Bandwidth Allocation for Multiple Federated Learning Services in Wireless Edge Networks

This paper studies a federated learning (FL) system, where multiple FL services co-exist in wireless network and share common resources. It fills the void of resource allocation for multiple simultaneous existing literature. Our method designs two-level framework comprising intra-service inter-service allocation. The intra-service problem aims to minimize length rounds by optimizing bandwidth among clients each service. Based on this, an inter-service is further considered, which distributes resources services. We consider both cooperative selfish providers For service providers, we design distributed algorithm optimize overall performance services, meanwhile catering it fairness privacy clients. new auction scheme designed with as bidders operator auctioneer. strikes balance between fairness. simulation results show that proposed algorithms outperform other benchmarks under various conditions.