航空高动态网络链路感知OLSR路由算法

针对航空高动态无人机（UAV）网络环境中节点移动速度快、网络拓扑变化快，导致网络链路稳定性差、数据到达率低和信息拥塞度高等问题，提出了一种航空高动态网络链路感知OLSR（OLSR-LA）路由算法，该算法利用接收的2个连续Hello消息的多普勒频移、能量等信号特征，计算出航空高动态无人机网络中2个相邻节点的相对速度和移动趋势，从而得出这2个节点之间链路的保持时间。根据节点MAC层接口队列长度衡量网络局部的负载程度，并利用ARIMA-WNN组合预测模型预测下一时刻节点负载的预测值，并通过Hello消息传递给邻居节点。根据链路感知情况，采用基于局部路由负载均衡（RRLB）算法避免拥塞的发生。仿真结果表明，与传统OLSR算法相比，本文提出的算法有效提高了分组交付率，降低了端到端的传输延时，增加了网络吞吐量，从而提高了整个无人机网络传输的有效性和实时性。 

Link awareness-based OLSR routing algorithm for airbonre highly dynamic networks

Due to the high mobility of the unmanned aerial vehicle (UAV) node, quick changes of the network topology structure, the airborne highly dynamic UAV network suffers some problems such as poor stability of the network link, low data delivery ratio and high data congestion information. In order to overcome these problems, a link awareness-based OLSR (OLSR-LA) routing algorithm for airborne highly dynamic networks is proposed. The charactreistics of two received consecutive Hello messages, such as the Doppler shift and the power strength of received Hello messages, can be used to obtain the relative speed and direction of motion between the two adjacent nodes in the airborne highly dynamic UAV network. Then the link connection lifetime is estimated by the relative speed. The OLSR-LA routing algorithm uses the queue length in the buffer of the MAC layer to indicate the local load level. Then the predicted value of the node's load level in the next time can be predicted by the ARIMA-WNN combination forecasting model and passed to neighbor by Hello message. Finally, according to the conditions of link awareness, regional routing load-balancing (RRLB) algorithm is employed to avoid network congestion. The simulation results show that compared to the traditional OLSR algorithm, the proposed OLSR-LA routing algorithm can effectively improve the packet delivery rate, increase the traffic of network, reduce the end-to-end transmission latency and enhance the real-time and effectiveness of the data transmission in the whole UAV networks.