基于离线学习的无人机网络抗干扰通信方案  

An Offline Learning-Based Anti-Interference Communication Scheme for UAV Networks

作　　者：唐韬 赵润晖[1,2,3] 冯学炜 石伟宏 文红 彭钰琳 TANG Tao;ZHAO Runhui;FENG Xuewei;SHI Weihong;WEN Hong;PENG Yulin(University of Electronic Science&Technology of China(UESTC),Chengdu Sichuan 611731,China;Aircraft Swarm Intelligent Sensing and Cooperative Control Key Laboratory of Sichuan Province,Chengdu Sichuan 611731,China;Sichuan Provincial Engineering Research Center of Communication Technology for Intelligent IoT,Chengdu Sichuan 611731,China)

机构地区：[1]电子科技大学航空航天学院,四川成都611731 [2]电子科技大学飞行器集群智能感知与协同控制四川省重点实验室,四川成都611731 [3]电子科技大学四川省智慧物联通信技术工程研究中心,四川成都611731

出　　处：《通信技术》2024年第5期495-499,共5页Communications Technology

基　　金：国家自然科学基金(U23B2021,62201132)。

摘　　要：无人机面临先进干扰技术的挑战,易受恶意节点攻击、数据截取和篡改,传统的抗干扰决策存在一定局限,无法根据干扰信号的变化进行自适应调整,而基于深度强化学习(Deep Reinforcement Learning,DRL)的抗干扰通信模型需要长时间与环境交互,对抗干扰的环境要求较高。研究了基于Decision Transformer的离线抗干扰方法,其能快速稳定地获得实用的抗干扰决策模型。仿真试验验证了该算法在加性高斯白噪声信道和衰落信道环境下抗干扰决策的有效性,且该离线方案在训练迭代次数较少时便能达到预期奖励目标。UAVs(Unmanned Aerial Vehicles)face challenges from advanced interference technologies,making them susceptible to malicious node attacks,data interception and tampering.Conventional anti-interference decisions have limitations,as they cannot adaptively adjust to changes in interference signals.Moreover,DRL(Deep Reinforcement Learning)-based anti-interference communication models need prolonged interactions with the environment and require a high level of anti-interference environment.This paper investigates an offline anti-interference method based on Decision Transformer,which can quickly and stably obtain practical anti-interference decision models.Simulation experiments verified the effectiveness of the algorithm in making anti-interference decisions under AWGN(Additive White Gaussian Noise)and fading channel conditions.Furthermore,the offline scheme can achieve the expected reward goal with fewer training iterations.

关 键 词：无人机 抗干扰决策 深度强化学习 Decision Transformer 

分 类 号：V279[航空宇航科学与技术—飞行器设计] TN975[电子电信—信号与信息处理]