Reinforcement Learning Based Techniques for Radar Anti-Jamming

摘要

With the increasing dependence on Radar technology in modern warfare, the ability of a radar to operate in a hostile environment is becoming more important. For this purpose, modern radar systems incorporate technologies such as adaptive beam-forming, frequency hopping and adaptive waveforms. More recently, researchers have been looking at combining all these technologies to obtain a more jamming resilient radar system. However, such systems can have a very large number of states and designing an optimal strategy for selection of next state becomes a difficult problem. To overcome this issue, recently reinforcement learning has been applied for optimal state selection to avoid jamming. This paper provides an overview of this problem and provides a survey of literature proposing application of reinforcement learning to radar to overcome jamming.