes linear models with variance and low-rank constraints, to help it generalize better and reduce the number of parameters. The model is also interpretable in a field-wise manner. As the dimensionality of multi-field categorical data can be very high, the models applied to such data are mostly over-parameterized. Our theoretical analysis can potentially explain the effect of over-parametrization on the generalization of our model. It also supports the variance constraints in the learning objective. The experiment results on two large-scale datasets show the superior performance of our model, the trend of the generalization error bound, and the interpretability of learning outcomes. Our code is available at https://github.com/lzb5600/Field-wise-Learning.

Lian, J-W, Ban, Y-L, Zhu, H & Guo, YJ 1970, 'Uniplanar 2-D Butler Matrix for Multibeam Arrays', 2020 4th Australian Microwave Symposium (AMS), 2020 4th Australian Microwave Symposium (AMS), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>

A 2-D Butler matrix (BM) in uniplanar configuration for designing multibeam array antenna (MAA) is proposed using substrate integrated waveguide (SIW) technology. Firstly, a novel topology for building uniplanar 2D BM is proposed, which successfully transforms the traditional 3-D topology to a 2-D (or uniplanar) one. To realize the planarization of basic components, a novel design of eight-port hybrid couplers, is developed to transform four spatially intersected couplers to a planar structure. To address the issue of excessive path intersections, a novel SIW eight-port crossover is proposed to reduce the number of path intersections from 16 to merely 4. Using this proposed 2-D BM, a 2-D MAA with 16 (4 × 4) beams can be realized.

Lin, Z, Lv, T, Zhang, JA & Liu, RP 1970, 'Tensor-based High-Accuracy Position Estimation for 5G mmWave Massive MIMO Systems', ICC 2020 - 2020 IEEE International Conference on Communications (ICC), ICC 2020 - 2020 IEEE International Conference on Communications (ICC), IEEE, Dublin, Ireland, pp. 1-6.
View/Download from: Publisher's site
View description>>

Highly accurate localization is important for wire-less communications. In this paper, we propose a new tensor-based positioning method for 5G wideband mmWave massive MIMO systems. We first develop an extended multidimensional interpolation (E-MI)-based method as the preprocessing step to suppress the frequency-dependence of the array steering vectors. By using this method, the data across the whole frequency band can be processed jointly, and the high temporal resolution offered by wideband mmWave signals can be exploited. Then, we propose a parameter decoupling (PD)-based tensor multiparameter estimation algorithm. This algorithm can suppress the noises in all of temporal, spatial and frequency domains, and thus all the parameters can be precisely estimated. A simplified perturbation term (S-PT)-based method is also presented to match the estimated parameters at low complexity. Based on the quasi-optical property of mmWave signals, we propose a novel method to compute the 3D coordinates of the target. Simulation results demonstrate the effectiveness of the proposed positioning method in the end.

Liu, J, Zhang, JA, Xu, R, Pearce, A, Ni, W & Hedley, M 1970, 'Gaussian Mixture Model based Convolutional Sparse Coding for Radar Heartbeat Detection', 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, Adelaide, Australia, pp. 1-6.
View/Download from: Publisher's site
View description>>

Accurate detection of heartbeat through radar has many potential applications in, e.g., security and health. However, it is generally challenging to obtain clear heart-beat signature, due to its weak signal and relatively large interference caused by, e.g., body and respiration movement. In this paper, we propose an advanced algorithm based on convolutional sparse coding (CSC) and Gaussian mixture model (GMM) for suppressing the interference and extracting clear heartbeat signals. In this study, heartbeat signals are modelled by CSC and recovered by exploiting the sparsity of the signal. GMM is introduced to model the unknown noise, which could be a mixture from multiple noise/interference sources. The parameters of GMM, dictionary and codes are computed via the expectation maximization (EM) algorithm. To achieve faster processing, convolution computing is proposed to be processed in the frequency domain. The proposed method is tested and validated by simulation and experiments. The results show that our proposed algorithm can accurately extract the heartbeat components.

Lyu, B, Hoang, DT, Gong, S & Yang, Z 1970, 'Intelligent Reflecting Surface Assisted Wireless Powered Communication Networks', 2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW), 2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW), IEEE, Seoul, Korea (South), pp. 1-6.
View/Download from: Publisher's site
View description>>

In this paper, we propose to use an intelligent reflecting surface (IRS) in a wireless powered communication network to boost both downlink energy transfer (ET) and uplink information transmission (IT) efficiency, where the IRS consisting of a large number of low-cost passive reflecting elements is deployed between a hybrid access point (HAP) and multiple wireless-powered users. In particular, all passive reflecting elements collaboratively adjust their phase shifts to first construct beamforming for ET from the HAP to all users and then provide additional transmission links for IT from the users to the HAP. Then, we formulate a sum-rate maximization problem by jointly optimizing the time scheduling for network, the phase shift matrix for ET, and the phase shift matrices for all users' IT. Since the formulated problem is non-convex, we first design the phase shift matrices for IT independently by exploiting the characteristics of IT and obtain an approximate solution by using the semidefinite relaxation technique and the Gaussian randomization method. After that, we propose a block coordinated decent based algorithm to solve the simplified problem by iteratively optimizing the time scheduling and the ET's phase shift matrix, the convergence of which is further analyzed. Simulation results confirm that the proposed scheme can achieve up to 350% sum- rate gain compared to the benchmarks.

Madhuri, M, Gill, AQ & Khan, HU 1970, 'IoT-Enabled Smart Child Safety Digital System Architecture.', ICSC, 2020 IEEE 14th International Conference on Semantic Computing, IEEE, San Diego, CA, USA, pp. 166-169.
View/Download from: Publisher's site
View description>>

Safety of a child in a large public event is a major concern for event organizers and parents. This paper addresses this important concern and proposes an architecture model of the IoT-enable smart child safety tracking digital system. This IoT-enabled digital system architecture integrates the Cloud, Mobile and GPS technology to precisely locate the geographical location of a child on an event map. The proposed architecture model describes the people, information, process, and technology architecture elements, and their relationships for the complex IoT-enable smart child safety tracking digital system. The proposed architecture model can be used as a reference or guide to assist in the safe architecture driven development of the various child tracking digital systems for different public events.

Makhdoom, I, Tofigh, F, Zhou, I, Abolhasan, M & Lipman, J 1970, 'PLEDGE: A Proof-of-Honesty based Consensus Protocol for Blockchain-based IoT Systems', 2020 IEEE International Conference on Blockchain and Cryptocurrency (ICBC), 2020 IEEE International Conference on Blockchain and Cryptocurrency (ICBC), IEEE, Toronto, ON, Canada, pp. 1-3.
View/Download from: Publisher's site
View description>>

Exhibition of malicious behavior during blockchain consensus, threats against reputation systems, and high TX latency are significant issues for blockchain-based IoT systems. Hence, to mitigate such challenges we propose 'Pledge', a unique Proof-of-Honesty based consensus protocol. Initial experimentation shows that Pledge is economical with low computations and communications complexity and low latency in transaction confirmation.

