基于遗传-群体智能融合算法的干扰决策方法  

An Interference Decision-making Method Based onGenetic-Population Intelligent Fusion Algorithm

作　　者：阎潇 王青平[1] 胡卫东[1] 朱虹宇 王超[1] 施庆展 YAN Xiao;WANG Qingping;HU Weidong;ZHU Hongyu;WANG Chao;SHI Qingzhan(College of Electronic Science,National University of Defense Technology,Changsha 410003,China)

机构地区：[1]国防科技大学电子科学学院,长沙410003

出　　处：《电讯技术》2025年第2期275-282,共8页Telecommunication Engineering

基　　金：国家自然科学基金资助项目(62301570);国防科技创新特区计划(163计划)。

摘　　要：针对无人机集群协同多目标干扰实时决策问题,提出了一种基于遗传-群体智能融合算法。首先根据多功能雷达工作模式及特点构建自适应协同干扰效益评估函数,分析动态干扰决策这个NP-hard问题的特点,建立多要素的干扰资源分配模型,然后采用遗传-群体智能融合算法进行优化求解。所提算法通过分布式决策实现算法初期的次优解快速收敛,再通过集中式优化实现算法全局寻优,通过种群内个体间的竞争与协作降低算法陷入局部最优的风险。仿真结果表明,融合算法相比较于遗传算法和改进蚁群算法收敛迭代次数分别减少了42.25%和15.43%,在实时处理性能方面相较于遗传算法提升了9.53%,具有更好的收敛效率和实时处理能力。For the real-time decision-making problem of unmanned aerial vehicle(UAV)swarm cooperative multi-target jamming,a genetic-group intelligent fusion algorithm is proposed.Firstly,the adaptive cooperative jamming benefit evaluation function is constructed according to the working mode and characteristics of the multi-function radar.The characteristics of the NP-hard problem of dynamic jamming decision-making are analyzed,and the multi-factor jamming resource allocation model is established.Then,the genetic-population intelligent fusion algorithm is used to optimize the solution.The proposed algorithm realizes the rapid convergence of the sub-optimal solution in the initial stage of the algorithm through distributed decision-making,and then realizes the global optimization of the algorithm through centralized optimization,and reduces the risk of the algorithm falling into local optimum through the competition and cooperation among individuals in the population.The simulation results show that the fusion algorithm reduces the number of convergence iterations by 42.25%and 15.43%respectively compared with the genetic algorithm and the improved ant colony algorithm.It improves the real-time processing performance by 9.53%compared with the genetic algorithm.The proposed fusion algorithm has better convergence efficiency and real-time processing ability.

关 键 词：无人机集群 协同多目标干扰 干扰资源分配 遗传算法 实时决策 群体智能 

分 类 号：TN974[电子电信—信号与信息处理]