Research On Optimal Resource Allocation And MEC Server Selection In Internet Of Vehicles

With the vigorous development of Internet of Vehicles(IoV)and self-driving technology,a large number of computing-intensive and delay-sensitive applications such as videos and images have emerged in vehicle terminals.And there are massive data sensed by multiple sensors on the body of a car,which increases the local processing burden of unmanned vehicles greatly and makes it difficult to process tasks in time.At the same time,with the rapid development of the Internet of Things(IoT),roadside IoT devices provide real-time and reliable traffic information and emergency warning information for assisting unmanned vehicles.It is necessary to collect surrounding environment data in real time,and the huge amount of data is difficult to realize full localization processing.In this context,Mobile Edge Computing(MEC)emerged as an effective solution application.In response to the high latency of traditional cloud computing,mobile edge computing sinks the data processing center to the edge server close to the user for processing,and uses the powerful computing power of the edge server to solve the problem of limited terminal resources,so as to meet the demand of real-time response to the tasks.In the real traffic scenarios,factors such as the high-speed mobility of vehicles,a large number of IoT devices on the roadside,different types of computing tasks,and MECrestricted communication and computing resources have increased the difficulty of optimal resource allocation,making it difficult to achieve efficient real-time processing of computing tasks.In addition,for vehicles that are ready to perform task offloading,when there are multiple MEC servers to choose from,how to comprehensively consider the above factors and choose the best MEC server is a research difficulty that deserves attention.In response to the above challenges,this thesis studies the two issues,the joint optimization scheduling method of communication and computing resources in the single MEC scenario and the optimal selection of the MEC server in the multi-MEC scenario: 1)In the single MEC scenario,we study a tripartite joint task offloading scheme for vehicle terminals,IoT devices,and MEC.Considering different types of services comprehensively,a three-side dynamic joint task offloading and resource allocation scheme is proposed from a global perspective,which is solved by the optimization algorithm,particle swarm optimization(PSO),so as to minimize the total delay and overhead of vehicle terminals and roadside IoT devices with weights and minimize the overhead of MRSU;2)In multi-MEC scenarios,we introduce virtual channels,establish a matching model between MEC servers and vehicle terminals,and propose an optimal selection strategy for MEC servers based on bipartite graphs,whose goal is to select the optimal MEC server node for the vehicles to minimize the transmission cost of the vehicle users.