Research On Task Allocation And Offloading Mechanism In Internet Of Vehicles Based On Fog Computing

Interenet of Vehicles based on fog computing is a new network architecture model,which aims to provide fast and powerful computing services for mobile vehicle users.By introducing the fog computing paradigm into the Internet of Vehicles,this new network architecture model makes services closer to mobile vehicle users and improves the performance of delay-sensitive applications.However,the number of vehicle users offloading independent and interrelated computing tasks to fog servers is increasing rapidly,which leads to overload of fog servers and expands the risk of service interruption.At the same time,mobile vehicles have abundant idle computing resources that have not been utilized,which provides another idea for task allocation and offloading in Interenet of vehicles based on fog computing.In view of the task offloading problem faced by Interenet of vehicles based on fog computing,this paper with the task collaborative computing as its core,focusing on the construction of task collaborative computing mechanism and encouraging mobile vehicles to open their idle computing resources,and discusses the solutions to the task allocation and offloading problem in Interenet of Vehicles based on fog computing.The research work of this paper is as follows:(1)Aiming at the construction of task collaborative computing mechanism in Interenet of Vehicles based on fog computing,this paper proposes a location-aware task collaborative computing scheme in Interenet of Vehicles based on fog computing.In this scheme,the principle of proximity is adopted to allocate tasks between the required tasks and the cooperating vehicles.Under the condition of satisfyied the task requirements,the fog server selects a more reliable cooperative vehicles to complete the related sub-tasks required by the customer’s vehicle according to the principle of proximity.In addition,we further propose an improved Hungarian algorithm to complete the allocated between the required tasks and the cooperative vehicles,so as to reduce the fragmentation of vehicle computing resources and improve the utilization rate of vehicle idle resources.Finally,the experimental results show that our location-aware reliable task collaboration computing scheme is effective in the vehicle networking environment based on fog computing.(2)Aiming at the problem of encouraging mobile vehicles to open up their idle computing resources in Interenet of Vehicles based on fog computing,this paper proposes a task collaborative computing scheme based on Stackelberg game in Interenet of Vehicles based on fog computing.Considering the mobility of vehicles,we first propose a vehicle usability evaluation model to evaluate the usability of mobile vehicles.After selecting the candidate cooperative vehicles,we designed the corresponding utility functions for the customer vehicle and the cooperative vehicles respectively.In addition,we also use the Stackelberg game theory to model the game process between the customer vehicle and the cooperative vehicles,and deploy an iterative algorithm on the fog server to find the optimal task assignment scheme.Finally,the experimental results prove the effectiveness and superiority of the iterative algorithm.