Prof. Li Qiu IEEE Fellow, IFAC Fellow, The Chinese University of Hong Kong, Shenzhen, China | Experience: Professor Li Qiu received the Ph.D. degree in electrical engineering from the University of Toronto in 1990. After briefly working in the Canadian Space Agency, the Fields Institute for Research in Mathematical Sciences (Waterloo), and the Institute of Mathematics and its Applications (Minneapolis), he joined the Hong Kong University of Science and Technology, Hong Kong SAR, China in 1993 and is now a Professor Emeritus in the Department of Electronic and Computer Engineering. In 2023, he took a leave to work in Southern University of Science and Technology, as a Chair Professor. He joined the Chinese University of Hong Kong, Shenzhen in 2024, as a Presidential Chair Professor. Prof. Qiu's research interests include system, control, optimization theory, and mathematics for information technology, as well as their applications in manufacturing industry and energy systems. He is also interested in control education and co-authored an undergraduate textbook "Introduction to Feedback Control" which was published by Prentice-Hall in 2009. He served as an Associate Editor of the IEEE Transactions on Automatic Control and Automatica. He was the General Chair of the 7th Asian Control Conference, which was held in Hong Kong, China in 2009.He was a Distinguished Lecturer from 2007 to 2010 and was a Member of the Board of Governors in 2012 and 2017 of the IEEE Control Systems Society. He was a Vice President of the Asian Control Association and the Founding Chairperson of the Hong Kong Automatic Control Association. He is a Fellow of IEEE, a Fellow of IFAC, and an inaugural Fellow of ACA (Asian Control Association). Title:The Diversity and Interaction Quality (IQ) of a Multi-Agent Network Abstract: If we have a heterogeneous population of agents interconnected by an interaction network, then whether the multi-agent network can be made to exhibit a desirable collective behavior depends on the diversity of the population and the interaction quality of the network. In this talk, for several scenarios, we will define the population diversity and the interaction quality, and we will then present such a dependence. The results presented in this talk come from the recent study of matrix phases. |
Prof. Abdelhak M. Zoubir IEEE Fellow, Technische Universität Darmstadt, Germany | Experience: Abdelhak M Zoubir is an IEEE Life Fellow, a Fellow of EURASIP, and a Distinguished Lecturer (Class 2010-2011). He holds since 2003 the position of Professor at Darmstadt University of Technology, Germany and is the Head of the Signal Processing Group. His research interest lies in statistical methods for signal processing with emphasis on bootstrap techniques, robust detection and estimation and array processing applied to telecommunications, radar, sonar, car engine monitoring and biomedicine. He published over 500 journal and conference papers on these areas. Professor Zoubir was General Chair, Technical Program Chair and Member of Organization Committees of numerous international conferences and workshops, most notably he was Technical Program Co-Chair of the largest and flagship conference on signal processing, The International Conference on Acoustics, Speech, and Signal Processing (ICASSP), held in Florence in 2014. He served on numerous editorial boards, most notably, he was the Editor-In-Chief of the IEEE Signal Processing Magazine (2012-2014). Dr Zoubir was elected as Chair (2010-2011) of the IEEE SPS Technical Committee Signal Processing Theory and Methods (SPTM), and as Member (2007-2012) of the IEEE SPS Technical Committee Sensor Array and Multi-channel Signal Processing (SAM). He served on the Board of Governors of the IEEE SPS as Member-at-Large (2015-2017), and Member of the Board of Directors (BoD) of the European Association for Signal Processing (EURASIP) from 2009-2016, and its President from 2017 until 2018. He has been inducted to the German National Academy of Science and Engineering (acatech) in January 2024. Title:Robust Sequential Detection Abstract:Robust statistics continue to gain importance due to an increase of impulsive measurement environments and outliers in practical engineering systems. Classical estimation or detection theory does not apply in such situations and robust statistical methods are sought for. The talk is on robust sequential detection and aims at discussing the most recent efforts in this area. First, we briefly revisit robust fixed sample size tests and provide an overview of fundamentals of sequential testing. A sequential test is fundamentally different from a fixed sample size test in that it typically continues observing until the evidence strongly favors one of the two hypotheses. The aim is to design robust sequential tests, which cannot be easily drawn from the design of robust fixed sample size tests. We first formulate the problem of robust sequential testing and show some recent results supported by examples. A parallel to robust fixed sample size tests is given and conclusions are drawn in view of minimax optimality. |
