Research On Mobile Edge Computing In Satellite-terrestrial Network

With the continuous development of 6G technology,seamless communication coverage in all airspace,all time domain,all frequency domain and all terminal forms has become an important development direction of future communication.Satellite communication is characterized by high coverage,high flexibility and low limitation.Combined with ground communication network and edge computing technology,a satellite-terrestrial network with high reliability,high capacity and low delay can be built,which is of great significance for the future development of new communication services.However,due to the short cycle,fast movement and limited resources of low-orbit satellites,there are still some problems such as architecture design,connection management and resource allocation in integrating edge computing into the satellite-terrestrial network.In addition,because the period of low-orbit satellite is different from the rotation of the earth,it will receive complex signals from different regions,leading to security problems in the edge computing server of low-orbit satellite.To solve the above problems,This thesis carries out the following three aspects of research:Firstly,aiming at the problem of high delay caused by the distance of the satellite-earth link in the satellite-terrestrial network,the mobile edge computing architecture of the satellite-terrestrial network is proposed to sink the edge computing node to the low-orbit satellite,so as to reduce the user communication delay.Based on this architecture,an ephemeris beam independent switching method is proposed to solve the problems of frequent beam switching and edge computing data migration caused by the high mobility of low-orbit satellites,which optimizes the random access process of users,reduces the dropout rate and improves the reliability of edge computing services.Secondly,aiming at the problem of resource allocation and computation offloading in the edge computing of the satellite-terrestrial network,a service,delay,link and other element model of the edge computing system of the satellite-terrestrial network is established.Under this model,resource allocation algorithm based on Lagrange multiplier method and unloading strategy based on game theory and particle swarm optimization algorithm are proposed for the purpose of reducing the system delay.This strategy increases the delay by 33%compared with uncalculated unload,27%compared with random allocation for low number of users,and continuously increases with the increase of users in the case of high number of users.Finally,in view of the interference and security problems caused by a large number of aliasing signals from the ground brought by the low-orbit satellite edge computing nodes in the satellite-terrestrial network,a satellite-Earth signal recognition system is built.The system includes acquisition module,identification module,display module and storage module,which can collect satellite-Earth signals in real time and identify the channel coding mode of the signals.When the bit error rate is 2×10^(-3),the encoding type recognition rate of linear block code,convolutional code and TPC code is 95.7%,in which the encoding parameter recognition rate of linear block code,convolutional code and TPC code is 93.8%,92.7%and 90.4%.