Wireless Federated Learning over Resource-Constrained Networks: Digital versus Analog Transmissions

TL;DR

This paper compares digital and analog wireless transmission schemes for federated learning, analyzing their differences, scenarios, and performance under resource constraints, with a focus on convergence, efficiency, and error sensitivity.

Contribution

It provides a unified framework and convergence analysis for digital and analog schemes, highlighting their fundamental differences and optimal scenarios in resource-constrained wireless FL.

Findings

Analog communication supports over-the-air computation with better spectrum use.

Digital schemes are limited by bandwidth and decouple communication from computation.

Sampling optimization significantly improves FL performance.

Abstract

To enable wireless federated learning (FL) in communication resource-constrained networks, two communication schemes, i.e., digital and analog ones, are effective solutions. In this paper, we quantitatively compare these two techniques, highlighting their essential differences as well as respectively suitable scenarios. We first examine both digital and analog transmission schemes, together with a unified and fair comparison framework under imbalanced device sampling, strict latency targets, and transmit power constraints. A universal convergence analysis under various imperfections is established for evaluating the performance of FL over wireless networks. These analytical results reveal that the fundamental difference between the digital and analog communications lies in whether communication and computation are jointly designed or not. The digital scheme decouples the communication design from FL computing tasks, making it difficult to support uplink transmission from massive devices with limited bandwidth and hence the performance is mainly communication-limited. In contrast, the analog communication allows over-the-air computation (AirComp) and achieves better spectrum utilization. However, the computation-oriented analog transmission reduces power efficiency, and its performance is sensitive to computation errors from imperfect channel state information (CSI). Furthermore, device sampling for both schemes are optimized and differences in sampling optimization are analyzed. Numerical results verify the theoretical analysis and affirm the superior performance of the sampling optimization.