CoCo-Fed: A Unified Framework for Memory- and Communication-Efficient Federated Learning at the Wireless Edge

TL;DR

CoCo-Fed is a federated learning framework designed for wireless edge devices that reduces memory and communication costs through gradient compression and superposition, while ensuring convergence even in challenging unsupervised scenarios.

Contribution

It introduces a unified approach combining gradient low-rank approximation and subspace superposition to enhance federated learning efficiency at the wireless edge.

Findings

Significantly reduces backhaul traffic and memory usage.

Achieves robust convergence in non-IID and unsupervised settings.

Outperforms state-of-the-art methods in simulation experiments.

Abstract

The deployment of large-scale neural networks within the Open Radio Access Network (O-RAN) architecture is pivotal for enabling native edge intelligence. However, this paradigm faces two critical bottlenecks: the prohibitive memory footprint required for local training on resource-constrained gNBs, and the saturation of bandwidth-limited backhaul links during the global aggregation of high-dimensional model updates. To address these challenges, we propose CoCo-Fed, a novel Compression and Combination-based Federated learning framework that unifies local memory efficiency and global communication reduction. Locally, CoCo-Fed breaks the memory wall by performing a double-dimension down-projection of gradients, adapting the optimizer to operate on low-rank structures without introducing additional inference parameters/latency. Globally, we introduce a transmission protocol based on orthogonal subspace superposition, where layer-wise updates are projected and superimposed into a single consolidated matrix per gNB, drastically reducing the backhaul traffic. Beyond empirical designs, we establish a rigorous theoretical foundation, proving the convergence of CoCo-Fed even under unsupervised learning conditions suitable for wireless sensing tasks. Extensive simulations on an angle-of-arrival estimation task demonstrate that CoCo-Fed significantly outperforms state-of-the-art baselines in both memory and communication efficiency while maintaining robust convergence under non-IID settings.