Fed-PELAD: Communication-Efficient Federated Learning for Massive MIMO CSI Feedback with Personalized Encoders and a LoRA-Adapted Shared Decoder

2025-10-29

AI Summary

To address the high communication overhead, data heterogeneity, and privacy leakage inherent in deep learning-based CSI feedback for massive MIMO systems, this paper proposes a lightweight personalized federated learning (FL) framework. The method introduces a personalized encoder coupled with a LoRA-adapted shared decoder, transmitting only low-rank adaptation parameters; it further incorporates a learning-rate-calibrated alternating freezing strategy to stabilize local training and global aggregation. By synergistically integrating FL, model personalization, LoRA fine-tuning, and controlled gradient updates, the framework achieves efficient, privacy-preserving CSI feedback. Evaluated on the 3GPP-standardized channel model, it reduces uplink communication overhead by 42.97%, improves CSI reconstruction accuracy by 1.2 dB under heterogeneous data conditions, and simultaneously enhances efficiency, generalization, and privacy protection.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Multiagent Systems: Multiagent Learning

Application Category

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

Abstract

This paper addresses the critical challenges of communication overhead, data heterogeneity, and privacy in deep learning for channel state information (CSI) feedback in massive MIMO systems. To this end, we propose Fed-PELAD, a novel federated learning framework that incorporates personalized encoders and a LoRA-adapted shared decoder. Specifically, personalized encoders are trained locally on each user equipment (UE) to capture device-specific channel characteristics, while a shared decoder is updated globally via the coordination of the base station (BS) by using Low-Rank Adaptation (LoRA). This design ensures that only compact LoRA adapter parameters instead of full model updates are transmitted for aggregation. To further enhance convergence stability, we introduce an alternating freezing strategy with calibrated learning-rate ratio during LoRA aggregation. Extensive simulations on 3GPP-standard channel models demonstrate that Fed-PELAD requires only 42.97% of the uplink communication cost compared to conventional methods while achieving a performance gain of 1.2 dB in CSI feedback accuracy under heterogeneous conditions.

Problem

• Reducing communication overhead in massive MIMO CSI feedback systems
• Addressing data heterogeneity across user equipment in federated learning
• Maintaining privacy while improving CSI feedback accuracy performance

Innovation

• Personalized encoders capture device-specific channel characteristics locally
• Shared decoder uses LoRA adaptation for global parameter aggregation
• Alternating freezing strategy stabilizes convergence with calibrated learning rates