Radar and Jammer Intelligent Game under Jamming Power Dynamic Allocation

In modern electronic warfare, the intelligence of jammer greatly worsens anti-jamming performance traditional passive suppression methods. How to actively design strategies deal with intelligent jammers is crucial radar system. existing research on strategies’ design, assumption too ideal. To establish a model that closer real this paper explores game between subpulse-level frequency-agile (FA) and transmit/receive time-sharing under jamming power dynamic allocation. Firstly, discrete allocation established, multiple-round sequential interaction described based an extensive-form game. A detection probability calculation method signal-to-interference-pulse-noise ratio (SINR) accumulation gain criterion (SAGC) proposed evaluate results. Secondly, considering competition has feature imperfect information, we utilized neural fictitious self-play (NFSP), end-to-end deep reinforcement learning (DRL) algorithm, find Nash equilibrium (NE) Finally, simulation results showed can converge approximate NE established model. The are better than elementary from perspective probability. addition, comparing NFSP Q-network (DQN) illustrates effectiveness in solving information games.