Makhdoom, I, Tofigh, F, Zhou, I, Abolhasan, M & Lipman, J 1970, 'PLEDGE: An IoT-oriented Proof-of-Honesty based Blockchain Consensus Protocol', 2020 IEEE 45th Conference on Local Computer Networks (LCN), 2020 IEEE 45th Conference on Local Computer Networks (LCN), IEEE, Australia, pp. 54-64.
View/Download from: Publisher's site
View description>>

The existing lottery-based consensus algorithms, such as Proof-of-Work, and Proof-of-Stake, are mostly used for blockchain-based financial technology applications. Similarly, the Byzantine Fault Tolerance algorithms do provide consensus finality, yet they are either communications intensive, vulnerable to Denial-of-Service attacks, poorly scalable, or have a low faulty node tolerance level. Moreover, these algorithms are not designed for the Internet of Things systems that require near-real-time transaction confirmation, maximum fault tolerance, and appropriate transaction validation rules. Hence, we propose 'Pledge, 'a unique Proof-of-Honesty based consensus protocol to reduce the possibility of malicious behavior during blockchain consensus. Pledge also introduces the Internet of Things centric transaction validation rules. Initial experimentation shows that Pledge is economical and secure with low communications complexity and low latency in transaction confirmation.

Marsh, L, Cochrane, M, Lodge, R, Sims, B, Traish, J & Xu, R 1970, 'Autonomous Target Allocation Recommendations', 2020 IEEE Symposium Series on Computational Intelligence (SSCI), 2020 IEEE Symposium Series on Computational Intelligence (SSCI), IEEE, Australia, pp. 1403-1410.
View/Download from: Publisher's site
View description>>

We consider the problem of land vehicles under attack from a number of unmanned aerial systems. As the number of unmanned aerial systems increase, it may become difficult for human operators to coordinate actions across vehicles in a timely manner. In this paper, we study a number of algorithms designed to recommend actions to operators that will maximise the survivability of the vehicle fleet. We present a comparison of several assignment approaches including evolutionary strategies, genetic algorithms, multi-armed bandits, probability trees and basic heuristics. The performance of these algorithms is analysed across six different simulated scenarios. Our findings indicate that while there was no single best approach, Evolution Strategies, Ensemble and Genetic Algorithms were the strongest performers. It was also seen that a number of heuristic algorithms and the multi-armed bandits approach offered reliable performance in a number of scenarios without the need for any training.

More, FJ, Chaczko, Z & Kulbacka, J 1970, 'Early Detection of Coronary Artery Diseases Using Endocrine Markers', Intelligent Information and Database Systems, Asian Conference on Intelligent Information and Database Systems, Springer International Publishing, Phuket, Thailand, pp. 593-601.
View/Download from: Publisher's site
View description>>

© 2020, Springer Nature Switzerland AG. Cardiovascular diseases including coronary artery disease is the leading cause of death in the well developed and developing countries of the 21st century and has a higher rate of mortality and morbidity. Dysfunction of the pituitary, thyroid, and parathyroid glands caused cardio/cardiovascular diseases including changes in blood pressure, contractility of myocardium - systolic and diastolic myocardial functions, endothelial and dyslipidemia. Dysfunction of thyroid, parathyroid and adrenocorticotropic hormones caused imbalance of endocrine system such as hyper and hypo function, effects on pathophysiology of the cardiovascular system.

Nguyen, CT, Nguyen, DN, Hoang, DT, Pham, H-A, Tuong, NH & Dutkiewicz, E 1970, 'Blockchain and Stackelberg Game Model for Roaming Fraud Prevention and Profit Maximization', 2020 IEEE Wireless Communications and Networking Conference (WCNC), 2020 IEEE Wireless Communications and Networking Conference (WCNC), IEEE, Seoul.
View/Download from: Publisher's site

Nguyen, D & Meixner, G 1970, 'A survey of gamified Augmented Reality systems for procedural tasks in industrial settings', IFAC-PapersOnLine, Elsevier BV, pp. 10096-10100.
View/Download from: Publisher's site

Nguyen, N-T, Nguyen, DN, Hoang, DT, Van Huynh, N, Nguyen, H-N, Nguyen, QT & Dutkiewicz, E 1970, 'Energy Trading and Time Scheduling for Energy-Efficient Heterogeneous Low-Power IoT Networks', GLOBECOM 2020 - 2020 IEEE Global Communications Conference, GLOBECOM 2020 - 2020 IEEE Global Communications Conference, IEEE, Taipei, Taiwan.
View/Download from: Publisher's site
View description>>

In this paper, an economic model is proposed to jointly optimize profits for participants in a heterogeneous IoT wireless-powered backscatter communication network. In the network under considerations, a power beacon and IoT devices (with various communication types and energy constraints) are assumed to belong to different service providers, i.e., energy service provider (ESP) and IoT service provider (ISP), respectively. To jointly maximize the utility for both service providers in terms of energy efficiency and network throughput, a Stackelberg game model is proposed to study the strategic interaction between the ISP and ESP. In particular, the ISP first evaluates its benefits from providing IoT services to its customers and then sends its requested price together with the service time to the ESP. Based on the request from the ISP, the ESP offers an optimized transmission power that maximizes its utility while meeting energy demands of the ISP. To study the Stackelberg equilibrium, we first obtain a closed-form solution for the ESP and propose a low-complexity iterative method based on block coordinate descent (BCD) to address the non-convex optimization problem for the ISP. Through simulation results, we show that our approach can significantly improve the profits for both providers compared with those of conventional transmission methods, e.g., bistatic backscatter and harvest-then-transmit communication methods.

Nguyen, TG, Phan, TV, Hoang, DT, Nguyen, TN & So-In, C 1970, 'Efficient SDN-Based Traffic Monitoring in IoT Networks with Double Deep Q-Network', Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), International Conference on Computational Data and Social Networks, Springer International Publishing, Dallas, TX, USA, pp. 26-38.
View/Download from: Publisher's site
View description>>

In an Internet of Things (IoT) environment, network traffic monitoring tasks are intractable to achieve due to various IoT traffic types. Recently, the development of Software-Defined Networking (SDN) enables outstanding flexibility and scalability abilities in network control and management, thereby providing a potential approach to mitigate challenges in monitoring the IoT traffic. In this paper, we propose an IoT traffic monitoring approach that implements deep reinforcement learning technique to maximize the fine-grained monitoring capability, i.e., level of traffic statistics details, for several IoT traffic groups. Specifically, we first study a flow-rule matching control system constrained by different expected levels of statistics details and by the flow-table limit of the SDN-based gateway device. We then formulate our control optimization problem by employing the Markov decision process (MDP). Afterwards, we develop Double Deep Q-Network (DDQN) algorithm to quickly obtain the optimal flow-rule matching control policy. Through the extensive experiments, the obtained results verify that the proposed approach yields outstanding improvements in terms of the ability to simultaneously provide different required degrees of statistics details while protecting the gateway devices from being overflowed in comparisons with those of the conventional Q-learning method and the typical SDN flow rule setting.

