Practical Adversarial Attacks Against AI-Driven Power Allocation in a Distributed MIMO Network

TL;DR

This paper demonstrates that adversarial attacks can significantly compromise AI-based power control in distributed MIMO networks, highlighting the need for robust defense mechanisms.

Contribution

It reveals the vulnerability of AI-driven power allocation in D-MIMO networks to adversarial attacks and emphasizes the importance of developing effective defense strategies.

Findings

Adversarial samples can drastically reduce network performance.

Threat level of adversarial attacks exceeds conventional threats.

Simulations confirm attack effectiveness and defense necessity.

Abstract

In distributed multiple-input multiple-output (D-MIMO) networks, power control is crucial to optimize the spectral efficiencies of users and max-min fairness (MMF) power control is a commonly used strategy as it satisfies uniform quality-of-service to all users. The optimal solution of MMF power control requires high complexity operations and hence deep neural network based artificial intelligence (AI) solutions are proposed to decrease the complexity. Although quite accurate models can be achieved by using AI, these models have some intrinsic vulnerabilities against adversarial attacks where carefully crafted perturbations are applied to the input of the AI model. In this work, we show that threats against the target AI model which might be originated from malicious users or radio units can substantially decrease the network performance by applying a successful adversarial sample, even in the most constrained circumstances. We also demonstrate that the risk associated with these kinds of adversarial attacks is higher than the conventional attack threats. Detailed simulations reveal the effectiveness of adversarial attacks and the necessity of smart defense techniques.