Federated Parameter-Efficient Adaptation for Interference Mitigation at the Wireless Edge

2026-04-17

AI Summary

This work addresses the challenge of heterogeneous co-channel interference in dense wireless networks, which exhibits significant spatial variation and renders conventional federated learning inefficient due to high communication overhead from transmitting full model updates. To overcome this, the authors propose Fed-LoRA, a novel framework that integrates Low-Rank Adaptation (LoRA) into the dilated convolutional layers of a temporal convolutional network. By freezing the backbone model and training only a small set of low-rank adapter parameters—comprising merely 5.1% of the backbone’s size—the approach preserves both local interference specificity and temporal modeling capability. Evaluated under non-IID interference conditions, Fed-LoRA reduces communication costs by 20×, achieves a 12.6% lower average bit error rate, and significantly outperforms local fine-tuning on data-scarce nodes, effectively avoiding the performance collapse observed in full-model federated learning.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Large Vision Models
• Natural Language Processing: Learning & Optimization for NLP

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: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Dense wireless deployments face co-channel interference from heterogeneous sources that vary across base stations (gNBs in 5G). While centralized DNN-based approaches to interference mitigation have shown strong performance, deploying and adapting these models across distributed gNBs via federated learning (FL) requires transmitting full model updates each round, resulting in a cost that scales poorly with network density. Parameter-efficient fine-tuning (PEFT) reduces this burden by training and communicating only a small fraction of parameters. While traditionally applied to large foundation models, we adapt Low-Rank Adaptation (LoRA) to temporal convolutional neural network architectures for interference suppression, placing low-rank adapters on the dilated convolutional layers. This placement enables LoRA to learn local interference-specific temporal patterns, while the frozen backbone retains the shared signal extraction capability. These lightweight adapters (5.1\% of backbone parameters) are federated via FedAvg, reducing per-round communication by up to 20$\times$ compared to federating full model updates. We evaluate various PEFT strategies across simulated distributed gNBs with non-IID interference environments. Results show that local LoRA achieves 12.8\% average BER improvement over the frozen backbone, while Fed-LoRA achieves comparable performance (12.6\%). Fed-LoRA outperforms local adaptation on data-starved nodes where federated knowledge transfer compensates for limited samples, all while avoiding the catastrophic degradation observed with full-model FedAvg under heterogeneous conditions.

Problem

• federated learning
• interference mitigation
• parameter-efficient fine-tuning
• wireless edge
• non-IID

Innovation

• Federated Learning
• Parameter-Efficient Fine-Tuning
• LoRA
• Interference Mitigation
• Temporal Convolutional Network