面向动态通信用户数量的多模Q学习干扰策略  

Jamming Strategy Based on Multi-mode Q Learning for Dynamically Changing Number of Communication Users

作　　者：贺兴 任国春 冯智斌 龚玉萍 王路广 朱卓丹 HE Xing;REN Guochun;FENG Zhibin;GONG Yuping;WANG Luguang;ZHU Zhuodan(College of Communications Engineering,Army Engineering University of PLA,Nanjing 210007,China;Unit 94676 of PLA,Shanghai 202178,China;Unit 93117 of PLA,Nanjing 211515,China)

机构地区：[1]陆军工程大学通信工程学院,江苏南京210007 [2]94676部队,上海202178 [3]93117部队,江苏南京211515

出　　处：《陆军工程大学学报》2023年第6期25-30,共6页Journal of Army Engineering University of PLA

基　　金：国家自然科学基金(62071488,U22B2002)。

摘　　要：通信干扰是频谱安全领域的重要研究方向之一。在复杂电磁环境下的无线通信对抗过程中,通信用户数量动态变化、用频规律未知,使得传统通信干扰方法难以达到理想的效果,干扰效率不高。为此,针对通信用户数量变化情况下的频率域干扰问题,提出了基于多模Q学习的多干扰机干扰信道策略优化算法。将动态变化的用户数量视为不同的模态,干扰机针对不同的模态,分别建立对应的Q值表,采用Q学习方法并以概率更新的方式,在不同模态下依据玻尔兹曼概率选择干扰信道策略,同时通过所获奖励更新维护各自Q值表。仿真结果表明,所提算法能够有效应对用户数量变化的情况,且相较于随机干扰算法,提高了干扰成功率。Communication jamming is one of the important research directions in the field of spectrum security.In the wireless communication confrontation in the complex electromagnetic environment,the number of communication users changes dynamically and the frequency rule is completely unknown,which makes it difficult for the traditional communication jamming methods to achieve the ideal effect and efficiency.Therefore,the multi-jammer jamming channel strategy optimization algorithm based on multimode Q learning is proposed for the dynamically changing number of communication users.Regarding the dynamically changing number of users as different modes,the jammers establish the corresponding Q value tables for different modes.The Q learning method will select the corresponding jamming strategy according to Boltzmann probability under different modes,and maintain the respective Q value tables by updating the reward.The simulation results show that the proposed algorithm can effectively deal with the situation of the dynamically changing number of users and improve the jamming success rate,compared with the random interference algorithm.

关 键 词：无线通信对抗 多模Q学习 多干扰机 干扰信道策略 

分 类 号：TN975[电子电信—信号与信息处理]