Ni, Z, Zhang, JA, Huang, X, Yang, K & Gao, F 1970, 'Parameter Estimation and Signal Optimization for Joint Communication and Radar Sensing', 2020 IEEE International Conference on Communications Workshops (ICC Workshops), 2020 IEEE International Conference on Communications Workshops (ICC Workshops), IEEE, Dublin, Ireland.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Joint communication and radar sensing (JCAS) integrates communication and radar sensing into one system, sharing one transmitted signal. In this paper, we study a JCAS system that uses a dedicated low-cost single-antenna receiver for sensing. We provide sensing parameter estimation algorithms for the JCAS system, and investigate the optimization of the precoding matrix to balance communication and sensing performance. A MUSIC-based estimation approach is proposed to obtain time delays and angle-of-arrivals of targets. A weighted signal optimization to balance between communication and sensing is then proposed. Numerical results are provided and verify the effectiveness of the proposed scheme.

Raza, MA, Abolhasan, M, Lipman, J, Shariati, N & Ni, W 1970, 'Statistical Learning-Based Dynamic Retransmission Mechanism for Mission Critical Communication: An Edge-Computing Approach', 2020 IEEE 45th Conference on Local Computer Networks (LCN), 2020 IEEE 45th Conference on Local Computer Networks (LCN), IEEE, Australia, pp. 393-396.
View/Download from: Publisher's site
View description>>

Mission-critical machine type communication (MC-MTC) systems in which machines communicate to perform various tasks such as coordination, sensing, and actuation, require stringent requirements of ultra-reliable and low latency communications (URLLC). Edge computing being an integral part of future wireless networks, provides services that support URLLC applications. In this paper, we use the edge computing approach and present a statistical learning-based dynamic retransmission mechanism. The proposed approach meets the desired latency-reliability criterion in MC-MTC networks employing framed ALOHA. The maximum number of retransmissions Nr under a given latency-reliability constraint is learned statistically by the devices from the history of their previous transmissions and shared with the base station. Simulations are performed in MATLAB to evaluate a framed-ALOHA system's performance in which an active device can have only one successful transmission in one round composed of (Nr + 1) frames, and the performance is compared with the diversity transmission-based framed-ALOHA.

Saputra, YM, Nguyen, DN, Hoangl, DT, Dutkiewicz, E & Mueck, MD 1970, 'Common Agency-Based Economic Model for Energy Contract in Electric Vehicle Networks', GLOBECOM 2020 - 2020 IEEE Global Communications Conference, GLOBECOM 2020 - 2020 IEEE Global Communications Conference, IEEE, Taipei, Taiwan, pp. 1-6.
View/Download from: Publisher's site
View description>>

The rapid adoption of electric or hybrid vehicles (EVs) has called for wide deployment of charging stations. These stations can be launched/owned by different owners, referred to as charging station providers (CSPs), which make energy contracts with a smart grid provider (SGP). However, there exists a shortage of mutual economic strategy between the SGP and CSPs in an energy request/transfer competition due to the selfish nature among them. In this paper, we propose an economic model leveraging a multi-principal single-agent (referred to as common agency) contract policy, aiming at maximizing the utilities of multiple CSPs while optimizing the utility of the SGP in an EV network. In particular, we first develop the common agency-based contract problem as a non-cooperative energy contract optimization problem, in which each CSP can maximize its utility given the common constraints from the SGP and the contracts of other CSPs. To deal with this problem, we develop an iterative energy contract algorithm to find an equilibrium contract solution where the contracts from the CSPs can produce maximum utilities of the CSPs and satisfy the constraints of the SGP. Through numerical results, we show that our proposed model can improve the social welfare of the EV network up to 54% and the utilities of CSPs up to 60% compared with the baseline method in which each CSP obtains the amount of energy that is proportional to its energy request.

Sarathy, N, Alsawwaf, M & Chaczko, Z 1970, 'Investigation of an Innovative Approach for Identifying Human Face-Profile Using Explainable Artificial Intelligence', 2020 IEEE 18th International Symposium on Intelligent Systems and Informatics (SISY), 2020 IEEE 18th International Symposium on Intelligent Systems and Informatics (SISY), IEEE, Serbia, pp. 155-160.
View/Download from: Publisher's site
View description>>

Human identification is a well-researched topic that keeps evolving. Advancement in technology has made it easy to train models or use ones that have been already created to detect several features of the human face. When it comes to identifying a human face from the side, there are many opportunities to advance the biometric identification research further. This paper investigates the human face identification based on their side profile by extracting the facial features and diagnosing the feature sets with geometric ratio expressions. These geometric ratio expressions are computed into feature vectors. The last stage involves the use of weighted means to measure similarity. This research addresses the problem of using an eXplainable Artificial Intelligence (XAI) approach. Findings from this research, based on a small data-set, conclude that the used approach offers encouraging results. Further investigation could have a significant impact on how face profiles can be identified. Performance of the proposed system is validated using metrics such as Precision, False Acceptance Rate, False Rejection Rate and True Positive Rate. Multiple simulations indicate an Equal Error Rate of 0.89.

Shi, Q, Wu, N, Wang, H, Nguyen, DN & Huang, X 1970, 'Joint Phase Noise Estimation and Decoding in OFDM-IM', GLOBECOM 2020 - 2020 IEEE Global Communications Conference, GLOBECOM 2020 - 2020 IEEE Global Communications Conference, IEEE, Taipei, Taiwan, pp. 1-6.
View/Download from: Publisher's site
View description>>

This paper proposes a low-complexity joint phase noise (PHN) estimation and decoding algorithm for orthogonal frequency division multiplexing relying on index modulation (OFDM-IM) systems. A factor graph (FG) is constructed based on the truncated discrete cosine transform (DCT) expansion model for the variation of PHN. In order to explicitly take into account the structured and sparse a priori information of the frequency-domain symbols provided by the soft-in soft-out (SISO) decoder, the generalized approximate message passing (GAMP) algorithm is employed. Furthermore, to solve the unknown and nonlinear transform matrix problem introduced by the PHN, the mean-field (MF) method is invoked at the observation nodes on the FG. Monte Carlo simulations show the superiority of the proposed algorithm over the existing variational inference (VI) and extended Kalman filter (EKF) methods in terms of their bit error rate (BER) performance and complexity. In addition, we demonstrate that the OFDM-IM scheme outperforms its conventional OFDM counterpart in the presence of PHN.

