Task Offloading And Resource Allocation Strategy Research Of Vehicular Network Based On Edge Computing

In recent years,technological innovation has made big strides in the areas of advanced road transportation and material living standards and has reached a new level.For the car,cars equipped with technologies such as autonomous driving,augmented reality(AR)navigation,and intelligent voice control are more competitive.However,such computing-intensive and latency-sensitive applications require the vehicle to have high computing power,which will undoubtedly increase the manufacturing cost of the vehicle.To overcome this challenge,this paper integrates Mobile Edge Computing(MEC)with the Internet of Vehicles,aiming to provide computing and storage services for vehicles near the roadside through edge computing technology and expand the task processing capabilities of vehicles.This paper mainly studies the computing offloading and resource allocation strategy on the Internet of Vehicles environment.The main research work is as follows:In the Vehicular Edge Computing(VEC)system,the processing efficiency of vehicle tasks is low,and urgent tasks cannot be prioritized.A computing offloading strategy based on the sparrow search algorithm is proposed.First,according to the indicators that affect the quality of service(Qo S)of users in the VEC system,a mathematical model is constructed to minimize the task processing delay and resource service charges.Then,according to the data volume of the task,the maximum tolerated delay,and other attributes of the vehicle,and use the analytic hierarchy process assigns the processing priority to each task.Finally,according to the utility function of the system,the sparrow search algorithm is used to obtain the optimal offloading decision.The simulation results are compared to commonly used benchmark algorithms,the algorithms in this article effectively reduce resource service charges and task processing time,improve task processing efficiency,and are optimally useful for the system.Aiming at the problem that the traditional heuristic search algorithm cannot perceive the changes of the VEC system channel environment in real-time and the load of the MEC server is unbalanced,this paper proposes a calculation offloading and resource allocation strategy based on meta-reinforcement learning.First,this paper divides the original optimization problem into two sub-problems: computing offloading and resource allocation.Scientific computing offloading decisions can effectively reduce the task processing delay,and a reasonable resource allocation strategy can balance the MEC server load.Then,this paper adopts a discrete meta-reinforcement learning algorithm to solve the optimal offloading decision under the current channel conditions according to the information of the channel state,task attribute,and system utility function value.Finally,once the offloading decision is determined,the resource allocation strategy can be solved directly by the KKT condition.The simulation results show that the algorithm has a fast convergence speed,can quickly adapt to the changes of the channel environment,effectively improve the task processing efficiency,and realize the load balancing of the MEC server.