Research On Techniques Of Video Content Caching And Delivery In D2D Communications

With the rapid development of mobile communication and multimedia technology,multimedia service will gradually become the main traffic in future network.How to load massive video data and improve the user's service experience is significant.Under this incentive,device-to-device(D2D)communications has been proposed.By sharing spectrum,D2 D communication effectively improves the network capacity.Moreover,it is considered as one of key technologies in the next generation mobile communication(5G).In this dissertation,we combine content popularity,user requirement,service characteristics and network framework to develop key technologies for content caching and delivery in D2 D communications.By offloading massive repetitive or similar video data,the traffic in the core network can be effectively reduced.At the same time,the user experience is improved.The four main contents and contributions of this dissertation are summarized as follows:Firstly,aiming at the problem about user random distribution and limited cache,a video content delivery scheme based on clustering is proposed.Using the user's media cloud and D2 D communication,we have completed a video content distribution.The adaptive media cloud mechanism and minimizing D2 D hopping algorithm are used to build a reasonable cache deployment based on popularity and user distribution.The simulation results show that the response time is reduced by 50%,which effectively offsets the impact of popularity changes.Secondly,we analyze the relationship among quality of experience(QoE),network-related parameters,based on different service requirements and personalization of user requests.A QoEdriven distribution scheme is proposed.And we design an adaptive and dynamic update algorithm based on QoE model and also propose a resource allocation algorithm.The system effectively utilizes caching in this multi-service scenario and reduces transmission interference.Compared with the traditional method,the total user quality of experience in the network is improved about 20%.Thirdly,aiming at the heterogeneity and transmission interference in the future communication networks,we analysis the cache hitting ratio,optimal D2 D communication radius,maximizing link access and other related performance in multi-layer content distribution networks.Further,we propose a link selection and scheduling algorithm,which guarantees the threshold about signal to interference plus noise ratio and signal to interference ratio.The algorithm obtains the relevant results to the theoretical analysis,effectively reducing the number of interference by 40% and increasing the system throughput by 10%.Finally,in order to meet the actual needs of different types of services and applications,we propose a multi-layer smart city framework.And we provide an optimum resource allocation strategy for service content deployment and computing.Moreover,we design an optimal deployment scheme for network services based on genetic algorithm.The algorithm implements global optimization of calculation and caching,shortens 20% service waiting time,and improves the quality of user service experience.