| As a reliable caching paradigm,edge caching technology has been widely applied in the Internet of vehicles.By deploying caching services on base stations and road side units,the request delay of users can be greatly decreased and the service pressure of the system can be effectively reduced.However,with the continuous development of Internet of vehicles technology,the amount of data required by on-board applications surges,and the edge nodes need to handle massive user requests.Due to the edge nodes’ poor caching capability,the caching decision performance on a single edge node deployment is bad,and it is also vulnerable to the dynamic network topology and wireless environment caused by the uneven distribution of vehicles.Therefore,this thesis studies the cooperative caching scheme and content delivery strategy among edge nodes.First,to solve the problems of high request delay and poor service experience of users at the edge of the cell,this thesis proposes a DDPG-based joint cooperative caching and beamforming coordination(JCCBC)algorithm jointly optimizing the decision on content caching and wireless transmission with the aim of minimizing content delivery delay.Based on the double time-scale scheduling model,the JCCBC algorithm considers the current cache state and the limited of wireless transmission resources on the base station at a small time scale,and jointly optimizes the user selection strategy and the beamforming coordination to minimize the content transmission delay.Considering the content request distribution during a large time scale,the cooperative cache strategy among base stations is optimized to minimize the content retrieval delay.The simulation results show that the JCCBC algorithm has better performance in content retrieve and transmission.Secondly,to solve the problem of high load in the base station,this thesis proposes a platoon caching strategy based on Stackelberg game,and designs an algorithm(Binary Search and Genetic Algorithm based Hybrid Algorithm,BGHA).Aiming to maximize the utility of base station and platoons,the bid decision of base station and cache decision of platoon are optimized respectively by binary search and genetic algorithm.In this game model,the base station will incentive platoons to cache with payment.According to the cache state of the platoons,the base station uses the bid decision algorithm based on binary search algorithm to optimize the current bid decision,so as to achieve the purpose of motivating the platoons to actively cache more contents.According to the bid decision of the base station and the cache decision of other platoons,platoons will consider the impacts of communication cost and caching cost comprehensively and then make the caching decision on the platoons with the cache decision algorithm based on genetic algorithm so as to maximize the utility of the platoon.Simulation results show that the BGHA algorithm can increase the service scope of contents and provide more rich content services.Compared with the vehicle caching strategy,BGHA algorithm can improve the cache hit ratio and reduce content delivery delay at the same time. |
