Energy-Efficient Federated Learning for Edge Real-Time Vision via Joint Data, Computation, and Communication Design

TL;DR

This paper introduces FedDPQ, an energy-efficient federated learning framework for real-time vision on edge devices, combining data augmentation, pruning, quantization, and power control to optimize training under wireless constraints.

Contribution

It presents the first joint optimization of data, computation, and communication in federated learning for edge vision, with a novel energy-convergence model and Bayesian tuning algorithm.

Findings

FedDPQ improves convergence speed and energy efficiency in real-world CV tasks.

The framework effectively mitigates wireless transmission issues through adaptive power control.

Joint optimization outperforms traditional methods in resource-constrained environments.

Abstract

Emerging real-time computer vision (CV) applications on wireless edge devices demand energy-efficient and privacy-preserving learning. Federated learning (FL) enables on-device training without raw data sharing, yet remains challenging in resource-constrained environments due to energy-intensive computation and communication, as well as limited and non-i.i.d. local data. We propose FedDPQ, an ultra energy-efficient FL framework for real-time CV over unreliable wireless networks. FedDPQ integrates diffusion-based data augmentation, model pruning, communication quantization, and transmission power control to enhance training efficiency. It expands local datasets using synthetic data, reduces computation through pruning, compresses updates via quantization, and mitigates transmission outages with adaptive power control. We further derive a closed-form energy-convergence model capturing the coupled impact of these components, and develop a Bayesian optimization(BO)-based algorithm to jointly tune data augmentation strategy, pruning ratio, quantization level, and power control. To the best of our knowledge, this is the first work to jointly optimize FL performance from the perspectives of data, computation, and communication under unreliable wireless conditions. Experiments on representative CV tasks show that FedDPQ achieves superior convergence speed and energy efficiency.