Research On Vehicle Edge Computing Task Offloading Strategy

In recent years,Internet of Things(Io T)technology and the automotive industry have experienced rapid development.At the same time,5G communication networks are becoming increasingly popular,and the era of true interconnectivity of everything has arrived.In the field of Internet of Vehicles(Io V),various compute-intensive and delay-sensitive applications are emerging,which presents a severe challenge to onboard systems with limited computing capabilities.Mobile Edge Computing(MEC),one of the key technologies of 5G,can effectively address the problem of insufficient computing power of user vehicles and reduce data processing latency and terminal energy consumption by sinking storage and computing capabilities to the edge.However,there are also many vehicles on the road with idle computing resources.How to design an offloading scheme to utilize these idle resources to further enhance the vehicle’s edge computing service capabilities is a hot research topic today.In this thesis,the problems of high vehicle maneuverability and distant central cloud distance in highway scenarios,which lead to large data unloading delays,are first addressed.To minimize unloading time and energy consumption,a distributed vehicle edge computing unloading solution based on Genetic Algorithm(GA)is proposed.The solution takes into account the cooperation among local vehicle terminals,edge servers on roadside base stations,and surrounding idle vehicles,and can minimize the total cost of task execution,including delay and energy consumption,while meeting the maximum delay constraints of current tasks.Finally,the correctness and stability of the proposed solution are verified through simulation,and comparisons with other solutions show that the designed solution has significant advantages.In addition,a centralized vehicle edge computing unloading solution based on Particle Swarm Optimization(PSO)is proposed in this thesis to address the problem of insufficient computing resources of vehicles to handle computationally intensive and delay-sensitive tasks,in the context of sufficient coverage of cellular base stations but relatively few roadside base station devices on current urban roads.The solution combines wide coverage cellular base stations in urban areas,fixed and mobile edge nodes within the area,and optimizes the weighted sum of delay and energy consumption as the optimization objective,subject to the maximum delay tolerance of user vehicles in the area,by using an improved Particle Swarm Optimization algorithm(POS-X)to solve the optimization problem.Simulation results demonstrate that,under the same conditions,the proposed solution has significantly lower average delay and energy consumption compared to other three algorithms,and the algorithm convergence speed is relatively fast,indicating the significant advantages of the designed solution.