Adversarial Reinforcement Learning in Dynamic Channel Access and Power Control
This addresses security risks in wireless communications for systems using RL, but is incremental as it builds on existing adversarial RL methods.
The paper studied the vulnerabilities of deep reinforcement learning (RL) agents in wireless resource allocation to adversarial attacks, showing that a designed RL-based jammer significantly degrades users' sum rate, and proposed an ensemble defense strategy using reloaded models with minimum transition correlation.
Deep reinforcement learning (DRL) has recently been used to perform efficient resource allocation in wireless communications. In this paper, the vulnerabilities of such DRL agents to adversarial attacks is studied. In particular, we consider multiple DRL agents that perform both dynamic channel access and power control in wireless interference channels. For these victim DRL agents, we design a jammer, which is also a DRL agent. We propose an adversarial jamming attack scheme that utilizes a listening phase and significantly degrades the users' sum rate. Subsequently, we develop an ensemble policy defense strategy against such a jamming attacker by reloading models (saved during retraining) that have minimum transition correlation.