Unanticipated Adversarial Robustness of Semantic Communication

2026-03-25

AI Summary

This work challenges the prevailing notion that semantic communication systems are inherently vulnerable to adversarial attacks by theoretically demonstrating, for the first time, that DeepJSCC-based semantic communication systems possess intrinsic adversarial robustness. To this end, the authors introduce two novel attack strategies—structure-aware fragile set attack and progressive gradient ascent attack—and establish a lower bound on the required attack power through Lipschitz smoothness analysis of the decoder. This analysis reveals that decoder smoothness induced by noise-aware training is a key source of robustness. Experimental results confirm that, to achieve comparable distortion levels, adversarial attacks on semantic communication systems require 14–16 times higher power than those on classical systems, thereby substantiating their significantly enhanced adversarial robustness.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Search and Optimization: Adversarial Search

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Semantic communication, enabled by deep joint source-channel coding (DeepJSCC), is widely expected to inherit the vulnerability of deep learning to adversarial perturbations. This paper challenges this prevailing belief and reveals a counterintuitive finding: semantic communication systems exhibit unanticipated adversarial robustness that can exceed that of classical separate source-channel coding systems. On the theoretical front, we establish fundamental bounds on the minimum attack power required to induce a target distortion, overcoming the analytical intractability of highly nonlinear DeepJSCC models by leveraging Lipschitz smoothness. We prove that the implicit regularization from noisy training forces decoder smoothness, a property that inherently provides built-in protection against adversarial attacks. To enable rigorous and fair comparison, we develop two novel attack methodologies that address previously unexplored vulnerabilities: a structure-aware vulnerable set attack that, for the first time, exploits graph-theoretic vulnerabilities in LDPC codes to induce decoding failure with minimal energy, and a progressive gradient ascent attack that leverages the differentiability of DeepJSCC to efficiently find minimum-power perturbations. Designing such attacks is challenging, as classical systems lack gradient information while semantic systems require navigating high-dimensional, non-convex spaces; our methods fill these critical gaps in the literature. Extensive experiments demonstrate that semantic communication requires up to $14$-$16\times$ more attack power to achieve the same distortion as classical systems, empirically substantiating its superior robustness.

Problem

• semantic communication
• adversarial robustness
• DeepJSCC
• adversarial perturbations
• channel coding

Innovation

• semantic communication
• adversarial robustness
• DeepJSCC
• Lipschitz smoothness
• vulnerable set attack