Model Splitting Enhanced Communication-Efficient Federated Learning for CSI Feedback

TL;DR

This paper proposes a model splitting method with shared and local models to reduce communication overhead in federated CSI feedback, incorporating a pipeline module to accelerate training.

Contribution

It introduces a novel model splitting approach with boundary parameter exchange reduction and a pipeline module for faster training in federated CSI feedback.

Findings

Significantly reduces communication overhead compared to benchmarks.

Accelerates training time with the pipeline module.

Maintains effective CSI feedback performance.

Abstract

Recent advancements have introduced federated machine learning-based channel state information (CSI) compression before the user equipments (UEs) upload the downlink CSI to the base transceiver station (BTS). However, most existing algorithms impose a high communication overhead due to frequent parameter exchanges between UEs and BTS. In this work, we propose a model splitting approach with a shared model at the BTS and multiple local models at the UEs to reduce communication overhead. Moreover, we implant a pipeline module at the BTS to reduce training time. By limiting exchanges of boundary parameters during forward and backward passes, our algorithm can significantly reduce the exchanged parameters over the benchmarks during federated CSI feedback training.