Research On Topology Connectivity Oriented Optimal Deployment Algorithms Of Unmanned Relay

With the rapid development of integrated circuit technology,large-scale UAV network deployment has become a possibility,especially in the field of war.Due to the concealment and all-weather characteristics of UAV,it plays a more and more important role in modern war.However,in the battlefield environment,UAVs may be destroyed for various reasons,resulting in the loss of connectivity of the overall network and hindering our military action.Therefore,in order to solve the problem that UAVs may be destroyed in war,thesis proposes an optimal deployment algorithm of unmanned relay based on clustering,which mainly optimizes the network based on the topological connectivity of the network,so as to improve the topological connectivity of the whole network.The main research contents of Thesis are as follows:Aiming at the possible destruction of existing networks,thesis proposes a network enhancement technology based on topological connectivity.In a UAV network,due to the uneven distribution of UAVs,the connectivity of some areas is poor.Through the analysis of network topology connectivity,thesis finds out the relatively important weak connectivity areas,and improves the overall network performance by introducing a small number of relay UAVs.It mainly solves the problem of determining the weakly connected area,the introduction algorithm of relay UAV and the overall network evaluation system.In the battlefield environment,no matter how designed,the UAV network may still be destroyed.For the above problems,in order to improve the destroyed network,by calculating the topological connectivity of each node of the network,a small number of relay UAVs are introduced to realize the connectivity of the whole network.The main problems to be solved are: how to connect as many UAVs in the network as possible,how to realize the communication of UAVs in the network and the corresponding evaluation system for the connection results.For the above problems,thesis mainly uses K-means++ algorithm to split the network,particle swarm optimization algorithm to optimize the problem,and combined with other relevant algorithms to solve the problems in different scenarios.The final simulation results show that the overall network performance can be improved by introducing a small number of UAVs.In addition,thesis proposes a general algorithm.In the actual situation,we can conduct more in-depth research on the problem according to the specific situation.