基于Q-learning的自适应链路状态路由协议  

Q-learning based adaptive link state routing protocol

作　　者：吴麒[1] 左琳立 丁建 邢智童 夏士超 WU Qi;ZUO Linli;DING Jian;XING Zhitong;XIA Shichao(Southwest Electronic Technology Research Institute,Chengdu 610036,P.R.China;School of Communications and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,P.R.China;School of Software Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,P.R.China)

机构地区：[1]西南电子技术研究所,成都610036 [2]重庆邮电大学通信与信息工程学院,重庆400065 [3]重庆邮电大学软件工程学院,重庆400065

出　　处：《重庆邮电大学学报（自然科学版）》2024年第5期945-953,共9页Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)

基　　金：重庆市自然科学基金资助项目(2022NSCQ-LZX0191);重庆市教委科学技术研究项目(KJQN202300638)。

摘　　要：针对大规模无人机自组网面临的任务需求多样性、电磁环境复杂性、节点高机动性等问题,充分考虑无人机节点高速移动的特点,基于无人机拓扑稳定度和链路通信容量指标设计了一种无人机多点中继(multi-point relay,MPR)选择方法;为了减少网络路由更新时间,增加无人机自组网路由策略的稳定性和可靠性,提出了一种基于Q-learning的自适应链路状态路由协议(Q-learning based adaptive link state routing,QALSR)。仿真结果表明,所提算法性能指标优于现有的主动路由协议。Large-scale unmanned aerial vehicle(UAV)ad-hoc networks face challenges such as diverse task requirements,complex electromagnetic environments,and high node mobility.To this end,this paper considers the characteristics of high-speed movement of UAV nodes and designs a UAV multi-point relay(MPR)selection method based on UAV topology stability and link communication capacity indicators.Additionally,to reduce network routing update time and increase the stability and reliability of UAV ad-hoc network routing strategies,a Q-learning based adaptive link state routing protocol(QALSR)is proposed.Simulation results demonstrate that the proposed algorithm outperforms existing proactive routing protocols in terms of performance metrics.

关 键 词：无人机自组网 路由协议 强化学习 自适应 

分 类 号：TN92[电子电信—通信与信息系统]