Research On Resource Allocation Based On DVB-RCS2 System

On the issue of satellite bandwidth resource allocation,the existing resource allocation algorithms have problems such as waste of bandwidth resources,small actual throughput,and insufficient consideration of priority of service flows.In addition,most algorithms consider only a few parameters.Optimization or special conditions are required to achieve the good optimization,and many solutions need to be adaptively adjusted with the application of technologies such as SDN in the 5G era.In this thesis,the resource allocation process of GSE packet of satellite DVB-RCS2 is deeply studied,and the technical characteristics of bandwidth and power resource allocation are deeply studied.The advantages of resource allocation in new scenarios and new problems to be solved are analyzed.Firstly,based on the DVB-RCS2 protocol,this thesis studies and analyzes GSE,SDN and other technologies,designs the overall framework of the satellite resource allocation system,and points out the key technical problems to be solved.Secondly,an intelligent selection bandwidth allocation algorithm is proposed to automatically select a more suitable resource allocation optimization algorithm under different conditions.This is a new way to solve such optimization problems.It is different from the previous single algorithm in the mathematical optimization,using many excellent optimization algorithms to constantly improve themselves.Through mathematical modeling,a unified evaluation system for resource allocation algorithms is established,which can comprehensively consider the influencing factors such as delay and filling rate.It is better than the previous model to consider only one influencing factor.Using the naive Bayesian classification algorithm,the appropriate algorithm is selected among the various resource allocation optimization algorithms through the newly established evaluation criteria.This thesis proposes two resource allocation optimization algorithms,weight selection algorithm and potential genetic algorithm.Based on these two algorithms,the optimization of satellite resource allocation is realized,and the utilization of satellite resources is improved.Finally,the application of SDN and NOMA to satellites through simulation improves the shortage of satellite resources,and proposes solutions for the problem of user matching and uneven distribution of new resources in the process of resource allocation between NOMA and SDN satellites.For the first problem,this thesis proposes ACM technology with channel environment awareness by collecting past information,eliminates the impact of incomplete channel state information estimation,completes user matching,and establishes a fairness model to balance the total throughput in user matching.Basic requirements with links.For the second problem,under the principle of considering QoS and fairness,through the joint allocation of bandwidth resources,power resources and computing resources introduced by SDN technology,the problem of uneven distribution of satellite computing power that needs to be consumed during the instantiation of virtual network functions is solved.In this thesis,aiming at improving satellite throughput,the feasibility and effectiveness of the proposed joint resource allocation scheme based on NOMA and SDN satellites are verified by simulation.