The Research Of Resource Collaborative Optimization Between Terminal And Base-Station In Mobile Edge Computing

Mobile Internet and Internet of things has been developing rapidly,and constantly have sprung up a lot of new business,such as high complexity business,high energy consumption business,and delay sensitive business,that bring great challenge to terminal computing power and endurance,at the same time of increasing network load.In order to solve the above problems,the concept of edge mobile computing arises at the historic moment.Through the edge near the mobile users of wireless access network,providing IT services and computing power,calculation is migrated to the edge nodes,which effectively reduce the center network load and reduce the demand for data transmission bandwidth.Mobile edge computing has the characteristics of high bandwidth and low delay.It has greatly promoted the development of 5G technology and has received extensive attention in recent years.Computing migration is the core content and primary service mode of mobile edge computation,which is conducive to the breakthrough of the resource limitation of mobile terminals,providing powerful computing power and improving user experience.It is the time and energy expenditure that are the main considerations for the terminal users to comply the computation offloading or not.Therefore,user's expenses can be minimized by reasonable computation offloading strategy.According to the thought of service localization,storage capabilities and computing capabilities should be provided in mobile edge computing service deployment.The service deployment and computing migration of multi-terminal single base station scene has been studied in the thesis firstly,and the resource scheduling strategy applicable to single base station scenario has been proposed.Then,based on the multi-terminal multi-base station scene service delivery and computing migration,the resource coordination strategy for the multi-base station scenario was also proposed in the thesis.The following research work and innovative points was involved in this thesis:The service delivery process has been studied in this thesis on the scene of multi-terminal with single base station,moreover,considering computing task classification,task information,and intelligent base station information,and designed base station caching strategy and terminal computing migration strategy.Subclass tasks has been defined by using computation and data volumes and the poisson process has been used to simulate the terminal application generation and terminal tasks to the base station.In order to maximize the benefit of the system,including time gain and energy consumption,the migration problem has been transformed into optimization problem in the thesis.Also,simulated annealing method has been used to solve the 01 knapsack problem based on the minimum transmission delay,and the time delay was calculated by the gradient projection method to solve the minimum convex optimization problem,using the steepest descent method to solve the minimum energy consumption of linear constrained optimization problems,Modeling maximization experience function,the method of simulated annealing was used to get the base station caching strategies and the gradient projection method was used to get the optimal terminal migration strategy.Finally,compared with the other three strategies via Matlab software simulation,the response time delay was reduced by 44 percent and energy consumption by 28 percent.Based on multiple terminal station of resource synergy,the master-slave framework of multi base station was designed in the thesis,and put forward with the innovative storage and computing power at the center of the base station and have the computing power service base station service,center base station is responsible for resource scheduling and calculation data caching policy,mission computing strategy through the service stations.Finally,compared with the other three strategies via Matlab software simulation,the response time delay was reduced by 54 percent and energy consumption by 36 percent.In the study of this thesis,the resource scheduling strategy between the terminal and the base station has been improved significantly in both the terminal energy consumption and the system response time delay.The research results have some reference value for research work and engineering implementation of mobile edge computing.