Tutorial/Invited Talks

 

Tutorial/TalksTopicSpeakers/Organizers
TutorialGraph neural networks for wireless communicationsProf. Jun Zhang, Hong Kong University of Science and Technology, IEEE Fellow
Invited TalkChannel modelling and measurementsProf. Jianhua Zhang, Beijing University of Posts and Telecommunications
Invited TalkFast and Robust Low-Rank Matrix CompletionProf. H.C. So, City University of Hong Kong, IEEE Fellow

Topic:
Graph Neural Networks for Wireless Communications
Speaker:
Prof. Jun Zhang, Hong Kong University of Science and Technology, IEEE Fellow

Abstract:

Deep learning techniques have been recently applied to solve various challenging problems in wireless communications. The effectiveness of such approaches highly depends on the neural network architecture. Early attempts adopted architectures inherited from applications such as computer vision, e.g., multi-layer perceptrons (MLPs) convolutional neural networks (CNNs). Unfortunately, methods based on these classic architectures often require huge amounts of training samples (i.e., poor generalization), and yield poor performance in large-scale networks (i.e., poor scalability). This tutorial introduces the graph neural network (GNN) as a promising neural architecture to solve generic design problems in wireless communications, which enjoys good generalization, high computational efficiency, and is with theoretical guarantees. It will present recent advancements in theoretical analysis (of generalization and training performance) of GNN-based approaches to solve general optimization problems in wireless communications, and provide design guidelines for the GNN architecture of different types of problems.

Biography:

Jun Zhang received his Ph.D. degree in Electrical and Computer Engineering from the University of Texas at Austin. He is an Associate Professor in the Department of Electronic and Computer Engineering at the Hong Kong University of Science and Technology. His research interests include wireless communications and networking, mobile edge computing and edge AI, and cooperative AI. Dr. Zhang co-authored the book “Fundamentals of LTE” (Prentice-Hall, 2010). He is a co-recipient of several best paper awards, including the 2021 Best Survey Paper Award of IEEE Communications Society, the 2019 IEEE Communications Society & Information Theory Society Joint Paper Award, and the 2016 Marconi Prize Paper Award in Wireless Communications. Two papers he co-authored received the Young Author Best Paper Award of the IEEE Signal Processing Society in 2016 and 2018, respectively. He also received the 2016 IEEE ComSoc Asia-Pacific Best Young Researcher Award. He is an Editor of IEEE Transactions on Communications, and was an editor of IEEE Transactions on Wireless Communications (2015-2020). He served as a MAC track co-chair for IEEE Wireless Communications and Networking Conference (WCNC) 2011 and a wireless communications symposium co-chair of IEEE International Conference on Communications (ICC) 2021. He is an IEEE Fellow.


Topic:
Channel Modelling and Measurements
Speaker:
Prof. Jianhua Zhang, Beijing University of Posts and Telecommunications

Abstract:

With the advancement of 5G commercialization all over the world, 6G research has been initiated in order to meet future demands for higher rates, lower latency, and new services. A wireless channel is a medium for information transmission between the transmitter and the receiver, and the characteristics of the wireless channel determine the performance limit of the wireless communication system. Therefore, 6G channel research is a prerequisite for the design and development of 6G wireless communication systems. This report first introduces the background and significance of 6G channel research, as well as the challenges brought by “large frequency span”, “complex scenarios”, and “various technologies”. Then, the key steps of 6G channel research based on channel measurement are introduced, including channel sounder construction, channel data acquisition, multipath parameter estimation and clustering, and channel modeling. Then, the research progress in massive MIMO channel, terahertz channel, intelligent modeling based on cluster-nucleus, and channel fading tracking and prediction is summarized. Finally, the conclusions and prospects of the 6G channel research to be further studied are given.

Biography:

Jianhua ZhangJianhua Zhang received her Ph.D. degree in circuit and system from Beijing University of Posts and Telecommunication (BUPT) in 2003 and now is a professor at BUPT, China Institute of Communications Fellow, and director of BUPT-CMCC Joint Research Center. She has published more than 200 papers and authorized 50 patents. She received several paper awards, including 2019 SCIENCE China Information Hot Paper, 2016 China Comms Best Paper, 2008 JCN Best Paper, etc. She received several prizes for her contribution to ITU-R 4G channel model (ITU-R M.2135), 3GPP Relay channel model (3GPP TR 36.814), and 3GPP 3D channel model (3GPP TR 36.873). She was also a member of 3GPP “5G channel model for bands up to 100 GHz”. From 2016 to 2017, she was the Drafting Group (DG) Chairwoman of the ITU-R IMT-2020 channel model and led the drafting of the ITU-R M.2412 Channel Model Section. Now she is the Chairwomen of China IMT-2030 tech group-channel measurement and modeling subgroup and works on 6G channel model. Her current research interests include Beyond 5G and 6G, artificial intelligence, and data mining, especially in massive MIMO and millimeter wave channel modeling, channel emulator, OTA test, etc.


Topic:
Fast and Robust Low-Rank Matrix Completion
Speaker:
Prof. H.C. So, City University of Hong Kong, IEEE Fellow

Abstract:

Matrix completion refers to the recovery of a low‐rank matrix from only a subset of its possibly noisy entries, and has a variety of important applications such as collaborative filtering, image restoration, system identification and node localization. It is because many real-world signals can be approximated as a matrix whose rank is much smaller than the row and column lengths. Nevertheless, most existing techniques for matrix recovery rely on -norm minimization, implying their inadequacy to resist gross errors or outliers, and require knowing the matrix rank which is difficult to accurately determine in practice. In this talk, these two challenges will be tackled with a robust rank-one matrix matching pursuit approach. Computer simulations using synthetic data and experimental results of real-world images demonstrate that the proposed algorithm achieves outstanding recovery performance and high computational efficiency.

Biography:

Hing Cheung So was born in Hong Kong. He received the B.Eng. degree from the City University of Hong Kong and the Ph.D. degree from The Chinese University of Hong Kong, both in electronic engineering, in 1990 and 1995, respectively. From 1990 to 1991, he was an Electronic Engineer with the Research and Development Division, Everex Systems Engineering Ltd., Hong Kong. During 1995–1996, he was a Postdoctoral Fellow with The Chinese University of Hong Kong. From 1996 to 1999, he was a Research Assistant Professor with the Department of Electronic Engineering, City University of Hong Kong, where he is currently a Professor. His research interests include detection and estimation, fast and adaptive algorithms, multidimensional harmonic retrieval, robust signal processing, source localization, and sparse approximation.

He has been on the editorial boards of IEEE Signal Processing Magazine (2014–2017), IEEE Transactions on Signal Processing (2010–2014), Signal Processing (2010–), and Digital Signal Processing (2011–). He was also Lead Guest Editor for IEEE Journal of Selected Topics in Signal Processing, special issue on “Advances in Time/Frequency Modulated Array Signal Processing” in 2017. In addition, he was an elected member in Signal Processing Theory and Methods Technical Committee (2011–2016) of the IEEE Signal Processing Society where he was chair in the awards subcommittee (2015–2016). He has been named a 2015 IEEE Fellow in recognition of his contributions to spectral analysis and source localization.