Shi, Z, Zhang, JA, Xu, R, Cheng, Q & Pearce, A 1970, 'Towards Environment-Independent Human Activity Recognition using Deep Learning and Enhanced CSI', GLOBECOM 2020 - 2020 IEEE Global Communications Conference, GLOBECOM 2020 - 2020 IEEE Global Communications Conference, IEEE, Taipei, Taiwan, pp. 1-6.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Deep learning has shown a strong potential in device-free human activity recognition (HAR). However, a fundamental challenge is ensuring accuracy, without re-training, when exposing a previously trained architecture to a new or unseen environment. To overcome the aforementioned challenge, this paper proposes an environment-robust channel state information (CSI) based HAR by leveraging the properties of a matching network (MatNet) and enhanced features (HAR-MN-EF). To improve the CSI quality, we propose a CSI cleaning and enhancement method (CSI-CE) that includes two key stages: activity-related information extraction (ARIE) and correlation feature extraction based on principal component analysis (CFE-PCA). The ARIE stage is able to effectively enhance the activity-dependent features whilst mitigating behavior-unrelated information. The CFE-PCA stage further improves the extracted features by filtering out the residual activity-unrelated data and the residual noise contained in signals from the former stage. The extracted features are then sequenced into the MatNet to create an environment-robust HAR. Experimental results confirm that an architecture trained by the proposed HAR-MN-EF can be directly adapted to a new environment, achieving reliable sensing accuracies without requiring additional effort.

Shi, Z, Zhang, JA, Xu, RY & Cheng, Q 1970, 'WiFi-Based Activity Recognition using Activity Filter and Enhanced Correlation with Deep Learning', 2020 IEEE International Conference on Communications Workshops (ICC Workshops), 2020 IEEE International Conference on Communications Workshops (ICC Workshops), IEEE, Dublin, Ireland, pp. 1-6.
View/Download from: Publisher's site
View description>>

Device-free WiFi sensing utilizing channel state information (CSI) is attractive for human activity recognition (HAR). However, several challenging problems are yet to be resolved, e.g., difficulty in extracting proper features from input signals, susceptibility to the phase shift of CSI and difficulty in identifying similar behaviors (e.g., lying and standing). In this paper, we aim to tackle these problems by proposing a novel scheme for CSI-based HAR that uses activity filter-based deep learning network (HAR-AF-DLN) with enhanced correlation features. We first develop a novel CSI compensation and enhancement (CCE) method to compensate for the timing offset between the WiFi transmitter and receiver, enhance activity-related signals and reduce the dimension of inputs to DLN. Then, we design a novel activity filter (AF) to differentiate similar activities (e.g., standing and lying) based on the enhanced CSI correlation features obtained from CCE. Extensive simulation results demonstrate that our proposed HAR-AF-DLN scheme outperforms state-of-the-art methods with significantly improved recognition accuracy (especially for similar activities) and notably reduced training time.

Song, B, Jing, Z, Jay Guo, Y, Liu, RP & Zhou, Q 1970, 'A Novel Measure to Quantify the Robustness of Social Network Under the Virus Attacks', Springer Singapore, pp. 189-200.
View/Download from: Publisher's site

Song, L-Z, Qin, P-Y, Liu, Y-H & Guo, YJ 1970, 'Dual-Layer Huygens Element Based Conformal Transmitarray With A High-Efficiency', 2020 IEEE Asia-Pacific Microwave Conference (APMC), 2020 IEEE Asia-Pacific Microwave Conference (APMC 2020), IEEE.
View/Download from: Publisher's site

Sun, H-H, Ding, C, Zhu, H & Guo, YJ 1970, 'A Method for Bandwidth Enhancement of Cross-Dipole Antennas with Compact Configurations', 2020 IEEE International Symposium on Antennas and Propagation and North American Radio Science Meeting, 2020 IEEE International Symposium on Antennas and Propagation and North American Radio Science Meeting, IEEE, Montreal, QC, Canada, pp. 571-572.
View/Download from: Publisher's site
View description>>

In this paper, a new method to broaden the bandwidth of dual-polarized cross-dipole antennas is presented. By connecting a thin loop to a traditional cross-dipole, additional resonant points are introduced and the bandwidth is broadened. This method does not increase the physical dimension of the antenna and has little influence on radiation performances. A loop-connected cross-dipole antenna is presented to verify the method. The bandwidth it achieves is 66.7% from 1.65 GHz to 3.30 GHz with a very compact radiator size. The antenna also has a high port isolation level and stable radiation performances, making it highly suitable for the base station application.

Sun, X, Jiang, Y & Li, W 1970, 'Residual Attention Based Network for Automatic Classification of Phonation Modes', 2020 IEEE International Conference on Multimedia and Expo (ICME), 2020 IEEE International Conference on Multimedia and Expo (ICME), IEEE, pp. 1-6.
View/Download from: Publisher's site

Sun, Z, Hua, X-S, Yao, Y, Wei, X-S, Hu, G & Zhang, J 1970, 'CRSSC: Salvage Reusable Samples from Noisy Data for Robust Learning', Proceedings of the 28th ACM International Conference on Multimedia, MM '20: The 28th ACM International Conference on Multimedia, ACM, pp. 92-101.
View/Download from: Publisher's site

Van Huynh, N, Nguyen, DN, Hoang, DT & Dutkiewicz, E 1970, 'Optimal Beam Association in mmWave Vehicular Networks with Parallel Reinforcement Learning', GLOBECOM 2020 - 2020 IEEE Global Communications Conference, GLOBECOM 2020 - 2020 IEEE Global Communications Conference, IEEE, Taipei, Taiwan, pp. 1-6.
View/Download from: Publisher's site
View description>>

This paper develops a beam association framework for mm Wave vehicular networks to improve the system performance in terms of handover, disconnection time, and data rate under the high mobility of vehicles. In particular, we recruit the semi Markov decision process to capture the uncertainty and dynamic of the environment such as locations of beams, received signal strength indicator profiles, velocities, and blockages. Instead of adopting complex deep learning structures such as deep dueling and double deep Q-learning, we develop a lightweight yet very effective parallel Q-learning algorithm to quickly derive the optimal beam association policy by simultaneously learning from various vehicles on the road. Through extensive simulation results, we demonstrate that the proposed framework can reduce the average disconnection time by 33% and increase the data rate by 60% compared to other solutions. We also observed that the proposed parallel Q-learning algorithm converges much faster to the optimal solution than state-of-the-art deep-learning based algorithms.

Wang, Z, Xu, M, Ye, N, Huang, H, Wang, R & Xiao, F 1970, 'RF-Mirror: Mitigating Mutual Coupling Interference in Two-Tag Array Labeled RFID Systems', 2020 17th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON), 2020 17th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON), IEEE.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Recent RFID systems start attaching a tag array consisting of two or more tags on an object to deal with polarization mismatch and RF phase periodicity for battery-free sensing and localization. The multi-tag solution can also provide target orientation estimation. However, when these tags are closely spaced apart, mutual coupling will be induced, producing the unexpected changes in reported RSSI and RF phase. In this paper, we present RF-Mirror that enables compensating the distortion in a two-tag array labeled RFID system. The system would output the accurate difference in tag-to-antenna distances between two tags, which is a fundamental parameter in previous works for use. Firstly, we model the backscatter signal of a responding tag in a two-tag scenario, and then formulate novel RSSI- and RF phase-distance models with coupling terms. Secondly, we design an algorithm to characterize the coupling effect on tag gain by fusing RSSI and RF phase. Thirdly, we design a decoupling algorithm based on an observation that tag mutual coupling is independent of the position of a tag array relative to a reader antenna. Our experiments show the effectiveness of our models and RF-Mirror achieves the decoupling error of 0.197 cm in calculating the tag-to-antenna distance difference.

