Research On Joint Computing Offloading And Resource Allocation Base On MEC In Internet Of Vehicles

With the rapid development of the Internet of Vehicles,various delay-sensitive invehicle applications have emerged,such as autonomous driving,augmented reality,image processing,etc.However,due to the limited computing resources and energy of existing vehicles,they cannot meet the requirements of these applications for time delay and energy consumption.Mobile cloud computing alleviates the application’s demand for computing resources,but because the remote cloud server is far from the vehicle,the communication delay of task offloading is relatively large.Mobile Edge Computing(MEC)reduces the processing time delay and energy consumption of vehicles by sinking computing resources to the edge of the road.However,due to the limited resources of the MEC server,it may still not be able to meet the offloading requests of all vehicle computing tasks.Therefore,it is very important to comprehensively consider the characteristics of each computing platform and task,and design a reasonable and efficient computing offloading and resource allocation strategy.This thesis focuses on the joint computing offloading and resource allocation algorithm base on MEC in Internet of Vehicles.The main contents are as follows:Aiming at the problem of concurrent offloading of computing tasks for multiple vehicles,an adaptive joint computing offloading and resource allocation algorithm based on multitasking is proposed.Considering the task data volume of each vehicle,the tolerable delay threshold,computing resources,and network bandwidth factors,the computational offloading decision and resource allocation are modeled as a mixed-integer nonlinear optimization problem.A particle matrix coding method is proposed to adaptively adjust the offloading platform and offloading ratio of computing tasks,and at the same time allocate the computing resources of the MEC server.Based on the multi-level penalty function and the particle swarm optimization algorithm with constriction coefficient,a particle correction algorithm is proposed to solve the optimization problem.The simulation results show that the algorithm can minimize the total cost of the system while meeting the maximum tolerable delay.Aiming at the dependency delay caused by the dual dependency of timing and data between subtasks when offloading,and the problem of load unbalance of the MEC server,a joint computing offloading and resource allocation algorithm based on dual dependency tasks is proposed.First,a directed acyclic graph is used to describe the task dependency,and the dual dependency of the subtasks are considered to establish the time delay and energy consumption model of each computing platform.Then,based on the bidding thought of the auction algorithm,the allocation model of computing resources and communication resources is established,and the load balance model of the MEC server cluster is established according to the load status.Secondly,the optimization problem of joint optimization offloading decision,resource allocation,and load balance is constructed.Finally,an adaptive particle swarm genetic hybrid algorithm is proposed to solve the optimization problem.The simulation results show that the algorithm can reduce the total system overhead while meeting the maximum tolerable delay,and effectively improve the load balance of the edge server cluster.