Energy-Aware Federated Learning with Distributed User Sampling and Multichannel ALOHA
This addresses energy efficiency for edge devices in federated learning, but it is incremental as it builds on existing FL and energy harvesting methods.
The paper tackles the problem of energy depletion in federated learning on edge devices by integrating energy harvesting with multi-channel ALOHA, proposing a method to reduce energy outage probability and ensure task execution. Numerical results show it outperforms a norm-based solution in convergence time and battery level, especially when average energy income is insufficient.
Distributed learning on edge devices has attracted increased attention with the advent of federated learning (FL). Notably, edge devices often have limited battery and heterogeneous energy availability, while multiple rounds are required in FL for convergence, intensifying the need for energy efficiency. Energy depletion may hinder the training process and the efficient utilization of the trained model. To solve these problems, this letter considers the integration of energy harvesting (EH) devices into a FL network with multi-channel ALOHA, while proposing a method to ensure both low energy outage probability and successful execution of future tasks. Numerical results demonstrate the effectiveness of this method, particularly in critical setups where the average energy income fails to cover the iteration cost. The method outperforms a norm based solution in terms of convergence time and battery level.