Reconfigurable Intelligent Surfaces Assisted NLOS Radar Anti Jamming Using Deep Reinforcement Learning

The complexity of the radar environment increases with technological advancement, especially when considering the difficulties presented by repeating jammers. These jammers can impede radar detection, especially when they create false targets in non-line-of-sight (NLOS) situations. This study focuses on optimizing the phase shifts of Reconfigurable Intelligent Surfaces (RIS) to address the problem of NLOS between a target and radar for detection in order to address these NLOS issues. Specifically, we investigate RIS phase shift optimization using a Genetic Algorithm (GA) to address the challenges posed by repeating jammers across various dynamic scenarios. Our objective is to increase the radar system's ability to detect actual targets in non-LOS scenarios when repeater jammers are present in the environment. According to the experimental results, this method offers a practical way to mitigate the effects of repeater jammers by improving radar detection performance in NLOS environments.