Abstract
With the rapid development of the Wireless Computing Power Network (WCPN), the urgent need for data privacy protection and communication efficiency led to the emergence of the Federated learning (FL) framework. However, the time delay leads to dragging problems and reduces the convergence performance of FL in the training process. In this article, we propose a FL resource scheduling strategy based on information age perception in WCPN, which can effectively reduce the time delay and enhance the convergence performance of FL. Moreover, a data cache buffer and a model cache buffer are set up at the user and the central server respectively. Next, we formulate the parametric age-aware problem to minimize the age of global parameter, energy consumption and FL service latency simultaneously. Taking account of the dynamic WCPN environment, the optimization target is modeled as a Markov decision process (MDP), and Proximal Policy Optimization (PPO) algorithm is used to achieve the optimal solution. Finally, a large number of numerical simulation results prove the superiority and outstanding performance of the proposed method.

This publication has 22 references indexed in Scilit: