Research And Simulated Implementation Of Satellite Link Selection Based Network Capacity Analysis Method For Multi-Layer Satellite Networks

With the development of onboard processing and satellite link technology, satellite networks have become a new internet covering the global. Its status is improved in the field of communication. Satellite networks have covered the characteristics of wide coverage, rapid layout, flexible networking and strong robustness, and don’t subject to geographical constraints. As the effective supplement of traditional ground networks, satellite networks are playing an important role in both military and civil fields. For adapting to the growing of user groups and the increasing demand for multimedia services, the way to build efficient and reliable satellite networks is of great economic value.This thesis regards the network capacity concept as a performance index of designing and planning satellite networks. It provides direct reference to improve the efficiency of satellite resources and enhance the ability of supporting services. A satellite link selection based network capacity analysis method for multi-layer satellite networks is proposed by fully analyzing the characteristics of multi-layer satellite network. First of all, to depict the impact of channel bandwidth, transmission power and noise on the performance of satellite links, this thesis sets up the network model, node model, channel model, transmission model and interference model. Secondly, the services of satellite networks including the geographical factors, time factors, equipment factors, service types and QoS (Quality of Service) demands are described to dynamic simulate the traffic flow of satellite network. Thirdly, this thesis proposes satellite routing strategy, considering the QoS satisfaction, service pricing satisfaction, lifetime fitness and load fitness for multiple attribute decision making to simulate satellites dynamic switching on links, which builds satellite network topologies in operating conditions. At last, it employs the time slot division way to overcome the dynamic characters of satellite networks and proposes an improved distributed genetic algorithm to search the optimal solution of satellite resource allocations, which decides the upper or lower bounds for the satellite network capacity.With the help of STK (Satellite Tool Kit) software, the trajectories of satellites and satellite visibility are analyzed, and the proposed network capacity analysis method is implemented in the VC++ simulation software. This thesis compares this method with the two reference methods and analyzes the influences of satellite routing strategy, burst interference and constellation structure on network capacity. The simulation results show that the proposed network capacity analysis method for multi-layer satellite network is feasible and effective, and has better convergence performance and search accuracy when compared with the reference algorithms. And the analysis results of factors, which influence network capacity, are useful for designing and planning satellite networks.