Wu, K, Guo, YJ, Huang, X & Heath, RW 1970, 'Accurate Channel Estimation for Frequency-Hopping Dual-Function Radar Communications', 2020 IEEE International Conference on Communications Workshops (ICC Workshops), 2020 IEEE International Conference on Communications Workshops (ICC Workshops), IEEE, Dublin, Ireland, pp. 1-6.
View/Download from: Publisher's site
View description>>

Dual-function radar communications (DFRC) is proposed recently to embed information into radar waveform, and hence performs data communications by sharing radar apertures and frequency resources. Exploiting a frequency-hopping (FH) MIMO radar, DFRC can achieve the symbol rate that is larger than the radar pulse frequency. However, this requires an accurate channel estimate, which is challenging to achieve due to the radar-prioritized transmission and the fast-changing FH waveform. In this paper, we propose an accurate channel estimation method for the DFRC based on FH-MIMO radars. We design a new FH-MIMO radar waveform which incurs no change to the ranging performance of the radar. The new waveform also enables a communication receiver to estimate the channel without knowing the pairing between hopping frequencies and antennas. We also develop a new angle estimation method at a single-antenna communication receiver using as few as one symbol, i.e., a single hop. Simulations are provided to validate the efficacy of the proposed channel estimation method. Specifically, the symbol error rate achieved based on the estimated channel approaches that based on the ideal channel.

Xie, H-B, Li, C, Mengersen, K, Wang, S & Xu, RYD 1970, 'Nonparametric Bayesian Nonnegative Matrix Factorization', Modeling Decisions for Artificial Intelligence, International Conference on Modeling Decisions for Artificial Intelligence, Springer International Publishing, Sant Cugat, Spain, pp. 132-141.
View/Download from: Publisher's site
View description>>

© Springer Nature Switzerland AG 2020. Nonnegative Matrix Factorization (NMF) is an important tool in machine learning for blind source separation and latent factor extraction. Most of existing NMF algorithms assume a specific noise kernel, which is insufficient to deal with complex noise in real scenarios. In this study, we present a hierarchical nonparametric nonnegative matrix factorization (NPNMF) model in which the Gaussian mixture model is used to approximate the complex noise distribution. The model is cast in the nonparametric Bayesian framework by using Dirichlet process mixture to infer the necessary number of Gaussian components. We derive a mean-field variational inference algorithm for the proposed nonparametric Bayesian model. Experimental results on both synthetic data and electroencephalogram (EEG) demonstrate that NPNMF performs better in extracting the latent nonnegative factors in comparison with state-of-the-art methods.

Xu, J, Yu, L, Zhang, J & Wu, Q 1970, 'Automatic Sheep Counting by Multi-object Tracking', 2020 IEEE International Conference on Visual Communications and Image Processing (VCIP), 2020 IEEE International Conference on Visual Communications and Image Processing (VCIP), IEEE, China, pp. 257-257.
View/Download from: Publisher's site
View description>>

Animal counting is a highly skilled yet tedious task in livestock transportation and trading. To effectively free up the human labour and provide accurate counts for sheep loading/unloading, we develop an auto sheep counting system based on multi-object detection, tracking and extrapolation techniques. Our system has demonstrated more than 99.9% accuracy with sheep moving freely in a race under optimal visual conditions.

Yang, T, Ding, C, Ziolkowski, RW & Guo, YJ 1970, 'Achieving a Terahertz Photonic Crystal Fiber with Enhanced Birefrigence', 2020 4th Australian Microwave Symposium (AMS), 2020 4th Australian Microwave Symposium (AMS), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. A method to achieve a photonic crystal fiber (PCF) with high birefringence for polarization maintenance in short range THz communication systems is introduced in this paper. Rectangular air slots are etched in the core region of the fiber; they make the X-polarized (XP) and Y-polarized (YP) propagation modes have different propagation constants, thus leading to the higher birefringence. In contrast to the widely-used fully-slotted (FS) configuration in which the fiber core is almost fully occupied by air slots, the proposed PCF has a partially-slotted (PS) core. The air slot in the core center is absent; only the dielectric background is present. Comparisons are made between the fully-slotted and partially-slotted PCFs to illustrate that the PS PCF overperforms the FS PCF. After optimization, the PS PCF attains a high birefringence value of 0.069 and a total loss of 0.071 cm-1 at 1.0 THz. Over a broad 0.4 THz working band, from 0.53 to 0.93 THz, the dispersion is within 0.06 ps/THz/cm.

Yang, T, Ding, C, Ziolkowski, RW & Jay Guo, Y 1970, 'An Ultra-Wideband Single-Polarization-Single-Mode Terahertz Photonic Crystal Fiber', 2020 IEEE International Conference on Computational Electromagnetics (ICCEM), 2020 IEEE International Conference on Computational Electromagnetics (ICCEM), IEEE, Singapore, SINGAPORE, pp. 21-22.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. A terahertz (THz) photonic crystal fiber (PCF) with an ultra-wide bandwidth and single-polarization-single-mode (SPSM) operation is designed and analyzed. Two air slots are introduced in the core region and epsilon-near-zero (ENZ) material is deposited in four specific air holes in the cladding of the PCF. The design achieves significantly different electric (E)-field distributions of the X-polarized (XP) and Y-polarized (YP) modes. The E-field components overlapping the ENZ material are attenuated because it is lossy. Gain material is then deposited in a rectangular slot in the core center to provide amplification of the E-field components overlapping this gain region. Changing the dimensions of the PCF modifies the amplification and attenuation rates to the wanted XP mode, the unwanted YP mode, and any unwanted higher order (HO) modes. The amplification of the wanted mode and the attenuation of the unwanted modes are maximized through optimization. The result is a PCF with an ultra-wide SPSM spectrum of 0.53 THz, from 1.00 to 1.53 THz. The minimum loss difference (MLD) across this bandwidth between the wanted mode and any unwanted modes is over 7.4 dB/cm. To the best of our knowledge, this is the widest SPSM bandwidth of a PCF fiber reported in THz regime.

