Research On Resource Allocation Strategy Based On Latency And Energy Efficiency In Internet Of Vehicle With Cloud-edge Collaboration

With the continuous development of Intelligent Transportation Systems(ITS),the level of automobile intelligence and autonomous driving ability has steadily improved,and users of the Internet of Vehicles(IoV)have thus experienced intelligent,comfortable,safe,energy-efficient,and efficient integrated services.The development of cloud computing and Mobile Edge Computing(MEC)offloads large-scale computing tasks in the IoV to the cloud,improving the overall computing speed.Furthermore,the cloud-edge collaboration technology in the vehicle network flexibly utilizes computing and storage resources to achieve the orderly unity of MEC and cloud computing functions,and improve the performance of the entire IoV system.In the context of the IoV,Vehicle Users(VUs)perform data transmission,including video data,navigation data,vehicle driving parameters,etc.Among them,the transmission of vehicle driving parameters and other data is a high-reliability and low-latency communication that meets strict requirements for timing and reliability.Meanwhile,an increasing number of new energy vehicles are entering the IoV,and it is also an urgent issue to consider data transmission processing and vehicle charging to reduce energy consumption.Therefore,the thesis constructs an IoV communication system based on cloud-edge collaboration and investigates resource allocation strategies based on latency and energy efficiency.After the problem formulation,the Lagrange dual method and sub-gradient iteration method are used to solve the problem,and simulation verification is conducted.The main works of this thesis includes:(1)Considering the limited computing resources of a single MEC server and the possibility of forming service islands,an IoV system model based on MEC collaboration is constructed to minimize the latency.This model consists of multiple VUs and multiple Edge Nodes(ENs),where the ENs are a combination of base stations and MEC servers.After a VU generates a computing task,it selects an EN and offloads the task through wireless communication.Furthermore,this EN selects other ENs for cooperation,divides the task,and unloads some sub-tasks to other ENs to complete.On this basis,with the goal of minimizing the total latency of all VUs and the constraints of the number of cooperating ENs and the number of users served by each EN,an optimization problem is formulated.The optimization problem is a mixed-integer non-linear problem(MINLP)and is solved using the Lagrange dual theory.The problem is decomposed into task proportion allocation sub-problem and EN selection sub-problem.The KKT condition and the winner-takes-all method are used to solve the problem and propose a resource allocation strategy for task proportion allocation and EN selection.Simulation experimental results show that using the proposed resource allocation strategy can effectively reduce system latency.(2)Considering the increasing proportion of new energy vehicles,a cloud-edge collaborative IoV system model based on energy efficiency maximization is constructed.The system includes a cloud computing center,multiple VUs,multiple ENs,and multiple charging stations.VUs select the ENs with the highest energy efficiency based on the user service situation,channel quality,and energy consumption of the computation process,and work collaboratively with the cloud to complete the computing tasks.Charging tasks involve VUs selecting the ENs with the highest charging power based on the number of user services provided by the charging station in the previous time slot and then recharging their vehicles.On this basis,the optimization problem is formulated to maximize the energy efficiency of all VUs,with constraints on the user service quantity of ENs and charging stations,and the ratio of task allocation.The fractional objective function is transformed into a subtractive form and solved through two nested loops.In the inner loop,the original optimization problem is decomposed into two sub-problems to solve the task allocation and EN selection variables.In the outer loop,the optimal solution of energy efficiency is output according to the set precision requirements.Simulation results show that the proposed resource allocation strategy can effectively balance latency and energy consumption and improve the performance of the system for cloud-edge collaborative IoV.