ZTE Communications ›› 2023, Vol. 21 ›› Issue (1): 38-45.DOI: 10.12142/ZTECOM.202301005

• Special Topic • Previous Articles    

Hierarchical Federated Learning: Architecture, Challenges, and Its Implementation in Vehicular Networks

YAN Jintao1, CHEN Tan1, XIE Bowen1, SUN Yuxuan2, ZHOU Sheng1(), NIU Zhisheng1   

  1. 1.Tsinghua University, Beijing 100084, China
    2.Beijing Jiaotong University, Beijing 100044, China
  • Received:2022-11-04 Online:2023-03-25 Published:2024-03-15
  • About author:YAN Jintao is a PhD student at Tsinghua University, China. His research interests include federated learning and vehicular edge computing and vehicular networks.
    CHEN Tan is a PhD student at Tsinghua University, China. His research interests include federated learning and vehicular networks.
    XIE Bowen is a PhD student at Tsinghua University, China. His research interests include federated learning and vehicular networks.
    SUN Yuxuan is an associate professor with the School of Electronic and Information Engineering, Beijing Jiaotong University, China and was previously a postdoctoral researcher with Tsinghua University, China. Her research interests include edge computing and edge learning.
    ZHOU Sheng (sheng.zhou@tsinghua.edu.cn) is an associate professor with the Department of Electronic Engineering, Tsinghua University, China. His research interests include vehicular networks, mobile edge computing, and green wireless communications.
    NIU Zhisheng is a professor with the Department of Electronic Engineering, Tsinghua University, China. His major research interests include queueing theory, traffic engineering, radio resource management of wireless networks, and green communication and networks.
  • Supported by:
    The work of YAN Jintao, CHEN Tan, XIE Bowen, ZHOU Sheng and NIU Zhisheng are sponsored in part by the National Key R&D Program of China(2020YFB1806605);the National Natural Science Foundation of China(62022049);OPPO. The work of SUN Yuxuan is supported by the Fundamental Research Funds for the Central Universities(2022JBXT001)

Abstract:

Federated learning (FL) is a distributed machine learning (ML) framework where several clients cooperatively train an ML model by exchanging the model parameters without directly sharing their local data. In FL, the limited number of participants for model aggregation and communication latency are two major bottlenecks. Hierarchical federated learning (HFL), with a cloud-edge-client hierarchy, can leverage the large coverage of cloud servers and the low transmission latency of edge servers. There are growing research interests in implementing FL in vehicular networks due to the requirements of timely ML training for intelligent vehicles. However, the limited number of participants in vehicular networks and vehicle mobility degrade the performance of FL training. In this context, HFL, which stands out for lower latency, wider coverage and more participants, is promising in vehicular networks. In this paper, we begin with the background and motivation of HFL and the feasibility of implementing HFL in vehicular networks. Then, the architecture of HFL is illustrated. Next, we clarify new issues in HFL and review several existing solutions. Furthermore, we introduce some typical use cases in vehicular networks as well as our initial efforts on implementing HFL in vehicular networks. Finally, we conclude with future research directions.

Key words: hierarchical federated learning, vehicular network, mobility, convergence analysis