Research On Deployment Of Edge Server In Internet Of Vehicles Environment

With the rapid growth of vehicles and the progressive increase of requirements for traffic safety and efficiency,traffic safety and roads have been given more intelligent capabilities.However,powerful computing power and stable communication capability are required to realize an intelligent system that connects vehicles to everything.Although the computing power and running memory of the vehicles are becoming more and more powerful,it is still underpowered for high-intensity computing and communication tasks limited by the structure,capacity and weight of the vehicle.The previous solution is generally to upload the vehicular computing tasks to the centralized cloud computing center for calculation.Despite the sufficient computing capacity of the centralized cloud computing center,the geographic location of the centralized cloud computing center is too far from the users,and therefore the edge computing model of migrating the computing capacity from the remote cloud computing center to the edge of the network is proposed.Edge computing not only reduces the pressure of the core network,but also makes the computing resources closer to the users.In the Internet of Vehicles,in order to ensure the communication of vehicles during driving,it is not only necessary to deploy a large number of roadside units on the roadside to realize local communication of Internet of Vehicles devices,but also to deploy a large number of base stations in the network to achieve wide area coverage.In addition,to provide computing services,a large number of edge computing servers need to be deployed in the Internet of Vehicles to realize the integration of computing and communication.Therefore,for the integration scenario of edge computing and the Internet of Vehicles,the deployment of edge servers should be considered firstly.From the perspective of deployment cost and operation and maintenance,edge servers are usually co-located with roadside units and macro base stations.Due to the different service capabilities and business scopes of the edge servers at the roadside and the area,it is necessary to deploy the roadside edge server and the area edge server at the roadside and the area respectively.To this end,this paper mainly studies the deployment of roadside edge servers and regional edge servers in the Internet of Vehicles.The key problems in the deployment of edge servers in the Internet of Vehicles are analyzed,and the goal is to minimize the response time of computing tasks,and the problem is expressed as an optimization problem under multiple constraints.A ranking-based deployment algorithm and a deployment algorithm that minimizes the response time are proposed to solve the optimal deployment loactions of the roadside edge server and the regional edge server in the Internet of Vehicles,so that the average response time of the edge computing system is minimized,and it is compared to several classical deployments algorithms.The experimental results show that compared with other algorithms,the algorithms proposed in this paper can find the best edge server deployment locations in various situations,greatly reducing the average response time of the system.