Research On Inter-satellite Handover Strategy Based On Dynamic Prediction

Compared with terrestrial cellular systems,satellite mobile communication systems have the advantages of seamless global coverage and are not easily destroyed by natural disasters.Among them,the Low Earth Orbit(LEO)satellite communication system has become the research focus of the satellite mobile communication system due to its advantages of low power consumption and low communication delay.LEO satellites move at high speed relative to the ground,so in order to ensure that user terminals can communicate continuously,service satellite switching is required.Efficient inter-satellite handover strategy can not only allocate satellite resources reasonably,but also reduce the handover failure rate and new call blocking rate of user terminals,and improve the communication performance of the entire system.At present,most studies on inter-satellite handover have neglected the influence of terminal motion on inter-satellite handover,resulting in some unnecessary handovers and increasing the terminal handover failure rate.In response to this problem,this thesis proposes a prediction-based multi-attribute no-preference switching strategy between satellites in the third chapter.First,predict the position information of the terminal in the future according to the mobile terminal's movement status,and use the predicted terminal position information combined with the satellite ephemeris to find the candidate satellites of each time slot terminal,and then according to the service of these candidate satellites to the terminal Duration,establish a terminal handover directed graph;Then,a multiobjective function is established by integrating the three inter-satellite switching properties of service duration,communication angle and the number of idle channels,and using the Niched Pareto Genetic Algorithm(NPGA)to screen the switching path of the terminal in the directed graph.The simulation results show that the proposed strategy can reduce the terminal handover failure rate and the new call blocking rate,and balance the load on the satellite.Because the strategy proposed in the third chapter is easy to cause different services to occupy the same channel resource,the switching failure rate and new call blocking rate of each service are still high.Therefore,in fourth chapter,a prediction based multi service switching strategy is proposed.This strategy first uses the analytic hierarchy process to set weights for each handover attribute and establishes an objective function according to different business preferences for each handover attribute of the service satellite.Then,according to their own objective functions,different services use genetic algorithms to screen out handover paths that meet their own characteristics in the terminal handover directed graph.In order to further ensure the success rate of handover of real-time services,a multi-service handover management method is proposed.The simulation results show that this strategy can further reduce the handover failure rate and new call blocking rate of various services.