IoT Thrust Seminar | Massive Random Access for Internet-of-Things: From Single-Cell to Cell-Free Massive MIMO
In the era of modern machine-type communications (MTC), efficiently identifying active Internet-of-Things (IoT) devices during the random access phase is crucial. Existing single-cell activity detection methods face significant computational complexity and delays, making real-time implementation difficult. The challenge intensifies with cell-free massive MIMO due to asynchronous IoT devices over large areas and capacity-limited fronthaul links. This talk will explore deep learning techniques for single-cell activity detection, ensuring real-time implementation and adaptability across varying device numbers. For cell-free massive MIMO, I will introduce a communication-efficient end-to-end learning framework that maintains high detection performance with minimal fronthaul overhead. These advancements promise to enhance the efficiency and reliability of IoT networks, paving the way for more robust MTC systems.
Dr. Yang Li obtained his Ph.D. in Electrical and Electronic Engineering from The University of Hong Kong (HKU), along with the B.E. and M.E. degrees in Electronic Information Engineering from Beihang University (BUAA). He currently works as a Research Scientist (副研究员) in Shenzhen Research Institute of Big Data, The Chinese University of Hong Kong, Shenzhen. His research expertise lies in distributed optimization and deep learning algorithms for optimization problems arising in Internet of Things, cell-free massive MIMO, and large-scale wireless networks. He has been the principal investigator of four national research projects/sub-projects. He has over 30 high-quality papers published in top international journals (SCI, Q1), such as IEEE Transactions on Wireless Communications and IEEE Transactions on Signal Processing. Among these, he is the first or corresponding author of more than 20 papers. He was a recipient of the 2020 Innovation Pioneer Award from Huawei.