Wirelessly Powered Federated Edge Learning

Abstract

The execution of a power-hungry learning task at energy-constrained devices is a key challenge confronting the implementation of federated edge learning (FEEL). To tackle the challenge, we propose the solution of powering devices using wireless power transfer (WPT). To derive guidelines on deploying the resultant wirelessly powered FEEL (WP-FEEL) system, this work aims at the derivation of the optimal tradeoffs between the model convergence and the settings of power sources: the transmission power and density of power-beacons, which are dedicated charging stations. To this end, the local-computation at devices (i.e., their mini-batch sizes and processor clock frequencies) is optimized to efficiently use the harvested energy for gradient estimation. The resultant optimal tradeoffs are derived to relate the accuracy of distributed stochastic-gradient estimation to the WPT settings, the randomness in communication links, and devices' computation capacities. They reveal simple scaling laws of the model-convergence rate with respect to the transferred energy as well as the devices' computational energy efficiencies. The results provide useful guidelines on WPT provisioning to provide a guarantee on learning performance.