Semantic Leakage and Privacy Preservation in Relay-Assisted Semantic Communications
For semantic communication systems with untrusted relays, this work addresses a critical privacy issue by providing a method to selectively suppress semantic leakage without degrading performance for the intended receiver.
The paper identifies a privacy vulnerability in relay-assisted semantic communication where a relay can infer semantic meaning from latent representations, and proposes an iterative adversarial training framework that enlarges the semantic accuracy gap between the legitimate receiver and the eavesdropper, achieving privacy preservation while maintaining reconstruction fidelity.
Semantic communication (SemCom) has emerged as a promising paradigm in which the transmission of task-relevant information is prioritized over raw data, enabling efficient and robust communication under resource and channel constraints. In this paper, the privacy implications of relay-assisted SemCom systems are studied, where the intermediate relay node operates directly on learned latent representations. It is shown that the relay, even without access to source data, can reliably infer semantic meaning and reconstruct signals with performance comparable to that of the legitimate receiver, revealing a fundamental privacy vulnerability of semantic representations. To address this issue, an iterative adversarial training framework is proposed in which a strong, adaptively trained eavesdropper at the relay is explicitly accounted for. The proposed approach alternates between optimizing the relay's eavesdropping function and the legitimate system, resulting in representations that preserve semantic decoding performance at the intended receiver while degrading semantic inference at the relay. The semantic accuracy gap between the legitimate receiver and the eavesdropper is significantly enlarged across channel conditions. Importantly, this protection is achieved in a stealthy manner, with high reconstruction fidelity maintained while semantic leakage is selectively suppressed.