Improving Wireless Federated Learning via Joint Downlink-Uplink Beamforming over Analog Transmission

2025-02-04

AI Summary

To address the slow convergence of over-the-air federated learning (AirFL) over time-varying noisy wireless channels, this paper proposes a joint downlink-uplink beamforming (JDUBF) framework—first integrating coordinated dual-link beamforming into the end-to-end AirFL model update dynamics. To enable efficient online optimization, we design a low-complexity JDUBF algorithm that combines greedy decomposition, block coordinate descent, and closed-form gradient updates, with projected gradient steps ensuring feasibility. Theoretical analysis quantifies how channel time-variation degrades convergence rate. Simulation results demonstrate that JDUBF significantly accelerates multi-round training convergence and improves final model accuracy compared to conventional independent single-link beamforming. These findings validate the critical role of joint downlink-uplink design in enhancing both the robustness and efficiency of AirFL under practical wireless conditions.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Planning, Routing, and Scheduling: Planning/Scheduling and Learning

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated learning (FL) over wireless networks using analog transmission can efficiently utilize the communication resource but is susceptible to errors caused by noisy wireless links. In this paper, assuming a multi-antenna base station, we jointly design downlink-uplink beamforming to maximize FL training convergence over time-varying wireless channels. We derive the round-trip model updating equation and use it to analyze the FL training convergence to capture the effects of downlink and uplink beamforming and the local model training on the global model update. Aiming to maximize the FL training convergence rate, we propose a low-complexity joint downlink-uplink beamforming (JDUBF) algorithm, which adopts a greedy approach to decompose the multi-round joint optimization and convert it into per-round online joint optimization problems. The per-round problem is further decomposed into three subproblems over a block coordinate descent framework, where we show that each subproblem can be efficiently solved by projected gradient descent with fast closed-form updates. An efficient initialization method that leads to a closed-form initial point is also proposed to accelerate the convergence of JDUBF. Simulation demonstrates that JDUBF substantially outperforms the conventional separate-link beamforming design.

Problem

• Optimizes wireless federated learning using analog transmission
• Designs joint downlink-uplink beamforming for better convergence
• Proposes low-complexity JDUBF algorithm for efficient optimization

Innovation

• Joint downlink-uplink beamforming
• Low-complexity JDUBF algorithm
• Block coordinate descent framework