Resource Management for Federated Learning

In this chapter, we introduce the joint optimization of FL parameters and wireless resource (i.e., computational resources, spectrum, transmit power) to optimize wireless FL performance metrics (i.e., FL training loss, convergence time, energy efficiency). In particular, we first analyze how wireless factors (i.e., transmit power, spectrum interference, packet error rates) affect the FL performance. Then, using the FL convergence analytical results, we introduce several optimization theory based methods for resource management aiming to optimize the wireless FL performance metrics. Finally, several simulations are implemented to demonstrate the performance of the designed FL.