Yao, Y, Hua, X, Gao, G, Sun, Z, Li, Z & Zhang, J 1970, 'Bridging the Web Data and Fine-Grained Visual Recognition via Alleviating Label Noise and Domain Mismatch', Proceedings of the 28th ACM International Conference on Multimedia, MM '20: The 28th ACM International Conference on Multimedia, ACM, Virtual, pp. 1735-1744.
View/Download from: Publisher's site
View description>>

To distinguish the subtle differences among fine-grained categories, a large amount of well-labeled images are typically required. However, manual annotations for fine-grained categories is an extremely difficult task as it usually has a high demand for professional knowledge. To this end, we propose to directly leverage web images for fine-grained visual recognition. Our work mainly focuses on two critical issues including 'label noise' and 'domain mismatch' in the web images. Specifically, we propose an end-to-end deep denoising network (DDN) model to jointly solve these problems in the process of web images selection. To verify the effectiveness of our proposed approach, we first collect web images by using the labels in fine-grained datasets. Then we apply the proposed deep denoising network model for noise removal and domain mismatch alleviation. We leverage the selected web images as the training set for fine-grained categorization models learning. Extensive experiments and ablation studies demonstrate state-of-the-art performance gained by our proposed approach, which, at the same time, delivers a new pipeline for fine-grained visual categorization that is to be highly effective for real-world applications.

Yuan, X, Feng, Z, Ni, W, Wei, Z & Liu, RP 1970, 'Waveform Optimization for MIMO Joint Communication and Radio Sensing Systems with Imperfect Channel Feedbacks', 2020 IEEE International Conference on Communications Workshops (ICC Workshops), 2020 IEEE International Conference on Communications Workshops (ICC Workshops), IEEE, Dublin, Ireland, pp. 1-6.
View/Download from: Publisher's site
View description>>

In this paper, we study optimal waveform design to maximize mutual information (MI) for a joint communication and (radio) sensing (JCAS, a.k.a., radar-communication) multi-input multi-output (MIMO) downlink system. We consider a typical packet-based signal structure which includes training and data symbols. We first derive the conditional MI for both sensing and communication under correlated channels by considering the training overhead and channel estimation error (CEE). Based on the derived MI, we provide optimal waveform design methods for three scenarios, including maximizing MI for communication only and for sensing only, and maximizing a weighted sum MI for both communication and sensing. We also present extensive simulation results that provide insights on waveform design and validate the effectiveness of the proposed designs.

Zhang, C, Yao, Y, Zhang, J, Chen, J, Huang, P, Zhang, J & Tang, Z 1970, 'Web-Supervised Network for Fine-Grained Visual Classification', 2020 IEEE International Conference on Multimedia and Expo (ICME), 2020 IEEE International Conference on Multimedia and Expo (ICME), IEEE, London, United Kingdom, pp. 1-6.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Fine-grained visual classification (FGVC) is a tough task due to its high annotation cost of the fine-grained subcategories. To build a large-scale dataset at low manual cost, straightforwardly learning from web images for FGVC has attracted broad attention. However, there exist two characteristics in the need of concerning for the web dataset: 1) Noisy images; 2) A large proportion of hard examples. In this paper, we propose a simple yet effective approach to deal with noisy images and hard examples during training. Our method is a pure web-supervised method for FGVC. Extensive experiments on three commonly used fine-grained datasets demonstrate that our approach is much superior to the state-of-the-art web-supervised methods. The data and source code of this work have been posted available at: https://github.com/NUST-Machine-Intelligence-Laboratory/WSNFG.

Zhang, H, Huang, X, Zhang, JA & Guo, YJ 1970, 'Adaptive Transmission Based on MMSE Equalization over Fast Fading Channels', 2020 IEEE 92nd Vehicular Technology Conference (VTC2020-Fall), 2020 IEEE 92nd Vehicular Technology Conference (VTC2020-Fall), IEEE, Victoria, BC, Canada, pp. 1-5.
View/Download from: Publisher's site
View description>>

The sixth generation (6G) mobile systems will enable high mobility applications in both space and ground based networks. In this paper, we investigate low-complexity equalization and adaptive transmission schemes to combat fast fading channels due to high mobility. We first derive signal and channel models in fast fading channels, which allow low complexity minimum mean square error (MMSE) equalization. We then analyze the output signal-to-noise ratio (SNR) using eigenvalue decomposition for a generalized modulation representation. Assuming the channel state information (CSI) is known at the transmitter, we propose an adaptive transmission technique which utilizes the CSI to precode data symbols in order to improve the output SNR at the receiver. Simulation results show that the adaptive transmission scheme effectively improves the MMSE equalization performance in non-line-of-sight channels especially when the transmission signal frame is short.

Zhang, JA, Hoang, L, Nguyen, D, Huang, X, Kekirigoda, A & Hui, K-P 1970, 'Multi-user MIMO Communications with Interference Mitigation in Time-varying Channels', 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), 2020 14th International Conference on Signal Processing and Communication Systems (ICSPCS), IEEE, Adelaide, Australia, pp. 1-7.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. In this paper, we present a technique for realizing reliable multi-user MIMO communications in the presence of interference in time-varying channels. The null space of interfering channels is estimated and exploited for interference mitigation. We first introduce an improved superframe structure to enable frequent tracking of user channels and the null space of interfering channels. The different natures of the received user signals and interference require different processing methods. We improve and compare several adaptive equalizers to deal with time-varying user channels, and propose to use a subspace-based tracking algorithm to handle time-varying interfering channels. We simulate the proposed tracking algorithms in various settings, including when the interference signals are correlated. Simulation results are provided and validate the effectiveness of the proposed technique.

Zhang, L, Zhang, J, Li, Z & Xu, J 1970, 'Towards Better Graph Representation: Two-Branch Collaborative Graph Neural Networks For Multimodal Marketing Intention Detection', 2020 IEEE International Conference on Multimedia and Expo (ICME), 2020 IEEE International Conference on Multimedia and Expo (ICME), IEEE, London, United Kingdom, pp. 1-6.
View/Download from: Publisher's site
View description>>

© 2020 IEEE. Inspired by the fact that spreading and collecting information through the Internet becomes the norm, more and more people choose to post for-profit contents (images and texts) in social networks. Due to the difficulty of network censors, malicious marketing may be capable of harming the society. Therefore, it is meaningful to detect marketing intentions online automatically. However, gaps between multimodal data make it difficult to fuse images and texts for content marketing detection. To this end, this paper proposes Two-Branch Collaborative Graph Neural Networks to collaboratively represent multimodal data by Graph Convolution Networks (GCNs) in an end-to-end fashion. We first separately embed groups of images and texts by GCNs layers from two views and further adopt the proposed multimodal fusion strategy to learn the graph representation collaboratively. Experimental results demonstrate that our proposed method achieves superior graph classification performance for marketing intention detection.

