ZTE Communications ›› 2023, Vol. 21 ›› Issue (1): 3-14.DOI: 10.12142/ZTECOM.202301002

• Special Topic • Previous Articles     Next Articles

Adaptive Retransmission Design for Wireless Federated Edge Learning

XU Xinyi, LIU Shengli, YU Guanding()   

  1. Zhejiang University, Hangzhou 310027, China
  • Received:2022-10-27 Online:2023-03-22 Published:2023-03-22
  • About author:XU Xinyi received her BE degree in communication engineering from Zhejiang University, China in 2021. Now she is working towards her MS degree with the College of Information Science and Electronic Engineering, Zhejiang University. Her research interest focuses on federated learning.|LIU Shengli received his BS degree in information engineering from Soochow University, China in 2017, and his PhD degree from the College of Information Science and Electronic Engineering, Zhejiang University, China in 2022. He currently holds a post-doctoral position at the College of Information Science and Electronic Engineering, Zhejiang University. In 2021, he was a Visiting Research Scholar with the Centre for Wireless Communication, University of Oulu, Finland and the VTT Technical Research Centre of Finland. His current research interests mainly include machine learning and federated learning.|YU Guanding (yuguanding@zju.edu.cn) received his BE and PhD degrees in communication engineering from Zhejiang University, China in 2001 and 2006, respectively. He joined Zhejiang University in 2006 and is now a professor with the College of Information and Electronic Engineering. From 2013 to 2015, he was also a visiting professor at the School of Electrical and Computer Engineering, Georgia Institute of Technology, USA. His research interests include 5G communications and networks, mobile edge computing, and machine learning for wireless networks.

Abstract:

As a popular distributed machine learning framework, wireless federated edge learning (FEEL) can keep original data local, while uploading model training updates to protect privacy and prevent data silos. However, since wireless channels are usually unreliable, there is no guarantee that the model updates uploaded by local devices are correct, thus greatly degrading the performance of the wireless FEEL. Conventional retransmission schemes designed for wireless systems generally aim to maximize the system throughput or minimize the packet error rate, which is not suitable for the FEEL system. A novel retransmission scheme is proposed for the FEEL system to make a tradeoff between model training accuracy and retransmission latency. In the proposed scheme, a retransmission device selection criterion is first designed based on the channel condition, the number of local data, and the importance of model updates. In addition, we design the air interface signaling under this retransmission scheme to facilitate the implementation of the proposed scheme in practical scenarios. Finally, the effectiveness of the proposed retransmission scheme is validated through simulation experiments.

Key words: federated edge learning, retransmission, unreliable communication, convergence rate, retransmission latency