Robust Adversarial Attacks Against DNN-Based Wireless Communication Systems

TL;DR

This paper introduces a novel, robust adversarial attack method for DNN-based wireless communication systems that remains effective even against defenses, highlighting significant vulnerabilities in current wireless AI applications.

Contribution

The work presents a universal, undetectable adversarial attack using GANs tailored for wireless DNNs, effective in both white-box and black-box scenarios, and resilient against countermeasures.

Findings

Attack outperforms existing methods against defenses

Effective in white-box and black-box scenarios

Remains undetectable with GAN-based constraints

Abstract

Deep Neural Networks (DNNs) have become prevalent in wireless communication systems due to their promising performance. However, similar to other DNN-based applications, they are vulnerable to adversarial examples. In this work, we propose an input-agnostic, undetectable, and robust adversarial attack against DNN-based wireless communication systems in both white-box and black-box scenarios. We design tailored Universal Adversarial Perturbations (UAPs) to perform the attack. We also use a Generative Adversarial Network (GAN) to enforce an undetectability constraint for our attack. Furthermore, we investigate the robustness of our attack against countermeasures. We show that in the presence of defense mechanisms deployed by the communicating parties, our attack performs significantly better compared to existing attacks against DNN-based wireless systems. In particular, the results demonstrate that even when employing well-considered defenses, DNN-based wireless communications are vulnerable to adversarial attacks.