Privacy-Preserving Semantic Communications via Multi-Task Learning and Adversarial Perturbations

TL;DR

This paper introduces a deep learning framework for semantic communications that enhances task performance while explicitly limiting information leakage to eavesdroppers through adversarial training and perturbations.

Contribution

It proposes a joint multi-task semantic communication system with an adversarial min-max training scheme and an auxiliary perturbation layer for end-to-end privacy protection.

Findings

Significant reduction in eavesdropper inference success.

Improved semantic accuracy and reconstruction quality.

Effective privacy preservation with minimal impact on legitimate performance.

Abstract

Semantic communications conveys task-relevant meaning rather than focusing solely on message reconstruction, improving bandwidth efficiency and robustness for next-generation wireless systems. However, learned semantic representations can still leak sensitive information to unintended receivers (eavesdroppers). This paper presents a deep learning-based semantic communication framework that jointly supports multiple receiver tasks while explicitly limiting semantic leakage to an eavesdropper. The legitimate link employs a learned encoder at the transmitter, while the receiver trains decoders for semantic inference and data reconstruction. The security problem is formulated via an iterative min-max optimization in which an eavesdropper is trained to improve its semantic inference, while the legitimate transmitter-receiver pair is trained to preserve task performance while reducing the eavesdropper's success. We also introduce an auxiliary layer that superimposes a cooperative, adversarially crafted perturbation on the transmitted waveform to degrade semantic leakage to an eavesdropper. Performance is evaluated over Rayleigh fading channels with additive white Gaussian noise using MNIST and CIFAR-10 datasets. Semantic accuracy and reconstruction quality improve with increasing latent dimension, while the min-max mechanism reduces the eavesdropper's inference performance significantly without degrading the legitimate receiver. The perturbation layer is successful in reducing semantic leakage even when the legitimate link is trained only for its own task. This comprehensive framework motivates semantic communication designs with tunable, end-to-end privacy against adaptive adversaries in realistic wireless settings.