多机制优化的可靠UANET路由协议  

Multi-Mechanism Optimized Reliable UANET Routing Protocol

作　　者：贾文杰 高峰 李大鹏[1] JIA Wenjie;GAO Feng;LI Dapeng(School of Communication and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing Jiangsu 210003,China)

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003

出　　处：《通信技术》2024年第10期1049-1058,共10页Communications Technology

摘　　要：在无人机自组网中,节点飞行速度较快会导致网络节点拓扑结构变化剧烈,传统自适应算法反应不够迅速,难以在不增加报文开销的情况下提升性能。提出了一种多机制优化的可靠无人机自组织网络(Unmanned Aerial Vehicle Network,UANET)路由协议,该协议能够根据节点之间的相对位置信息,计算出链路维持时间,从而调整感知Hello报文的发送间隔,及时感知链路变化。此外,在减少拓扑控制(Topology Control,TC)报文冗余的同时,利用Q学习算法以一跳对称邻居变化率为基础设计状态列表,通过历史变动信息设计奖励函数来调整TC报文的发送间隔,以及时更新网络拓扑;结合链路维持时间和邻居变化率2个指标,有效避免了选择链路不稳定的节点作为多点中继节点。实验结果表明,相较于其他协议,该协议在平均端到端时延、丢包率和吞吐量的性能上均有改善,能够更好地适应无人机自组织网络。In UAV(Unmanned Aerial Vehicle)self-organizing networks,the rapid flight speed of nodes leads to drastic changes in the topology structure of network nodes.Conventional adaptive algorithms are not responsive enough to improve performance without increasing message overhead.This paper proposes a multi-mechanism optimized adaptive and reliable UANET routing protocol,which is able to calculate the link maintenance time based on the relative position information between nodes,so as to adjust the sending interval of the Hello message and sense link changes in a timely manner.In addition,while reducing the redundancy of TC(Topology Control)messages,Q-learning algorithm is used to design a state list based on the one-hop symmetric neighbor change rate,and a reward function is designed using historical change information to adjust the sending interval of TC messages,which facilitates timely to update the network topology.By combining the indicators of link maintenance time and neighbor change rate,it effectively avoids selecting nodes with unstable links as MPR(Multi-Point Relay)nodes.The experimental results indicate that compared to other protocols,this protocol has improved performance in terms of average end-to-end latency,packet loss rate,and throughput,and can be better applied to UANET.

关 键 词：无人机自组网 Q学习 优化链路状态路由 多点中继 

分 类 号：TN929.5[电子电信—通信与信息系统]