Dynamic Clustering and Power Control for Two-Tier Wireless Federated Learning

TL;DR

This paper proposes a joint communication and learning framework for wireless federated learning that uses dynamic clustering and power control to improve efficiency and convergence, leveraging over-the-air computation in a two-tier structure.

Contribution

It introduces a novel two-tier wireless FL scheme with dynamic clustering and power control, and provides convergence analysis and an efficient optimization method.

Findings

The scheme achieves faster convergence compared to baselines.

Numerical results validate the effectiveness of the proposed method.

The hierarchical clustering improves communication efficiency.

Abstract

Federated learning (FL) has been recognized as a promising distributed learning paradigm to support intelligent applications at the wireless edge, where a global model is trained iteratively through the collaboration of the edge devices without sharing their data. However, due to the relatively large communication cost between the devices and parameter server (PS), direct computing based on the information from the devices may not be resource efficient. This paper studies the joint communication and learning design for the over-the-air computation (AirComp)-based two-tier wireless FL scheme, where the lead devices first collect the local gradients from their nearby subordinate devices, and then send the merged results to the PS for the second round of aggregation. We establish a convergence result for the proposed scheme and derive the upper bound on the optimality gap between the expected and optimal global loss values. Next, based on the device distance and data importance, we propose a hierarchical clustering method to build the two-tier structure. Then, with only the instantaneous channel state information (CSI), we formulate the optimality gap minimization problem and solve it by using an efficient alternating minimization method. Numerical results show that the proposed scheme outperforms the baseline ones.