Research On Task-oriented UAV Formation Networking Technology

In recent years,with the advancement of artificial intelligence and avionics technology,Unmanned Aerial Vehicle(UAV)has been widely used in military combat due to their advantage of "zero life risk",good invisibility,and high combat effectiveness,etc.Nowdays,since UAVs can perform multiple missions,the cluster tactics of a bee-like distribution network composed of multiple UAVs are more lethal comparing with single combat,thus the cluster operation has been the primary air combat mode in modern battlefield.And the premise of UAVs' cluster operations is to achieve efficient and reliable communication within the group,therefore,the UAV networking technology,especially routing protocol,has become the research priority of UAV network in recent years.This article focuses on analyzing different network scenarios of UAV combat and designing reliable and efficient routing protocols.The main work is as follows:Aiming at the communication protocols in multi-UAV cooperative combat,routing protocols and mobility models in the Flying Ad Hoc Network(FANET)are classified,and OLSR is selected for networking the UAV initially.For analyzing the UAV combat scenario,this paper studies RPGM and Pursue mobility models,which are usually used to simulate reconnaissance/surveillance and track/strike trajectory,according to the characteristics of the multi-UAV cooperative combat.In addition,the function and architecture of the simulation software NS-3 are also studied,and the functional design of the group mobility model is implemented in NS-3,which provides a software basis for the performance analysis of routing protocol.To solve the problem that the OLSR protocol cannot meet the business requirements in multiple combat scenarios,this article designs a UAV combat network architecture based on the U.S.combat model,proposes an Link-Sensing and Adaptive OLSR protocol,and implements the new routing module in NS-3.This protocol combines two mechanisms with OLSR: link awareness and parameter adaptation.Among them,the link awareness exploits the mobility and location information in the UAV network to measure the past,current,and future stability of the link for enhancing the ETX mechanism,which optimizes the MPR and routing selection algorithm.The parameter adaptive mechanism comprehensively considers the current network connectivity,network scale,and flight speed,and modifies the packet sending intervals appropriately.In view of a large amount of business demand for data distribution in UAV combat,using unicast routing only will result in low network bandwidth utilization,this article extends the multicast routing function of OLSR,and introduces a tree-based Multicast Optimized Link State Routing protocol,and completes modular design in NS-3.This protocol makes full use of the topology and neighbor information maintained by the unicast routing OLSR,optimizes the MPR selection algorithm with multicast information,and establishes a multicast tree between source node and multicast member by exchanging related control packets.The shortest path from the source to the destination is calculated based on the tree structure to forward the data.For the two routing protocols proposed,this paper analyzes the performance of routing protocols in different network scenarios through multiple simulations.Simulation results show that,compared with traditional OLSR,the link-sensing and adaptive OLSR protocol can adapt to network scenarios under different tactical types.Furthermore,the performance of the Multicast OLSR is simulated,the result indicates that the multicast routing protocol can meet the requirement of UAV data distribution,which improves the bandwidth utilization rate and the combat effectiveness of UAV.