As an important means of mobile communication of geostationary orbit satellites,multi-beam satellite can realize long-distance data transmission and telecommunication.In order to meet the diverse needs of users,it is necessary to consider the communication rates among different users,spectrum utilization efficiency and channel caching,which brings great challenges to the dynamic resource allocation of satellite communication.A simulated annealing-elitist non-dominated sorting genetic algorithm(SA-NSGA Ⅱ)based on feasible region boundary projection is proposed for the optimization of user satisfaction and spectrum efficiency with the limitation of multi-beam satellite cache.The proposed algorithm uses projection method to project the iteration points outside the feasible region to the feasible boundary to realize the effective retention of historical optimization information and to improve the optimization efficiency of non-dominated sorting genetic algorithm(NSGA Ⅱ).The simulation experimental results show that the proposed algorithm can not only control the user cache within the specified threshold,but also effectively improve the user satisfaction and optimize the spectrum efficiency.
维普
万方数据