Research On Dynamic Service Allocation Algorithm Of Vehicle Network Based On Mobile Edge Computing

In recent years,with the continuous growth of car ownership and the rise of computing intensive tasks,such as autonomous driving and vehicle multimedia services,the requirements for vehicle task processing performance in vehicle network are increasing.However,the traditional cloud computing architecture has been unable to cope with the challenges of the current complex vehicle network environment.In view of the disadvantages of the traditional cloud computing,it has become a more effective method to deploy edge nodes closer to the vehicle terminal,improve the complex task processing capacity in the vehicle network,and shorten the task processing delay.At present,the optimization of resource and service dynamic allocation based on edge computing in vehicle network has become one of the research hotspots in the field of vehicle network and intelligent transportation system.This paper focuses on the dynamic service allocation algorithm based on mobile edge computing in heterogeneous network environment.The main research contents are as follows:Firstly,aiming at the problem that there are many kinds of tasks in the vehicle Edge network,this paper proposes a task clustering algorithm based on task dependency.The algorithm clusters the complex task model of vehicle arrival and task generation in the vehicle Edge network.By constructing the task processing model of directed acyclic graph,the data matrix is set according to the standardized weighting number of the nearest neighbor shared between the target task itself and the selected task,and the task correlation learning is carried out to further build the correlation between the task sets.Cluster processing is based on the maximum value of dependency between services.This algorithm can divide and cluster the arriving tasks,avoid the repeated processing of high similar tasks,and simplify the complexity of tasks required by the edge server.The simulation results show that the clustering algorithm is superior to the traditional algorithm in clustering time and accuracy,can shorten clustering time and improve clustering accuracy,and can be applied to the vehicle network environment.Compared with the task processing without clustering,the delay is reduced by nearly 30%,and the energy consumption is reduced by about 20%.Secondly,in view of the limited computing power of edge server,a task adaptation algorithm based on mobile edge computing in the Internet of vehicles is proposed.The algorithm solves convex optimization problems by multiple polynomial iterations to increase the number of input tasks that can be processed in edge computing networks,and periodically adapts to the number of input tasks in edge computing networks to increase the number of tasks that can be processed in edge computing networks.The scheme in this paper is simple and convenient through linear programming.This algorithm can dynamically adjust the task load between the cloud server and the edge server,and make full use of the advantages of the cloud server and the edge server.The simulation results show that the algorithm can improve the efficiency of task processing,save energy consumption,improve the user experience,balance the processing capacity between the cloud server and the edge server,and play a good role in dealing with the server overload caused by a large number of arrival tasks in the vehicle network.