Research On High Speed Undirected Mobile Nodes Networking

With the rapid advancement in electronic,sensor,and communication technology,high-speed non-directional mobile devices represented by the Unmanned Aerial Vehicles(UAVs)are widely used in military and civilian areas.However,these applications are still limited to single UAV system and use the direct link between the ground base station and the devices for communication and control.This also give birth to some shortcomings that cannot give full play to the advantages of UAVs and only be used in uncomplicated application scenarios.The multi-UAVs systems with capabilities of communication,coordination and cooperation have the advantages of low cost,strong survivability,high task execution efficiency,and better concealment in the military.They can perform tasks that cannot be performed by single-UAV system and provide more diversified services.However,the primary problem in the realization of multi-UAVs is the networking of highspeed non-directional mobile nodes.This thesis deeply studies the problem of high-speed non-directional mobile nodes networking,and proposes a solution for the problem.The main contributions of this thesis are as follows:(1)Based on the analysis and summary of the network characteristics of high-speed and non-directional mobile nodes,and the problems in existing network architecture,the hierarchical structure of clustering is proposed by combining the network characteristics and the advantages of cluster architecture.This architecture fully considers the characteristics of high-speed and non-directional mobile nodes,and advantages of clustering architecture,and meets the networking requirements.(2)Aiming at the problem which existing clustering algorithm ignores is the distribution characteristics of nodes and also the mobility is not considered sufficiently.The thesis proposed a node distribution density based clustering algorithm as a solution to these problems.The algorithm selects the node with the highest V-graph density as the Cluster Head(CH)according to the topology distribution characteristics of the nodes,and then the neighbor nodes choose the most appropriate cluster to join according to the link duration and the relative mobility with CH.The simulation results show that the proposed algorithm outperforms in high-speed and non-direction network.(3)Moreover,a low energy adaptive clustering algorithm has been proposed in order to solve the problem of energy limitation in FANET.The algorithm comprehensively considers the aspects of relative mobility,node degree,remaining energy and link duration with neighbors,and calculates the stability factor of nodes by weighting these parameters.Subsequently,the node with the highest stability factor is selected as CH,and the other nodes select the appropriate cluster to join based on link duration and relative mobility of CH.Furthermore,in order to balance the load of CHs,the concepts of deputy cluster head nodes and gateway nodes are introduced,which are served by the member nodes satisfying specific conditions in a polling manner.The experimental results demonstrate the performance of our proposed algorithm in the high-dynamic network scenarios and effectively extend the life of the network.In brief,the research results in this thesis provide significant guideline for the implementation and application of high-speed non-directional mobile nodes network.