eduzhai > Applied Sciences > Engineering >

Joint Auction-Coalition Formation Framework for Communication-Efficient Federated Learning in UAV-Enabled Internet of Vehicles

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 18 pages

Abstract: Due to the advanced capabilities of the Internet of Vehicles (IoV) componentssuch as vehicles, Roadside Units (RSUs) and smart devices as well as theincreasing amount of data generated, Federated Learning (FL) becomes apromising tool given that it enables privacy-preserving machine learning thatcan be implemented in the IoV. However, the performance of the FL suffers fromthe failure of communication links and missing nodes, especially whencontinuous exchanges of model parameters are required. Therefore, we proposethe use of Unmanned Aerial Vehicles (UAVs) as wireless relays to facilitate thecommunications between the IoV components and the FL server and thus improvingthe accuracy of the FL. However, a single UAV may not have sufficient resourcesto provide services for all iterations of the FL process. In this paper, wepresent a joint auction-coalition formation framework to solve the allocationof UAV coalitions to groups of IoV components. Specifically, the coalitionformation game is formulated to maximize the sum of individual profits of theUAVs. The joint auction-coalition formation algorithm is proposed to achieve astable partition of UAV coalitions in which an auction scheme is applied tosolve the allocation of UAV coalitions. The auction scheme is designed to takeinto account the preferences of IoV components over heterogeneous UAVs. Thesimulation results show that the grand coalition, where all UAVs join a singlecoalition, is not always stable due to the profit-maximizing behavior of theUAVs. In addition, we show that as the cooperation cost of the UAVs increases,the UAVs prefer to support the IoV components independently and not to form anycoalition.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...