Zhang, R, Xu, M, Shi, Y, Fan, J, Mu, C & Xu, L 1970, 'Infrared Target Detection Using Intensity Saliency And Self-Attention', 2020 IEEE International Conference on Image Processing (ICIP), 2020 IEEE International Conference on Image Processing (ICIP), IEEE.
View/Download from: Publisher's site

Zhang, S, Luo, L, Li, Z, Wang, Y, Chen, F & Xu, R 1970, 'Simultaneous Customer Segmentation and Behavior Discovery', Neural Information Processing, International Conference on Neural Information Processing, Springer International Publishing, Bangkok, Thailand, pp. 122-130.
View/Download from: Publisher's site
View description>>

© 2020, Springer Nature Switzerland AG. Customer purchase behavior segmentation plays an important role in the modern economy. We proposed a Bayesian non-parametric (BNP)-based framework, named Simultaneous Customer Segmentation and Utility Discovery (UtSeg), to discover customer segmentation without knowing specific forms of utility functions and parameters. For the segmentation based on BNP models, the unknown type of functions is usually modeled as a non-homogeneous point process (NHPP) for each mixture component. However, the inference of these models is complex and time-consuming. To reduce such complexity, traditionally, economists will use one specific utility function in a heuristic way to simplify the inference. We proposed to automatically select among multiple utility functions instead of searching in a continuous space. We further unified the parameters for different types of utility functions with the same prior distribution to improve efficiency. We tested our model with synthetic data and applied the framework to real-supermarket data with different products, and showed that our results can be interpreted with common knowledge.

Zhang, Z, Da Xu, RY, Jiang, S, Li, Y, Huang, C & Deng, C 1970, 'Illumination Adaptive Person Reid Based on Teacher-Student Model and Adversarial Training', 2020 IEEE International Conference on Image Processing (ICIP), 2020 IEEE International Conference on Image Processing (ICIP), IEEE.
View/Download from: Publisher's site

Zhang, Z, Yu, L, Zhang, J & Wu, Q 1970, 'A Vision Based Fish Processing System', 2020 IEEE International Conference on Visual Communications and Image Processing (VCIP), 2020 IEEE International Conference on Visual Communications and Image Processing (VCIP), IEEE, Macau, China, pp. 260-260.
View/Download from: Publisher's site
View description>>

The digital fish provenance and quality tracking system is essential for the seafood supply chain. As a part of this system, we develop a vision-based fish processing system to automatically perform fish freshness estimation, size measurement and species classification. Under the constrained illumination environment, our system is able to auto-process the fish selection, thus greatly reduce the human labour and bring trust and efficiency to the seafood supply chain from catch to market.

Zhang, Z, Yu, L, Zhang, J & Wu, Q 1970, 'A Vision Based Fish Processing System', 2020 IEEE INTERNATIONAL CONFERENCE ON VISUAL COMMUNICATIONS AND IMAGE PROCESSING (VCIP), IEEE International Conference on Visual Communications and Image Processing (VCIP), IEEE, ELECTR NETWORK, pp. 260-260.
View/Download from: Publisher's site

Zhou, W, Cheng, Z & Guo, YJ 1970, 'A Dual-Polarized Patch Antenna With Electric and Magnetic Coupling Feed for 5G Base Stations', 2020 IEEE 5th International Conference on Integrated Circuits and Microsystems (ICICM), 2020 IEEE 5th International Conference on Integrated Circuits and Microsystems (ICICM), IEEE.
View/Download from: Publisher's site

Zhu, H & Guo, YJ 1970, 'A Wideband Differentially Fed Multi-beam Antenna Array', 2020 14th European Conference on Antennas and Propagation (EuCAP), 2020 14th European Conference on Antennas and Propagation (EuCAP), IEEE, Copenhagen, Denmark, pp. 1-3.
View/Download from: Publisher's site
View description>>

A differential Butler matrix is presented in this paper using a new type of wideband unbalanced-to-balanced power dividers. The differential Butler matrix has the merit of high levels of common-mode signal suppression. A differential array with four elements is also designed, fabricated and tested. By feeding the differential array with the differential Butler matrix, two beams are produced in the E-plane radiation pattern. The differentially fed array achieves very low cross-polarization level due to the excellent common-mode suppression from the Butler matrix. The design approach is verified experimentally, and the measured result agrees well with the predicted one, demonstrating the application potential for the presented differential beam-forming networks.

Zhu, H & Guo, YJ 1970, 'Circularly-Polarized Differential Antenna Array Fed by Single-Ended-to-Balanced Power Dividers with High Common-Mode Rejection', 2020 4th Australian Microwave Symposium (AMS), 2020 4th Australian Microwave Symposium (AMS), IEEE, Sydney, Australia.
View/Download from: Publisher's site
View description>>

This paper presents a differential feeding network comprising of a new type of single-ended-to-balanced power dividers with high level of common-mode rejection. The single-ended-to-balanced power dividers are built based on slotline-to-microstrip transitions, which are able to provide high common-mode suppression and low differential-mode-to-common-mode conversion levels. A wideband differential circularly-polarized (CP) antenna array is designed fabricated and tested using the differential feeding network. The experimental results verify that the presented differential feeding network can be used in feeding differential CP arrays to achieve high gain, symmetrical patterns and a wide axial ratio bandwidth.

Zhu, J, McGloin, D, Yang, Y & Liu, B 1970, '0.32 THz dual circularly polarized reflectarray', 14th Pacific Rim Conference on Lasers and Electro-Optics (CLEO PR 2020), Conference on Lasers and Electro-Optics/Pacific Rim, Optica Publishing Group, ELECTR NETWORK.
View/Download from: Publisher's site
View description>>

A terahertz (THz) reflect-array is proposed. Dual circularly polarized (left- and right-hand-circular-polarizations) collimated beams are independently manipulated. In our model, the left-hand-circularly-polarized and right-hand-circularly-polarized beams reflect at 23-degrees along the y-direction and x-direction respectively.

Zou, Y, Gong, S, Xu, J, Cheng, W, Hoang, DT & Niyato, D 1970, 'Joint Energy Beamforming and Optimization for Intelligent Reflecting Surface Enhanced Communications', 2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW), 2020 IEEE Wireless Communications and Networking Conference Workshops (WCNCW), IEEE, Seoul, Korea (South), pp. 1-6.
View/Download from: Publisher's site
View description>>

To face the scarcity of wireless spectrum resources and explosive energy consumption due to rapid growth of mobile devices and Internet of Things terminals, intelligent reflecting surface (IRS) has recently gained a lot of attention and become as one of the promising solutions. In this paper, we consider an IRS-enhanced multiple-input single- output (MISO) system, in which the IRS is wireless powered by the access point (AP) in power splitting scheme. We aim to maximize the signal-to-noise ratio (SNR) of the end user by jointly optimizing the AP's beamforming as well as the phase-shift and the power splitting ratio of the IRS elements. To tackle the non-convexity of the formulated problem due to the coupling of optimization variables, we devise a two-stage approximation algorithm by analyzing and then decomposing the structure of the problem. Specifically, the algorithm first tunes the phase-shift of IRS elements to align the equivalent channel of IRS reflected path to that of the direct link. After that, we adopt a successive convex approximation based method to achieve a near optimal solution for the reformulated problem iteratively. The simulation results show that our proposed two-stage approximation algorithm can solve the jointly SNR maximization problem efficiently.

