Task Offloading Of Internet Of Vehicles System Based On MEC

In the Internet of Vehicles,mobile edge computing(MEC)can offload the computing task of the target vehicle to the edge node for processing,thus reducing the computing energy consumption of the target vehicle and shortening the processing delay of the computing task.However,due to high vehicle mobility,the vehicle may drive out of the communication coverage of RSUs before receiving the computing results of the task.Therefore,in this paper,the task offloading-migration of MEC-assisted vehicle network is studied.The main contributions are as follows:(1)An intelligent city road system model is proposed to minimize the computation overhead of task offloading.In order to reduce computation overhead,the target vehicle needs to offload the computing task to a nearby vehicle or RSU for processing.Considering that the amount of data of the computing results is small and can be ignored,the offloading decision problem is modeled as a non cooperative game with vehicle/system computation overhead as the utility function,and an effective solution algorithm is given.The simulation results show that the algorithms reduce the computation overhead and the number of iterations.(2)Considering that the amount of data of the computing results is big and cannot be ignored,high-speed vehicles may drive out of the RSU communication coverage area before receiving the computing results,so reliable data transmission and effective computing offload cannot be guaranteed.In this paper,vehicle-to-vehicle(V2V)links and infrastructure-to-infrastructure(12I)links are studied to migrate the computing results,and the V2V migration number and 121 migration number are deduced.(3)A highway system model is proposed to minimize the computation overhead of task offloading-migration.In this system model,target vehicles need offload their tasks to nearby vehicles via V2V links,or offload their tasks to nearby RSUs via vehicle-to-infrastructure(V2I)links.In order to avoid that the target vehicle cannot receive the computing results from the RSU,the offloading-migration decision problem of the vehicle is studied in this paper,and the offloading-migration decision problem is modeled as a complete information sequential game combining with the game theory.The finite improvement property is used to prove that the game can always reach Nash equilibrium and has convergence.In the end,the offloading-migration algorithms are designed to reduce the computation overhead and improve the success rate of task processing.