ZTE Communications ›› 2023, Vol. 21 ›› Issue (1): 25-37.DOI: 10.12142/ZTECOM.202301004

• Special Topic • Previous Articles    

RIS-Assisted Federated Learning in Multi-Cell Wireless Networks

WANG Yiji, WEN Dingzhu(), MAO Yijie, SHI Yuanming   

  1. ShanghaiTech University, Shanghai 201210, China
  • Received:2022-12-04 Online:2023-03-22 Published:2023-03-22
  • About author:WANG Yiji received his BS degree from Zhejiang University City College, China in 2020. He is currently pursuing his master’s degree with the School of Information Science and Technology, ShanghaiTech University, China. His research interests include federated learning and wireless communications.|WEN Dingzhu (wendzh@shanghaitech.edu.cn) received his bachelor’s and master’s degrees from Zhejiang University, China in 2014 and 2017, respectively, and PhD degree from The University of Hong Kong, China in 2021. Subsequently, he joined ShanghaiTech University, China. He is currently an assistant professor at the School of Information Science and Technology there. His research interests include edge intelligence, integrated sensing, computation and communication, over-the-air computation, in-band full-duplex communications, etc.|MAO Yijie is an assistant professor at the School of Information Science and Technology, ShanghaiTech University, China. She received her BE degree from Beijing University of Posts and Telecommunications, China and BE (Hons.) degree from the Queen Mary University of London in 2014. She received her PhD degree from the Electrical and Electronic Engineering (EEE) Department, The University of Hong Kong, China in 2018. She was a postdoctoral research fellow at The University of Hong Kong from 2018 to 2019 and a postdoctoral research associate with the Department of the EEE at the Imperial College London, UK from 2019 to 2021. She is a senior member of China Institute of Communications. She is currently serving as an editor of IEEE Communications Letters and a guest editor of two special issues of IEEE Journal on Selected Areas in Communications and IEEE Open Journal of the Communications Society.|SHI Yuanming received his BS degree in electronic engineering from Tsinghua University, China in 2011. He received his PhD degree in electronic and computer engineering from The Hong Kong University of Science and Technology (HKUST), China in 2015. Since September 2015, he has been with the School of Information Science and Technology, ShanghaiTech University, China, where he is currently a tenured associate professor. He visited University of California, Berkeley, USA from October 2016 to February 2017. His research areas include optimization, machine learning, wireless communications, and their applications to 6G, IoT and edge AI. He was a recipient of the 2016 IEEE Marconi Prize Paper Award in Wireless Communications, the 2016 Young Author Best Paper Award by the IEEE Signal Processing Society, and the 2021 IEEE ComSoc Asia-Pacific Outstanding Young Researcher Award. He is also an editor of IEEE Transactions on Wireless Communications, IEEE Journal on Selected Areas in Communications, and Journal of Communications and Information Networks.

Abstract:

Over-the-air computation (AirComp) based federated learning (FL) has been a promising technique for distilling artificial intelligence (AI) at the network edge. However, the performance of AirComp-based FL is decided by the device with the lowest channel gain due to the signal alignment property. More importantly, most existing work focuses on a single-cell scenario, where inter-cell interference is ignored. To overcome these shortages, a reconfigurable intelligent surface (RIS)-assisted AirComp-based FL system is proposed for multi-cell networks, where a RIS is used for enhancing the poor user signal caused by channel fading, especially for the device at the cell edge, and reducing inter-cell interference. The convergence of FL in the proposed system is first analyzed and the optimality gap for FL is derived. To minimize the optimality gap, we formulate a joint uplink and downlink optimization problem. The formulated problem is then divided into two separable nonconvex subproblems. Following the successive convex approximation (SCA) method, we first approximate the nonconvex term to a linear form, and then alternately optimize the beamforming vector and phase-shift matrix for each cell. Simulation results demonstrate the advantages of deploying a RIS in multi-cell networks and our proposed system significantly improves the performance of FL.

Key words: federated learning (FL), reconfigurable intelligent surface (RIS), over-the-air computation (AirComp), multi-cell networks