Zou, Y, Xie, Y, Zhang, C, Gong, S, Hoang, DT & Niyato, D 1970, 'Optimization-driven Hierarchical Deep Reinforcement Learning for Hybrid Relaying Communications', 2020 IEEE Wireless Communications and Networking Conference (WCNC), 2020 IEEE Wireless Communications and Networking Conference (WCNC), IEEE, Seoul, Korea (South), pp. 1-6.
View/Download from: Publisher's site
View description>>

In this paper, we employ multiple wireless-powered user devices as wireless relays to assist information transmission from a multi-antenna access point to a single-antenna receiver. To improve energy efficiency, we design a hybrid relaying communication strategy in which wireless relays are allowed to operate in either the passive mode via backscatter communications or the active mode via RF communications, depending on their channel conditions and energy states. We aim to maximize the overall SNR by jointly optimizing the access point's beamforming strategy as well as individual relays' radio modes and operating parameters. Due to the non-convex and combinatorial structure of the SNR maximization problem, we develop a deep reinforcement learning approach that adapts the beamforming and relaying strategies dynamically. In particular, we propose a novel optimization-driven hierarchical deep deterministic policy gradient (H-DDPG) approach that integrates the model-based optimization into the framework of conventional DDPG approach. It decomposes the discrete relay mode selection into the outer-loop by using deep Q-network (DQN) algorithm and then optimizes the continuous beamforming and relays' operating parameters by using the inner-loop DDPG algorithm. Simulation results reveal that the H-DDPG is robust to the hyper parameters and can speed up the learning process compared to the conventional DDPG approach.

Amirgholipour, S, He, X, Jia, W, Wang, D & Liu, L 2020, 'PDANet: Pyramid Density-aware Attention Net for Accurate Crowd Counting', arXiv.

Amiri, M, Abolhasan, M, Shariati, N & Lipman, J 2020, 'Soil Moisture Remote Sensing Using SIW Cavity Based Metamaterial Perfect Absorber', Research Square Platform LLC.
View/Download from: Publisher's site

Barlass, T & Thompson, A 2020, 'Transport for NSW investigating system outage amid 'hacking' fears', The Age.

Chu, NH, Hoang, DT, Nguyen, DN, Huynh, NV & Dutkiewicz, E 2020, 'Fast or Slow: An Autonomous Speed Control Approach for UAV-assisted IoT Data Collection Networks'.

Gill, A 2020, 'EA for Digital – Digital for EA'.

Huang, H, Zhang, J, Zhang, J, Wu, Q & Xu, C 2020, 'TOAN: Target-Oriented Alignment Network for Fine-Grained Image Categorization with Few Labeled Samples'.

Kieu, BT, Unanue, IJ, Pham, SB, Phan, HX & Piccardi, M 2020, 'Learning Neural Textual Representations for Citation Recommendation'.
View description>>

With the rapid growth of the scientific literature, manually selecting appropriate citations for a paper is becoming increasingly challenging and time-consuming. While several approaches for automated citation recommendation have been proposed in the recent years, effective document representations for citation recommendation are still elusive to a large extent. For this reason, in this paper we propose a novel approach to citation recommendation which leverages a deep sequential representation of the documents (Sentence-BERT) cascaded with Siamese and triplet networks in a submodular scoring function. To the best of our knowledge, this is the first approach to combine deep representations and submodular selection for a task of citation recommendation. Experiments have been carried out using a popular benchmark dataset - the ACL Anthology Network corpus - and evaluated against baselines and a state-of-the-art approach using metrics such as the MRR and F1-at-k score. The results show that the proposed approach has been able to outperform all the compared approaches in every measured metric.

Klement, P, Zeichner, F, Subirana, B & Gill, A 2020, 'How Digital Transformation and IoT Can Contribute to the UN Sustainable Development Goals', The Industrial Internet Consortium (IIC) and IoT Alliance Australia (IoTAA).

Nguyen, C, Saputra, Y, Nguyen, H, Nguyen, T, Tran, K, Bui, T, Nguyen, D, Dinh, H, Vu, T, Dutkiewicz, E, Chatzinotas, S & Ottersten, B 2020, 'Enabling and Emerging Technologies for Social Distancing during Pandemics like COVID-19: A Comprehensive Survey'.

Nguyen, CT, Nguyen, DN, Hoang, DT, Pham, H-A, Tuong, NH, Xiao, Y & Dutkiewicz, E 2020, 'BlockRoam: Blockchain-based Roaming Management System for Future Mobile Networks'.

Nguyen, N-T, Nguyen, DN, Hoang, DT, Huynh, NV, Nguyen, N-H, Nguyen, Q-T & Dutkiewicz, E 2020, 'Time Scheduling and Energy Trading for Heterogeneous Wireless-Powered and Backscattering-based IoT Networks'.

Ni, Z, Zhang, JA, Huang, X, Yang, K & Yuan, J 2020, 'Uplink Sensing in Perceptive Mobile Networks with Asynchronous Transceivers'.

Ni, Z, Zhang, JA, Yang, K, Huang, X & Tsiftsis, TA 2020, 'Waveform Optimization with Multiple Performance Metrics for Broadband Joint Communication and Radar Sensing', arXiv.
View/Download from: Publisher's site

Pham, Q-V, Nguyen, DC, Mirjalili, S, Hoang, DT, Nguyen, DN, Pathirana, PN & Hwang, W-J 2020, 'Swarm Intelligence for Next-Generation Wireless Networks: Recent Advances and Applications'.

Tan, S-L 2020, 'Australia targeted in ‘malicious’ cyberattack by state actor, PM Morrison says', South China Morning Post.

Thompson, A 2020, 'NSW government was warned over cyber security weaknesses', The Sydney Morning Herald.

Unanue, IJ, Esmaili, N, Haffari, G & Piccardi, M 2020, 'Leveraging Discourse Rewards for Document-Level Neural Machine Translation'.
View description>>

Document-level machine translation focuses on the translation of entire documents from a source to a target language. It is widely regarded as a challenging task since the translation of the individual sentences in the document needs to retain aspects of the discourse at document level. However, document-level translation models are usually not trained to explicitly ensure discourse quality. Therefore, in this paper we propose a training approach that explicitly optimizes two established discourse metrics, lexical cohesion (LC) and coherence (COH), by using a reinforcement learning objective. Experiments over four different language pairs and three translation domains have shown that our training approach has been able to achieve more cohesive and coherent document translations than other competitive approaches, yet without compromising the faithfulness to the reference translation. In the case of the Zh-En language pair, our method has achieved an improvement of 2.46 percentage points (pp) in LC and 1.17 pp in COH over the runner-up, while at the same time improving 0.63 pp in BLEU score and 0.47 pp in F_BERT.

Wu, K, Zhang, JA, Huang, X & Guo, YJ 2020, 'Integrating Secure and High-Speed Communications into Frequency Hopping MIMO Radar'.

Yu, L, Gao, Y, Zhou, J & Zhang, J 2020, 'Parameter Efficient Deep Neural Networks with Bilinear Projections'.

Yu, L, Zhang, J & Wu, Q 2020, 'Dual Attention on Pyramid Feature Maps for Image Captioning'.

Zhang, H, Huang, X & Zhang, JA 2020, 'Low-Complexity Frequency Domain Equalization over Fast Fading Channels'.