Research On Directional Antenna Networking Technology For Unmanned Swarm

Unmanned swarm technology can be widely applied to fields such as search and rescue,reconnaissance,and surveillance,realizing task allocation and information sharing,and improving the accuracy and efficiency of task execution.With the continuous development of unmanned systems,the research and development of unmanned swarm technology is of great significance,providing strong support for the realization of intelligent and autonomous unmanned systems.In the communication networking technology for unmanned swarms,neighbor discovery is the first step.The neighbor discovery process is a necessary prerequisite for establishing connections between nodes and maintaining network topology.To achieve neighbor discovery between any two nodes in the communication networking scenario of an unmanned swarm,it is necessary to use a sector with a certain beamwidth to cover the three-dimensional omnidirectional scanning.The fewer sectors,the lower the complexity of the neighbor discovery algorithm.In order to achieve the three-dimensional omnidirectional coverage of the beam with the smallest number of sectors when the beamwidth is known,this thesis proposes a three-dimensional directional beam coverage algorithm based on Goldberg polyhedra.Simulation results show that,compared with the equal latitude and longitude node distribution algorithm in existing research,the algorithm proposed in this thesis has superior performance,i.e.,under the condition of covering the three-dimensional omnidirectional with a fixed beamwidth,the number of sectors required for the algorithm based on Goldberg polyhedra is less than that for the equal latitude and longitude node distribution algorithm.Moreover,among the three types of Goldberg polyhedra,the second type of Goldberg polyhedra GP(m,m)performs best when used to design the beam coverage model.Subsequently,this thesis proposes a high-performance semi-blind directional neighbor discovery algorithm.The algorithm is designed based on the Chinese Remainder Theorem and employs a sector scanning mechanism,enabling neighbor discovery in semi-blind scenarios.It optimizes the sector scanning mechanism by configuring dynamic parameters P and Q for each node.The algorithm improves its performance by adaptively adjusting parameters P and Q in a distributed manner.According to the simulation results,the average time to discover(ATTD)performance and maximum time to discover(MTTD)performance of this algorithm is significantly better than the scan based algorithm-directed(SBA-D)with the provision of semi-blind information;as the number of nodes and the node density increase,the average time to discover performance of this algorithm is far superior to the complete random algorithm(CRA).