Research On Task-oriented Cross-Layer Routing Algorithm For UAV Ad-hoc Networks

Under the background of the development of communication networks,the organization mode of unmanned aerial vehicle(UAV)clusters is developing in the field of network and intelligence.The UAV self-organizing network is a mobile Ad hoc network composed of UAV clusters.Each UAV node uses multi-hop wireless routing to communicate,which expands the information coverage and communication distance of the UAV network and enhance the invincibility and robustness of the network.The routing protocol of the UAV Ad hoc network provides establishment,maintenance and routing decisions for the distribution of control messages and data information in terms of transmission path and routing optimization.The performance of the routing protocol largely determines the performance of the UAV communication network.The task-oriented routing algorithm design is the key to meet the mission transmission requirements and communication performance indicators of the UAV network.According to the characteristics and needs of the UAV Ad hoc network,the main research work is as follows:Analyze the advantages and existing problems of the directional networking mode,closely combine with the characteristics of directional transmission,and study the network topology and the quality service of the transmission service.The routing protocol of the existing UAV ad hoc network focuses on the principle,performance advantages and disadvantages of the classic on-demand routing protocol AODV.The topology of the UAV Ad hoc network changes and the links between nodes are not stable,leading to problems such as large transmission delays,route failures,etc.,while considering the quality of service(QoS)requirements for task transmission.On this basis,the task-oriented cross-layer routing algorithm DABP-AOCR is studied and designed.The algorithm uses the available bandwidth of the directional link estimated by the MAC layer as the upper limit of the link transmission throughput,and predicts the link layer between nodes.The available bandwidth information is sent to the network layer as a criterion for routing.At the same time,the information of the neighboring nodes is periodically updated to ensure the validity of the route,and the channel access and routing control for the new network access service are realized,and the existing transmission service in the network is guaranteed.Available bandwidth prediction and motion prediction improve the QoS performance of the global network.When predicting the actual bandwidth consumption of the directional multi-hop link service,the algorithm also fully considers the problem of directional transmission "in-stream competition" for multi-hop transmission,making the routing decision more stable and reliable,and improving service quality.With the data packet arrival rate,average end-to-end delay and routing overhead as performance indicators,various routing protocols are simulated and analyzed on the Aeronet network simulation platform.The simulation results show that compared with the typical mobile ad hoc network routing protocol.The proposed cross-layer routing algorithm based on dynamic link available bandwidth DABP-AOCR can improve the transmission performance of the whole network,meet the QoS requirements of task transmission,effectively improve the packet arrival rate of the drone self-organizing network,reduce the average end-to-end delay,and reduce the routing overhead.