基于Q-learning的自适应链路状态路由协议

Q-learning based adaptive link state routing protocol

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.