Research On Handover And Resource Management Algorithms In Multi-beam LEO Mobile Satellite System

The Low Earth Orbit Mobile Satellite System(LEO-MSS)is favored by the industry because of its unique advantages such as rich radio frequency resources,long communication distance,and low transmission delay.Although LEO-MSS has many advantages,there are still many challenging problems that need to be solved to provide high-quality global coverage bandwidth communications.The high-speed low-orbit satellites make the terminal frequently switch between beams,which not only reduces the user’s communication experience,but also increases the burden on the system.At the same time,resources on satellites are limited,and reasonable and intelligent allocation of resources is the key to improving resource utilization and system performance.This dissertation focuses on the handover management and dynamic resource management of LEO-MSS.The main research contents are as follows:Aiming at the problem of poor system communication quality caused by frequent switching of LEO-MSS,using the periodicity and predictability of the low-orbit satellite trajectory,a multi-beam low-orbit satellite pre-switching scheme based on position information is proposed.First,by establishing a multi-beam coverage model based on the oblique projection mode,predict the user switching time.Secondly,according to the user’s location and satellite ephemeris information,the target beam is reserved for resources,and the network control center decides whether to allocate channels for the user according to the current resource allocation,and updates the user’s access information for the next pre-handover decision.Finally,in order to simulate and verify the LEO-MSS pre-handover process and communication performance,an OPNET-based low-orbit satellite communication system simulation platform was built.The simulation results show that when the call arrival rate is 1,the strategy switching failure rate in this paper is 87% lower than the non-priority strategy,and the channel utilization rate is 14% higher than that of the protected channel strategy.This scheme can effectively reduce the handover failure due to insufficient channel resources.Probability to improve the quality of system communication.Causing the problem of low-orbit satellite resources and low resource utilization caused by the dynamics of satellite networks,a multi-service low-orbit satellite resource allocation algorithm based on Deep Q Network(DQN)is proposed.First,to ensure the communication quality of handover users,a joint power and channel allocation model based on multi-services of low-orbit satellites is established.The model aims at maximizing system throughput,while being limited by low-orbit satellite coverage time and service queue stability.Second,in order to overcome the problem of unknown state probability in the optimization model,the multi-beam LEO-MSS resource allocation is mapped to the interactive learning of the agent in the environment to maximize long-term benefits.Finally,the solution is solved by state reconstruction and DQN algorithm to improve the decision-making performance and obtain the approximate optimal resource allocation strategy.The simulation results show that the proposed algorithm can improve the system throughput while meeting the needs of multi-service users and maintaining the stability of the service queue.When the service distribution is uniform and non-uniform,the proposed algorithm is 25% and 31% higher than the fixed allocation algorithm.