Over-the-Air Federated Learning with Enhanced Privacy

TL;DR

This paper introduces a novel federated learning framework that enhances privacy and reduces bandwidth usage through over-the-air aggregation with pairwise cancellable noises, ensuring stronger privacy and efficient communication.

Contribution

The paper proposes a new wireless federated learning scheme using pairwise cancellable artificial noises for improved privacy and bandwidth efficiency, with analytical guarantees.

Findings

Stronger privacy protection compared to existing schemes.

Analytical proof of secrecy capacity.

Convergence rate of the proposed algorithm.

Abstract

Federated learning (FL) has emerged as a promising learning paradigm in which only local model parameters (gradients) are shared. Private user data never leaves the local devices thus preserving data privacy. However, recent research has shown that even when local data is never shared by a user, exchanging model parameters without protection can also leak private information. Moreover, in wireless systems, the frequent transmission of model parameters can cause tremendous bandwidth consumption and network congestion when the model is large. To address this problem, we propose a new FL framework with efficient over-the-air parameter aggregation and strong privacy protection of both user data and models. We achieve this by introducing pairwise cancellable random artificial noises (PCR-ANs) on end devices. As compared to existing over-the-air computation (AirComp) based FL schemes, our design provides stronger privacy protection. We analytically show the secrecy capacity and the convergence rate of the proposed wireless FL aggregation algorithm.