摘要
针对电子战中雷达和干扰装备双方对抗行为的先验知识贫乏、可预测性差等问题,本文围绕复杂对抗环境数字仿真与抗干扰策略自适应学习技术展开研究,构建动态博弈对抗行为自学习智能仿真系统,为博弈双方技术发展提供训练平台,充分发挥决策自主性,实现对抗策略的自主学习和效果的实时评估量化,为雷达干扰对抗边界问题提供理论与数据支撑。
Aiming at the problems of scarce prior knowledge and poor predictability of the countermeasure behaviors between radar and jamming equipment in electronic warfare,this paper focuses on the research of digital simulation of complex countermeasure environments and adaptive anti-jamming strategy learning technology.An intelligent simulation system featuring self-learning of dynamic game-based countermeasure behaviors is constructed to provide a training platform for the technical development of both sides in the game.This system fully exerts the autonomy of decision-making and realizes the autonomous learning of countermeasure strategies and real-time assessment and quantification of their effects,providing theoretical and data support for the boundary issues of radar jamming and counter-jamming.
作者
翟颖
ZHAI Ying(No.20 Research Institute of CETC,Xi'an 710068)
出处
《火控雷达技术》
2025年第4期61-65,共5页
Fire Control Radar Technology
关键词
对抗行为自学习
环境数字仿真
深度Q学习算法
self-learning of countermeasure behaviors
digital environment simulation
deep Q-learning algorithm
