| 摘要: |
| 针对多功能雷达和认知电子战的快速发展所导致传统干扰决策方法难以适应现代化战争的问题，提出了一种基于Q-Learning的多功能雷达认知干扰决策方法。通过对比认知思想和干扰决策原理，将Q-Learning运用于认知干扰决策中并提出了认知干扰决策的算法步骤。以某多功能雷达为基础，通过分析其工作状态及对应干扰样式构建雷达状态转移图，通过仿真试验分析了各参数对决策性能的影响，为应对实际战场提供参考。仿真了在新状态加入下的决策过程、实际战场中转移概率对决策路径的影响以及四种主要干扰决策方法的决策性能对比。试验表明，该方法能够通过自主学习干扰效果完成干扰决策，更加贴合实际战场，对认知电子战的发展有一定的借鉴意义。 |
| 关键词: 多功能雷达 认知电子战 干扰决策 Q-Learning 强化学习 |
| DOI： |
|
| 基金项目:电子信息系统复杂电磁环境效应(CEMEE)国家重点实验室项目(2018Z0202B) |
|
| A cognitive jamming decision method for multi-functional radar based on Q-Learning |
| ZHANG Bokai,ZHU Weigang |
| (Department of Graduate Management,Space Engineering University，Beijing 101416，China;Department of Electronic and Optical Engineering,Space Engineering University，Beijing 101416，China) |
| Abstract: |
| For the problem that traditional jamming decision-making methods are difficult to adapt to modern warfare due to the rapid development of multi-functional radar and cognitive electronic warfare,a cognitive jamming decision-making method based Q-Learning for multi-functional radar is proposed.By comparing cognitive thought and jamming decision-making principle,Q-Learning is applied to cognitive jamming decision-making and the algorithm steps of cognitive jamming decision-making are proposed.On the basis of a multi-functional radar,the radar state transition diagram is constructed by analyzing its working state and corresponding jamming modes,and the influence of each parameter on the decision-making performance is analyzed through simulation tests,which can provide reference for dealing with actual battlefield.The decision-making process under the new state,the influence of transition probability on decision-making path in actual battlefield and the performance comparison of four main jamming decision-making methods are simulated.Tests show that this method can accomplish jamming decision through self-learning jamming effect,which is more suitable for the actual battlefield and has certain reference significance for the development of cognitive electronic warfare. |
| Key words: multi-functional radar cognitive electronic warfare jamming decision-making Q-Learning reinforcement learning |
