Research On MEC Mobility Management And Offloading Algorithms In The Internet Of Vehicles Scenario

With the wave of intelligence sweeping the world,vehicles are becoming more and more intelligent.In smart cars,there are a large number of computationally intensive tasks.These tasks are not only computationally intensive,but also require low latency and low energy consumption.The traditional Internet of Vehicles cannot meet these two requirements.Mobile Edge Computing(MEC,Mobile Edge Computing)provides nearby computing services with features such as low latency and low energy consumption.Since it was proposed in 2014,it has become a research hotspot.However,in actual application scenarios,the mobility of the vehicles may cause the phenomenon of base station switching,which will eventually lead to the failure of task offloading,as well as the problem of computing resource allocation and task offloading-deision-making under the limited computing resources.Faced with these challenges,this article has done the following research:(1)Aiming at the offloading failure of computing tasks caused by the mobility of vehicles,a genetic algorithm-Kalman Filtering model(Genetic Algorithm-Kalman Filtering,GA-KF)offloading algorithm is proposed.In the process of problem modeling,an adaptive dynamic weight calculation method is proposed,which converts the optimization of task delay and energy consumption into optimization of task overhead.The Kalman Filtering algorithm(KF)is used to predict the moving path of the vehicles during the driving process of the vehicle,and a stable base station is selected for task offloading.In order to improve the accuracy of the KF model for vehicle path prediction,this paper uses Genetic Algorithm(GA)to optimize the observation noise in the KF model.Simulation experiments show that the GA-KF offloading algorithm performs well in the optimization of overhead,time delay,and energy consumption,and improves the success rate of task completion.At the same time,compared to the offloading scheme with solid weights,the comprehensive performance of the offloading scheme with adaptive dynamic weights is better.(2)In the mobile edge computing of the Internet of Vehicles,an iterative optimization scheme based on block coordinate descent technology with joint convex optimization and gray wolf algorithm(BCD-CONGW)is proposed for joint computing resource allocation and task offloading deision-making in mobile edge computing for the Internet of Vehicles.In the process of problem modeling,local computing resources,edge computing server resources of small base stations,edge computing server resources of macro base stations,and cloud computing server resources are considered,and six offloading strategies are proposed to jointly optimize computing resource allocation and task offloading decisions.Get a mixed integer nonlinear programming problem.It is very difficult to solve this problem,and there is no mature method to solve it directly.Based on the block coordinate descent method,this paper decomposes the original problem into two sub-problems,namely computing resource allocation problem and offloading decision problem,and obtains the optimal computing resource allocation plan and offloading decision plan by circular iterative solving.The experimental results show that under the condition of limited computing resources,the proposed resource allocation scheme and offloading decision can produce lower overhead,delay,and energy consumption.