Prof. Lei Huang Distinguished Young Scientists of NSFC, Shenzhen University, China | Experience: Lei Huang received the B. Sc. and Ph. D. degrees in electronic engineering from Xidian University, Xi’an, China, in 2000 and 2005, respectively. He is currently with the College of Electronics and Information Engineering, Shenzhen University, as a Chair Professor, and established the Shenzhen Key Laboratory of Advanced Navigation Technology (ANT) as the Founding Director. He is now the Executive Dean of the College of Electronics and Information Engineering and the Executive Director of the State Key Laboratory of Radio Frequency Heterogeneous Integration, Shenzhen University. Dr. Huang’s research interests include spectral estimation, array signal processing, statistical signal processing, and their applications in radar, navigation and wireless communications. In these areas, he has published 130 IEEE journal papers, and undertaken 20 national and provincial key projects, such as the Key Project of the National Natural Science Foundation of China (NSFC) and Joint Project of NSFC-RGC (Hong Kong). He was the winner of the Distinguished Young Scientists of NSFC. Dr. Huang severed as a Senior Area Editor of IEEE Transactions on Signal Processing (2019-2023), and an Associate Editor of IEEE Transactions on Signal Processing (2015-2019). He also was on the editorial boards of Elsevier-Digital Signal Processing (2012-2019) and has been on the editorial boards of IET Signal Processing (2017-present), and an elected member of Sensor Array and Multichannel (SAM) Technical Committee of the IEEE Signal Processing Society (2016-2022). Title：One-Bit MIMO Radar under Colored Noise: From Covariance Estimation to Target Detection Abstract：One-bit radar systems have gained attention for their ability to reduce hardware complexity, data storage, and power consumption. Yet, the resulting quantized measurements pose new challenges for estimation and detection, especially under colored noise. This work unifies two complementary contributions. First, we address covariance matrix recovery from one-bit data using non-zero quantization thresholds. By analyzing the Fisher information, we show that optimal thresholds depend on varying parameters and propose a time-varying scheme that significantly improves estimation accuracy. Building on these enhanced estimates, we then develop a one-bit target detector for colocated MIMO radar that accounts for colored noise and ensures a constant false alarm rate. The derived detector takes the form of a weighted matched filter and achieves notable performance gains in challenging scenarios. Together, these results offer a cohesive framework for advancing one-bit radar systems toward more robust and accurate signal processing. |
Assoc. Prof. Yik-Chung Wu The University of Hong Kong, China | Experience: Yik-Chung Wu received the B.Eng. (EEE) degree in 1998 and the M.Phil. degree in 2001 from the University of Hong Kong (HKU). He received the Croucher Foundation scholarship in 2002 to study Ph.D. degree at Texas A&M University, College Station, and graduated in 2005. From August 2005 to August 2006, he was with the Thomson Corporate Research, Princeton, NJ, as a Member of Technical Staff. Since September 2006, he has been with HKU, currently as an Associate Professor. He was a visiting scholar at Princeton University, in summers of 2015 and 2017. His research interests are in general areas of machine learning and communication systems, and in particular Bayesian inference, distributed algorithms, and large-scale optimization. Dr. Wu served as an Editor for IEEE Communications Letters, and IEEE Transactions on Communications. He is currently a Senior Area Editor for IEEE Transactions on Signal Processing, an Associate Editor for IEEE Wireless Communications Letters, and an Editor for Journal of Communications and Networks. He was a symposium chair for many international conferences, including IEEE International Conference on Communications (ICC) 2023, and IEEE Globecom 2025. He received four best paper awards in international conferences, with the most recent one from IEEE International Conference on Communications (ICC) 2020. He was a top 1% scholar (ranked by Clarivate Analytics) for seven consecutive years (2015-2021). He was elected the Best Editor of the year 2023 in IEEE Wireless Communications Letters. He is a senior member of the IEEE. Title：Parameter tuning-free matrix and tensor decompositions: A Bayesian Approach Abstract：Matrix and Tensor factorizations are important data analytic tools in many applications, such as recommendation systems, image completion, social network data mining, wireless communications, etc. Traditionally, matrix and tensor factorizations are approached from optimization perspective. While proven to be effective, optimization-based matrix and tensor factorizations usually involve hyperparameters tuning, with one of the major hyperparameters being the matrix or tensor rank. However, when the number of hyperparameters is more than 3 or 4, tuning them becomes computationally expensive. This talk approaches the problem from the Bayesian perspective and shows how hyperparameter tuning can be eliminated while providing comparable or even better performance than corresponding optimization-based algorithms. |
