Research On Task Offloading Strategy In Edge Computing Of VANETs

In recent years,Vehicular edge computing(VEC)has brought the cloud computing model to the edge of the network,mitigating the impact of insufficient computing power and delay caused by a large number of data requests in the network.With the development of Intelligent Transportation Systems(ITS),there are more and more intelligent applications.These intelligent applications usually have the characteristics of computationally intensive,delay sensitive,and high bandwidth requirements,which have a great pressure on the computing ability of vehicle terminals,and the wireless access network also has a great burden on it.The emergence of Mobile Edge Computing(MEC)technology has effectively reduced the burden and delay of the network,but how to make task offload decisions and resource allocation has become a key issue in VEC.This thesis aims at improving the task offload efficiency of vehicles in the Internet of Vehicles and allocating edge computing resources reasonably.While satisfying the success rate of tasks with high priority as much as possible,let as many vehicles as possible obtain the resources of the edge server.In order that edge computing resources can be reasonably allocated,the main contents of this thesis are as follows:(1)Aiming at the edge server deployed on RSU,a centralized task offloading and resource allocation method based on greedy algorithm is proposed to enact a reasonable task offload strategy.First,the controller is used to collect the task information sent by the vehicle.Then,in an environment where multiple users compete for limited edge resources,in order to meet the needs of intelligent applications of vehicle,two ways of task completion are adopted.Then,the task offloading decision of the vehicle and the strategy of computing resource allocation are studied.Finally,the controller is used to uniformly allocate and manage edge computing resources for decision of task offloading to meet the task offloading needs of vehicles.The results show that the proposed algorithm can effectively increase the task offloading success rate of the vehicle,so that the edge computing resources can be reasonably allocated.(2)Aiming at the deployment of edge servers on edge network data centers,a joint offload method based on task urgency is proposed.In some areas with dense vehicles,the difference in the number of vehicles leads to differences in the load status of the edge server.From the perspective of rational allocation of vehicle offload task requests and computing resources,the tasks are divided by the processable area according to the task deadline,type and load of the MEC server.Then,RSU selects the edge server with the highest probability of completion according to the calculated probability to offload the task.When the vehicle travels to the designated area,they can directly receive the results of the task request.The results show that the proposed algorithm can significantly reduce the total number of tasks that miss the deadline date,effectively reduce the load pressure of a single edge server and optimize the load situation of the edge server in the entire area.