Research On Vehicle Task Offloading Algorithm Based On Mobile Edge Computing

With the emergence of computing-intensive applications such as emerging autonomous driving,the demand for vehicle communication continues to grow,and Internet of vehicles(Io V)terminals face the dual challenges of end-to-end latency and energy consumption.Mobile edge computing(MEC),as one of the key technologies of 5G,reduces data processing delay and terminal energy consumption by sinking computing and storage capabilities from the cloud to the edge side.The ability to provide users with lower-latency services will play a key role in the future of intelligent vehicle networking.This article consists of the following three parts:(1)A vehicle task offloading algorithm based on simulated annealing algorithm is proposed.The problems of inter-vehicle interference,unloading ratio allocation,MEC server communication resource allocation,and delay minimization are comprehensively considered.The optimization problem is decomposed into two sub-problems.On the one hand,the water injection algorithm is used to solve the channel resource allocation scheme,and the two-stage allocation is used to make all vehicle communication rates meet the task requirements;at the same time,the channel allocation result is used as input,and the simulated annealing algorithm is used to solve the unloading ratio allocation problem.The solution solves NP-hard problems that are difficult to solve by traditional mathematical methods.By continuously reducing the temperature,an approximate global optimal solution is obtained.Finally,the simulation proves the superiority of this algorithm compared with other algorithms,which minimizes the average delay of all users,takes into account the moderate network traffic,and can play the best performance in different network conditions.(2)Aiming at the problem of insufficient RSU computing resources in densely populated places such as urban roads,a joint optimization algorithm of RSU caching and vehicle computing task offloading based on D2 D assistance is proposed.On a large time scale,the data of hot tasks in a certain area is cached to reduce the upload of repeated data and reduce the task delay.On a small time scale,D2 D is used to unload to the idle oncoming vehicle to assist unloading,so as to solve the problem of insufficient computing resources.Firstly,the network model of the system is established,which is divided into V2 R communication model and V2 V communication model,and then the problem model is established with the combination of cache factors.The tasks in the terminal can be executed locally,D2 D offloaded or offloaded to RSU.If there are computing tasks cached in the RSU,you can directly compute without uploading the tasks.Then,the weighted sum minimization problem of vehicle task delay and energy consumption is established.After using the knapsack algorithm to determine the optimal cache resource allocation,the KM algorithm and the interior point method are used to solve the optimal task offloading strategy problem and communication resource allocation problem respectively.Finally,the comparison with other algorithms proves that this algorithm has advantages in both time delay and energy consumption.(3)Based on the idea of MEC offloading and shunting,an edge network environment is actually built to offload and shunt core network user requests to designated servers.There are two main functions,namely,the access control system based on LTE small cell and the user security control system based on MEC.The access control system can do user perception based on terminal signal strength and fully control user access network permissions.MEC can unload traffic by formulating an unloading strategy and build a user security management and control platform to manage and control the target URLs accessed by users,effectively reducing the infringement of bad websites.Each function is tested separately to verify the effectiveness of the system.