Assist. Prof. Qinglei Kong Harbin Institute of Technology (Shenzhen), China | Experience: Qinglei Kong received the Ph.D. degree from the School of Electrical and Electronics Engineering, Nanyang Technological University, Singapore, in 2018. She is currently working as an Assistant Professor with the school of aerospace science, Harbin Institute of Technology, Shenzhen. Her research interests include applied cryptography and satellite IoT. Title: Secure collaborative on-orbit processing in LEO satellite constellation Abstract:With the recent boom of LEO satellite constellations and space electronics, LEO satellites are equipped with various computation and storage devices, which enable onboard data processing in space. A typical LEO satellite constellation acts as a public platform shared by multiple sectors, reflecting the status of the observation targets that must be maintained within the satellite. However, due to the orbital characteristics and onboard energy supply, a typical data processing task requires collaboration across multiple satellites. Under the above challenges, we propose a secure, collaborative, on-orbit data processing framework that covers joint and continuous surveillance, joint deduplication, and joint anomaly identification. To achieve the continuous and constant surveillance of a target area, the proposed framework first achieves the secure task delegation between a pair of satellites, which protects the details before the successful authentication. Second, as the speed of onboard data generation exceeds downlink bandwidth, the proposed secure framework supports the on-orbit data deduplication between satellites without disclosing the differences towards each other. Third, with one satellite holding a set of anomalies and the other satellite collecting real-time data, the proposed framework also enables the secure identification on-orbit and allows the secure on-orbit query of the identified anomalies. |
Assist. Prof. Feng Yin The Chinese University of Hong Kong, Shenzhen, China | Experience: Feng Yin received his B.Sc. degree from Shanghai Jiao Tong University, China, and his M.Sc. and Ph.D. degrees from Technische Universitaet Darmstadt, Germany. From 2014 to 2016, he was a postdoc researcher with Ericsson Research, Linkoping, Sweden. Since 2016, he has been with The Chinese University of Hong Kong, Shenzhen and is currently an assistant professor of the School of Science and Engineering. His research interests include statistical signal processing, Bayesian learning and optimization, and sensory data fusion. He has published more than 40 top-tier journal papers, 50 conferences, and 20 patents/standards. He was a recipient of the Chinese Government Award for Outstanding Self-Financed Students Abroad in 2013 and the Marie Curie Young Fellowship from the European Union in 2014. He was the finalist for the IEEE CAMSAP best paper award in 2013 and received the best paper award of ICSINC conference in 2022. He has served as Associate Editor for the Elsevier Signal Processing Journal and currently serving as the Associate Editor for the IEEE Transactions on Signal Processing. He is an IEEE senior member and a core member of the IEEE Machine Learning and Signal Processing (MLSP) technical committee and the IEEE SPS scholarship committee. Title：Towards Flexibility and Learning Efficiency of Gaussian Process State-Space Models Abstract： The Gaussian process state-space model (GPSSM) has garnered considerable attention over the past decade. However, the standard GP with a preliminary kernel, such as the squared exponential kernel or Matérn kernel, that is commonly used in GPSSM studies, limits the model's representation power and substantially restricts its applicability to complex scenarios. To address this issue, I will present a new class of probabilistic state-space models called TGPSSMs, which leverage a parametric normalizing flow to enrich the GP priors in the standard GPSSM, enabling greater flexibility and expressivity. Additionally, I will present a scalable variational inference algorithm that offers a flexible and optimal structure for the variational distribution of latent states. The proposed algorithm is interpretable and computationally efficient due to the sparse GP representation and the bijective nature of normalizing flow. Lastly I will demonstrate some experimental results on synthetic and real datasets to corroborate that the proposed TGPSSM outperforms several state-of-the-art methods. |
