Dynamic Scheduling for Over-the-Air Federated Edge Learning with Energy Constraints

TL;DR

This paper proposes an energy-aware dynamic device scheduling algorithm for over-the-air federated edge learning that optimizes training performance under energy constraints by predicting gradient norms and balancing communication and computation energy.

Contribution

It introduces a novel scheduling algorithm that accounts for both communication and computation energy, incorporating gradient norm estimation to improve training efficiency under energy limits.

Findings

Increases CIFAR-10 accuracy by 4.9% under energy constraints

Effectively balances communication and computation energy consumption

Outperforms myopic benchmark in unbalanced data scenarios

Abstract

Machine learning and wireless communication technologies are jointly facilitating an intelligent edge, where federated edge learning (FEEL) is a promising training framework. As wireless devices involved in FEEL are resource limited in terms of communication bandwidth, computing power and battery capacity, it is important to carefully schedule them to optimize the training performance. In this work, we consider an over-the-air FEEL system with analog gradient aggregation, and propose an energy-aware dynamic device scheduling algorithm to optimize the training performance under energy constraints of devices, where both communication energy for gradient aggregation and computation energy for local training are included. The consideration of computation energy makes dynamic scheduling challenging, as devices are scheduled before local training, but the communication energy for over-the-air aggregation depends on the l2-norm of local gradient, which is known after local training. We thus incorporate estimation methods into scheduling to predict the gradient norm. Taking the estimation error into account, we characterize the performance gap between the proposed algorithm and its offline counterpart. Experimental results show that, under a highly unbalanced local data distribution, the proposed algorithm can increase the accuracy by 4.9% on CIFAR-10 dataset compared with the myopic benchmark, while satisfying the energy constraints.