Researches On Method Of Edge Cache And Task Offloading For Internet Of Vehicles

With the rapid progress of society,the mobile Internet and the Internet of Things are deeply integrated and new control application scenarios such as the Internet of Vehicles(Io V)and UAV networks are emerging rapidly.Applications in such scenarios put forward higher requirements for network throughput and network latency.This thesis comprehensively analyzes the two aspects of edge caching and task offloading for Internet of Vehicles,and proposes content caching strategies based on edge computing and task offloading strategies based on vehicle fog computing,aiming to reduce system energy consumption and delay.The main work of this thesis is as follows:In order to solve the problem of high response delay in the Internet of Vehicles,this thesis proposes an edge caching scheme of Internet of Vehicles based on multi-agent reinforcement learning.Each moving vehicle in the Io V can be seen as an agent,and agents within communication range can cooperate with each other,and they can make content caching and content access decisions adaptively according to the changing environment to minimize the delay in the process of content distribution.Experimental results show that the proposed edge caching scheme has a better performance in reducing the delay in content distribution and improving the hit rate and success rate compared with other methods.In most cases,roadside units(RSU)deployed on rural roads have limited energy and need to unload some tasks to atomized vehicles.Therefore,this thesis proposes a vehicle fog computing task offloading framework for efficiently offloading tasks generated by smart devices.On this basis,this thesis proposes a Q-learning task offloading algorithm based on fuzzy logic.The algorithm offloads the task to the atomized vehicle under the constraints of the maximum tolerable time delay and resource availability of the task,thereby reducing the energy consumption of the roadside unit and the response time and improving the quality of service of users.Experiments show that the task offloading strategy proposed in this thesis has better performance than other algorithms.