Research On Offloading Strategy Of Vehicle Cooperative Computing Task Based On Mobile Edge Computing

With the rapid development of Internet of Things technology and wireless communication technology,numerous applications related to Internet of vehicles have emerged in an endless stream,such as unmanned driving,intelligent assisted driving,etc.All the above applications have the characteristics of large amount of computation and short computation delay.However,the limited computing resources of individual vehicles may not be able to meet the above emerging application requirements,making it difficult to guarantee the required quality of service for Io V users.The introduction of Mobile Edge Computing technology into the Internet of Vehicles and the construction of Io V Edge by providing computing resources at the edges of wireless access networks and close to mobile users to respond to various applications has a great potential to enhance vehicle computing performance and meet the heavy computing needs of vehicles.However,the technology of edge Internet of Vehicles at the current stage is not perfect enough,and the coverage area of edge server cannot cover all users.Moreover,if all vehicles offload their tasks to the same edge server,the offloading scheme will limit the performance improvement due to overload.In addition,due to the long deployment of cloud server,too much delay in computing and offloading will seriously affect the user experience and fail to guarantee the quality of service for users.In order to solve this problem,this paper intends to optimize the utilization of more available low-latency computing resources and improve the resource scheduling strategy of the Internet of Vehicles.Specifically,this paper proposes a cooperative offloading strategy for vehicle computing tasks based on moving edge computing to tap the potential available vehicle computing resources near the task vehicles.The main work is as follows:(1)According to the offloading mode of computing tasks in the edge network of vehicles: local offloading,partial offloading and complete offloading.Combine computing tasks to four destinations for offloading: local,nearby collaborating vehicles,edge servers,and cloud servers.The existing calculation and offloading methods are summarized,the framework of calculation and offloading is constructed,and the system flow chart is designed.(2)In the Edge Car Network,the task vehicle can offload computing tasks to nearby collaborating vehicles.However,the number of collaborative vehicles directly determines the efficiency of the task offloading strategy for edge Io V.In this paper,a kind of efficient task offloading strategy for edge vehicle network is designed,and the optimal number of vehicles for cooperative computing offloading is studied.The objective function is designed and the offloading strategy for cooperative computing task with the optimal number of cooperative vehicles is worked out.Finally,the effect is verified by experimental simulation.(3)In this paper,a contract-based incentive mechanism for edge Io V is designed to encourage the collaborating vehicle to contribute its idle computing resources to the task vehicle,taking into account energy consumption.In order to solve the incentive problem of cooperative vehicle,the mathematical model of the benefit of task vehicle,the mathematical model of the benefit of cooperative offloading vehicle and the social welfare model of the system are built firstly.Under the condition of information asymmetry under the constraints of IR and IC,the maximum benefit of the task vehicle is obtained based on the contract theory.At the same time,the efficiency of the task vehicle is maximized under the condition of symmetric information.Finally,the reliability and effectiveness of the excitation model are verified by experimental simulation.