Mixed-Timescale Differential Coding for Downlink Model Broadcast in Wireless Federated Learning

2026-07-14

AI Summary

This work addresses the challenge in wireless federated learning where downlink transmission failures prevent devices from timely acquiring the latest global model, thereby severely degrading training efficiency and accuracy. To mitigate this issue, the paper proposes a Mixed Timescale Differential Coding (MTDC) scheme that introduces, for the first time, a dual-timescale differential encoding mechanism. MTDC integrates dynamic reference model updates with an age-aware device scheduling strategy, enabling devices to reconstruct the most recent global model even if they miss intermediate broadcasts between full model transmissions. Theoretical analysis and experimental results demonstrate that, under identical communication resources, MTDC significantly outperforms existing approaches—particularly in high packet-loss scenarios—by accelerating convergence and improving model accuracy.

Abstract

In standard federated learning systems, the parameter server broadcasts the global model to the participating devices in every iteration. Motivated by the temporal correlation between consecutive global models, differential coding can be applied to global model dissemination to reduce the information magnitude, thereby enabling communication with fewer quantization bits. However, due to wireless link failures, devices may occasionally miss differential updates and consequently fail to reconstruct the global model. As a result, they either continue local training based on an outdated model or remain idle until the next full-model broadcast becomes available. To address this challenge, we propose a mixed-timescale differential coding (MTDC) scheme that performs differential coding at two different levels by adjusting the reference model. With MTDC, a device can reconstruct the latest global model between two full-model broadcasts even if it misses a differential update. We provide a convergence analysis that motivates the design of an age-aware variant of MTDC, along with a device scheduling policy to further improve communication efficiency. Simulation results demonstrate that the proposed MTDC schemes achieve superior learning performance compared to baseline methods under similar communication resource budgets in the presence of downlink transmission failures.

Problem

• wireless federated learning
• differential coding
• downlink transmission failures
• global model reconstruction
• communication efficiency

Innovation

• mixed-timescale differential coding
• wireless federated learning
• downlink model broadcast
• age-aware coding
• device scheduling