The Performance Optimization For The Edge Cloud User Systems

With the development of 5G technology and the great progress of cellular communication system,there are more and more complex computing applications accompanied,and these bring great challenges to mobile terminals which has limited computing resources.In this situation,it is an important way to unload the complex computing applications of mobile terminals to the powerful cloud server.In the past Mobile Cloud Computing model,Cloud server mainly used to expand the Computing and storage capacity of Mobile terminals and users can offload their applications to a cloud server to reduce application completion time.However,the communication latency between them is high because the cloud server is far from users.Deploying small-scale cloud servers on the edge of the network has become an effective solution to reduce the delay in the traditional mobile cloud computing model,and this kind of small-scale cloud server can be named as edge servers.The process of offloading complex computing tasks of mobile terminals to edge cloud servers for processing is Mobile Edge Computing(MEC).For mobile edge computing system,the mobile edge computing system performance can be optimized by increasing the computing capacity of mobile devices and the capacity of transmission channels.According to this,the main research of this paper is as follows:(1)To solve the problem of energy saving calculation for complex application tasks running on multi-core mobile devices,,this paper designs a core scheduling algorithm for CUP core and edge cloud resources on mobile devices—KSA(Kernel Scheduling Algorithm)to process user offloading decision-making and task scheduling issue.The basic idea is to make the unloading decision according to the energy consumption comparison of different tasks in the local and cloud.Then the tasks of different applications are prioritized to ensure that the application tasks meet the time constraints.To reduce the energy and time consumption during task execution,the algorithm can schedule the tasks on the CPU core reasonably within the delay constraints.Finally,the CPU consumption is reduced by reducing the completion time.Simulation and experimental results show that the proposed scheduling algorithm can achieve the time constraints of application tasks and significantly reduce the energy consumption of the MEC system.(2)Aiming at the energy consumption optimization problem of the multi-user edge cloud computing system,this paper combines the Massive multiple-input multiple-output(MIMO)technology,which has the characteristics of high channel capacity and low latency,and then constructs a multi-user MEC system based on MIMO.A Nested gradient iteration algorithm,NGIA(Nested Gradient Iterative Algorithm),was designed based on MIMO channels.This algorithm was divided into external gradient iteration and internal gradient iteration.The optimization problem can be transformed into a delay assignment problem to be solved.Due to the complexity of the original problem constructed,the non-convex objective function is decomposed,and then the original problem is transformed into two subproblems of convex functions by constructing the Lagrange function,and the optimal solution can be obtained by the NGIA algorithm.Finally,the results of the Matlab simulation show that the NGIA algorithm has advantages in reducing the energy consumption of the MEC system.When considering the overall energy consumption optimization of the multi-user system,the energy loss in the downlink was not considered in previous studies.This thesis provides some analysis on this issue,and this indicates that this thesis analyzes comprehensively and has certain academic value.