Multi-Agent Adversarial Attacks for Multi-Channel Communications

no code implementations • 22 Jan 2022 • Juncheng Dong, Suya Wu, Mohammadreza Sultani, Vahid Tarokh

In particular, by modeling the adversaries as learning agents, we show that the proposed MAAS is able to successfully choose the transmitted channel(s) and their respective allocated power(s) without any prior knowledge of the sender strategy.

Reinforcement Learning (RL)