Research Of Comiputing Offloading Based On Multi-relay Cooperation For Internet Of Vehecles

With the convergence of computer,communication and intelligent automobile technologies,research in the field of vehicle networks has reached a new level.The market competition among a large number of smart car brands is the direct reason to promote the evolution of automotive networks towards intelligence,automation and security.The network can be specifically divided into vehicle ad-hoc network and Internet of Vehicles.Different from the single mode of vehicle-to-vehicle,vehicle-to-infrastructure connection in the ad-hoc network,the focus of the Internet of Vehicles is to integrate vehicles,things,people and the environment to form a larger network,which is designed for road safety,intelligent traffic management,real-time Information and data sharing services.In the development process of Io V,a large number of emerging applications have been developed.And these applications create higher demand for computing resources and processing delays.The existing cloud computing is difficult to meet the needs of requirements.How to solve the problem of computing offloading caused by insufficient resources in the Internet of Vehicles has become a severe challenge.At present,mobile edge computing is a key vehicle connection technology that adds computing capabilities to roadside units,and researchers have introduced MEC technology into the Internet of vehicles to solve the above problems.As one of the architectures introduced by 5G,mobile edge computing can meet the computing requirements of exponential growth.In addition,MEC is well-suited for IoV scenarios due to its advantages such as being more efficient,smarter,and more flexible.However,for edge computing,the mobility and dynamic network topology characteristics of the Internet of vehicles lead to many limitations of connection and service interruption for intelligent vehicles.What’s more,the computing power of the MEC server is not infinite.How to improve the scalability of MEC servers and provide services for more intelligent vehicles is also a challenge.Therefore,in a specific scenario,modeling after analyzing resource management and vehicle terminal mobility,and formulating a series of MEC strategies or specialized technical solutions to adapt to the Internet of vehicles deserve in-depth studies.For the sake of those problems,this paper studies the vehicle network supporting MEC service.In the IoV scenario where computing resources are scarce,we propose a computing offloading strategy based on multi-relay collaboration.This strategy can optimize the task processing delay of high-speed mobile users and improve service continuity.The main work of this article is as follows:Firstly,aiming at the problem of limited computing capacity of MEC server,a multi-relay collaboration-based computing task offloading strategy is proposed.It is found that the task offloading in the IoV supported by MEC can be optimized by relay vehicle assistance.In the computing offloading model,especially in the case of low scalability of the current Internet of Vehicles,the joint scheduling of auxiliary relay vehicles has an obvious effect on improving the situation.Specifically,there are some computing tasks in this area where it is difficult to obtain resources.Pre-offloading tasks to a resource-rich MEC server can reduce the overall latency of transmission and calculation.Because the combinatorial optimization problem is too complicated to be solved,a low complexity algorithm is designed.In addition,the computational transfer strategy has been proved to be useful through experiments.It implements load balancing among multiple MEC servers,and alleviates the low scalability of IoV.Finally,simulation results show that the proposed scheme optimizes task computing efficiency and offloading latency.Secondly,in order to solve the complex problem in the computing-intensive offloading scenario in the Internet of vehicles,we propose a cache-enhanced relay offloading strategy.It can reduce the resource consumption and task processing delay of MEC equipment for repetitive tasks.At first,by analyzing the computing task eigenvalues,the results of popular tasks are cached in the MEC server.Meanwhile,task eigenvalues are stored in categories based on task content.Compared with tasks that have cached results,they only need to complete characteristic value comparison and data download,other intelligent vehicle tasks need to perform offloading,calculation,and backhaul processes.Then,according to the mutual influence of caching and computing capacity,the optimal MEC server parameter information is obtained through analysis.On this basis,this paper proposes a cache-enhanced task computing offloading strategy to optimize the total delay of all task processing in IoV.Wha’s more,based on the relative relationship between pricing strategy and resources,a collaborative scheduling algorithm is designed to assist vehicles in determining when to obtain MEC assistance.Finally,simulation results show that the strategy can effectively reduce the calculation offloading delay.