EPFL-REMNet: Efficient Personalized Federated Digital Twin Towards 6G Heterogeneous Radio Environment

TL;DR

This paper introduces EPFL-REMNet, a personalized federated learning framework that efficiently constructs high-fidelity digital twins of 6G heterogeneous radio environments, addressing Non-IID data challenges and improving accuracy and communication efficiency.

Contribution

EPFL-REMNet's novel shared backbone and personalized head architecture enhances digital twin fidelity and reduces communication overhead in Non-IID federated learning for 6G radio environments.

Findings

Achieves higher accuracy than FedAvg and state-of-the-art methods.

Reduces uplink communication overhead.

Improves performance for long-tail clients.

Abstract

Radio Environment Map (REM) is transitioning from 5G homogeneous environments to B5G/6G heterogeneous landscapes. However, standard Federated Learning (FL), a natural fit for this distributed task, struggles with performance degradation in accuracy and communication efficiency under the non-independent and identically distributed (Non-IID) data conditions inherent to these new environments. This paper proposes EPFL-REMNet, an efficient personalized federated framework for constructing a high-fidelity digital twin of the 6G heterogeneous radio environment. The proposed EPFL-REMNet employs a"shared backbone + lightweight personalized head" model, where only the compressed shared backbone is transmitted between the server and clients, while each client's personalized head is maintained locally. We tested EPFL-REMNet by constructing three distinct Non-IID scenarios (light, medium, and heavy) based on radio environment complexity, with data geographically partitioned across 90 clients. Experimental results demonstrate that EPFL-REMNet simultaneously achieves higher digital twin fidelity (accuracy) and lower uplink overhead across all Non-IID settings compared to standard FedAvg and recent state-of-the-art methods. Particularly, it significantly reduces performance disparities across datasets and improves local map accuracy for long-tail clients, enhancing the overall integrity of digital twin.