基于强化学习的跳频抗干扰系统设计  

Design of frequency hopping anti-jamming system based on reinforcement learning

作　　者：邹雯雯 Zou Wenwen(School of Mechanical and Electrical Engineering,Henan Agricultural University,Zhengzhou 450003,China)

机构地区：[1]河南农业大学机电工程学院,河南郑州450003

出　　处：《无线互联科技》2023年第11期8-10,共3页Wireless Internet Technology

摘　　要：跳频技术在无线通信中应用广泛,能够通过不同频点之间的转换,提高通信过程的安全性和抗干扰性,但此类系统对随机干扰信号和线性扫频干扰信号的识别、抵抗能力较差,产生了一定的使用缺陷。文章基于蒙特卡洛法和时序差分法,分别设计了在线和离线的强化学习算法,旨在探索提高系统抗干性的途径。经过MATLAB仿真检验,相关算法能够在较少的迭代后实现收敛,具备较强的干扰识别和规避能力。Frequency-hopping technology is widely used in wireless communication,which can improve the security and anti-interference of the communication process through the conversion between different frequency points.However,such systems have poor recognition and resistance ability to random interference signals and linear sweep interference signals,resulting in certain use defects.The research process is based on the Monte Carlo method and the time series difference method,and the online and offline reinforcement learning algorithms are designed respectively to explore ways to improve the system's anti-interference.Through the MATLAB simulation test,the relevant algorithms can achieve convergence after less iterations,and have strong interference identification and avoidance ability.

关 键 词：强化学习 跳频抗干扰系统 设计方法 

分 类 号：TN975[电子电信—信号与信息处理]