| Satellite communication is a microwave relay communication technology that communicates with earth stations through artificial earth satellites.As an effective supplement to traditional terrestrial communication,it has the advantages of wide coverage,wide frequency band,and large communication capacity,and has been more and more widely used.However,with the continuous increase in the types of satellite communication services,such as high-definition video live broadcast,high-speed Internet access,etc.,the fixed resource allocation mode of traditional multi-beam satellites will lead to limited satellite capabilities and waste of resources.Therefore,the researchers came up with the beam hopping technology.This technology uses time slicing to provide the corresponding number of time slots for each beam according to traffic requirements,which improves the throughput and resource utilization of the satellite communication system.However,the size and shape of the beam position of the current beam hopping technology are relatively fixed,and cannot be changed with changes in traffic demand,and the system flexibility is not high enough;when multiple beams work at the same time,the problem of co-channel interference cannot be ignored;and the beam hopping service the more beams there are,the greater the queuing delay of the system,so the development of beam hopping technology still faces many challenges.Aiming at the above problems,this thesis firstly proposes a beam position optimization algorithm based on traffic distribution.In the case of uneven service distribution,the use of phased array antennas can flexibly change the characteristics of the pointing,width and shape of the hopping beam,and adjust the beam size according to the characteristics of traffic distribution.For the situation where users are scattered and the service demand is low,large beam positions are used to solve the access problem of users in a large range,so that the service demands in each beam are more balanced,and the situation that a few users have no resources available is improved.On this basis,from the perspective of matching the demand with the actual transmission capacity,the objective function of minimizing the n-order difference is established,and the convex optimization algorithm is used to solve the allocated time slot resources in each beam.The simulation results show that,compared with the traditional beam position division scheme,the algorithm can improve the overall satisfaction of the system,improve the success rate of user access and reduce the system delay.Secondly,considering that when more than one beam is lit at the same time,there will be cochannel interference between the beams,the beam position optimization strategy is improved,and a resource allocation algorithm that jointly optimizes time slot allocation and beam hopping pattern design is proposed.Firstly,the beam size is divided into several levels.When the traffic demand decreases,consider using large beam positions to solve the user access problem.When the service demand increases,consider configuring smaller beams to increase the access rate.Secondly,the genetic algorithm is used to design the beam hopping pattern to avoid co-channel interference.The simulation results show that the method can improve the system throughput and the overall satisfaction of the system.Finally,from the perspective of user location distribution,because the delay is related to the number of beam positions,the more the number of beam positions,the greater the delay.This thesis transforms the beam position optimization problem into a p-center problem,trying to use the least beam position The number of beams covers all user terminals,thereby reducing system delay.Firstly,a stepless and discontinuous beam position optimization algorithm based on user position distribution is proposed,and then the hopping beam pattern is directly obtained in one step by combining genetic algorithm,and finally the performance of beam position optimization is analyzed.The simulation results show that the beam position optimization shortens the queuing delay and improves the system throughput and resource utilization. |
