Research On Decision And Optimization Of Computation Offloading Ianvehicle Edge Computing Environments

Recently,the development of Cloud Computing technology has become more and more mature.,combining Cloud Computing with Internet of Vehicles,vehicles,roadside units,and the Central Cloud constitute the VANET Cloud Computing environment.In an VANET Cloud Computing environment,when tasks carried by vehicles are required to be processed,tasks can be processed on their own resources or offloaded to the Central Cloud for processing.The academia has called this method of migrating tasks on resource-constrained devices to resource-rich devices for processing is “computation offloading”.Computation offloading can not only prolong the battery life of the terminal device,but also reduce the task completion time.However,in the Internet of Vehicles,the high network delay and distant physical distance brought about by the Central Cloud of resource concentration may cause the cloud service to be unable to accurately locate the vehicle position and handle the task of low delay demand.With the rapid development of 5G communication technology,the edge computing is considered to be a promising method to improve the efficiency of the offloading.The deployment of edge servers on the roadside unit of the Internet of Vehicles has transformed the traditional two-tier computation offloading architecture vehicle-cloud into a three-tier computation offloading vehicle-edge-cloud.It is called the vehicle edge computing environments,this article is a study conducted in this environment.This dissertation focuses on the computation offloading mechanism in the vehicle edge computing environments of 5G heterogeneous networks.Firstly,it studies whether the introduction of the edge layer in the static vehicle edge computation offloading environment improves the task offloading performance.Then it studies how to make better offloading decisions for the tasks with different attributes.Finally,we focus on the study of the dynamic vehicle edge computation offloading environment,due to the movement of the communication link caused by vehicle movement,static environment under the offloading algorithm used in the dynamic environment can not be a good task to reduce the task’s completion time,proposed through the computation switching methods to reduce the dynamic vehicle edge computation offloading environment under the task completion time optimization program.This paper studies and analyzes how to improve the offloading efficiency in the vehicle edge computing environments,and simulation experiments are carried out based on the Matlab R2016 a platform.Under the constraints of the communication computation ratio,the performance parameters of the task execution time and completion time were analyzed and compared.The task’s computation offloading decision was instructed,and the experimental simulation of the proposed dynamic Vehicle Edge computation offloading optimization scheme was carried out.The results show that the three-tier computation offloading structure can complement the traditional two-tier computation offloading structure;the tasks with different attributes have different choices for the agent resource side;the optimization scheme of computation switching proposed in this paper can improve the offloading efficiency under the dynamic environment and shorten the task completion time.