Energy-Efficient Radio Resource Allocation for Federated Edge Learning

TL;DR

This paper proposes energy-efficient radio resource management strategies for federated edge learning, optimizing bandwidth allocation and scheduling to reduce device energy consumption while maintaining learning performance.

Contribution

It introduces novel adaptive RRM strategies that consider device channel states and capacities, improving energy efficiency in FEEL systems.

Findings

Significant energy reduction in device operation.

Optimal policies favor weaker channel devices for bandwidth.

Scheduling prioritizes devices with better channels and capacities.

Abstract

Edge machine learning involves the development of learning algorithms at the network edge to leverage massive distributed data and computation resources. Among others, the framework of federated edge learning (FEEL) is particularly promising for its data-privacy preservation. FEEL coordinates global model training at a server and local model training at edge devices over wireless links. In this work, we explore the new direction of energy-efficient radio resource management (RRM) for FEEL. To reduce devices' energy consumption, we propose energy-efficient strategies for bandwidth allocation and scheduling. They adapt to devices' channel states and computation capacities so as to reduce their sum energy consumption while warranting learning performance. In contrast with the traditional rate-maximization designs, the derived optimal policies allocate more bandwidth to those scheduled devices with weaker channels or poorer computation capacities, which are the bottlenecks of synchronized model updates in FEEL. On the other hand, the scheduling priority function derived in closed form gives preferences to devices with better channels and computation capacities. Substantial energy reduction contributed by the proposed strategies is demonstrated in learning experiments.