Research On Low Earth Orbit Satellite Network Routing Algorithms Based On Intelligent Optimization

With the rapid development of mobile communication technology and the continuous innovation of science and technology,building a global interconnected wireless network has become the future trend of communication.LEO satellite communication network has become a powerful complement to land communication network for its advantages of low launch cost,low delay,low loss and global coverage.However,the service distribution of LEO satellite network is extremely uneven,and the activity of users changes with time,which makes the service pressure of LEO satellite network extremely uneven under high service load.Because the computing power and energy of the satellite are limited,its service life will be greatly reduced when the satellite load is too large.Therefore,how to achieve efficient load balancing in LEO satellite network is an urgent problem to be solved.Intelligent optimization algorithm is more and more popular in solving various optimization problems because of its decentralization and adaptive ability,and has been widely used in network routing.In order to solve the above problems,this thesis studies the routing technology of LEO satellite network based on intelligent optimization,and the main contents are as follows:Firstly,an inter-satellite routing algorithm based on genetic algorithm was proposed for low earth orbit satellite networks.Based on the basic process of genetic algorithm,the fitness function with delay,the crossover method with LEO satellite networks characteristics and the mutation probability of satellite congestion are redesigned.Compared with the shortest path routing algorithm on NS2 platform,the simulation results showed that the proposed algorithm had lower packet loss rate,better average delay,lower transmission overhead ratio and higher throughput.Secondly,the artificial bee colony algorithm and genetic algorithm were combined to the LEO satellite networks scene,and a load balancing routing algorithm based on genetic optimization of artificial bee colony was proposed.The crossover operator of genetic algorithm was applied to the bee-collecting stage of artificial bee colony algorithm to avoid the artificial bee colony algorithm falling into local optimum.In addition,this thesis divided the current routing state into three congestion states,which are first congestion,second congestion and non-congestion.In order to balance the load of LEO satellite network,different updating principles were selected according to different states.In the experiment,this thesis used NS2 simulation platform to simulate.Compared with other routing algorithm in LEO satellite networks,the proposed algorithm demonstrated good performance in average delay,transmission overhead ratio,data delivery rate and throughput.