Research On Video Distribution Strategy In Cloud Radio Access Network

Cloud radio access network(C-RAN)has the advantages of energy-saving,high spectrum utilization,good extensibility,resource pool sharing,cooperative anti-jamming and so on.It can provide users with access services with large capacity,high reliability and low delay when combined with passive optical network.Meanwhile,with the wide application of multimedia intelligent terminal equipment,video service become the main business of the fifth generation of cellular system(5G).As the technology choice of 5G network,efficient video distribution strategy in C-RAN is particularly important to improve the quality of experience(QoE).Therefore,this thesis focuses on the video distribution strategy in C-RAN.The main researches are as follows:Firstly,the research background,architecture evolution,bearer mode and research hotspot in C-RAN are introduced,and the research hotspots in C-RAN are analyzed.Then it focuses on the video distribution strategy in C-RAN and the characteristics of the existing distribution strategy.Finally,the summary is given at the end of this chapter.Secondly,a video distribution strategy based on RRH collaboration in C-RAN is proposed to improve the poor performance of cell-edge users and heavy-load cell users,which caused by dense network and load imbalance respectively.According to the degree of interference between cells,node transmission bandwidth and request content similarity,cooperative gain-aware virtual passive optical network is established to provide cooperative caching and transmission for video streaming by adopting collaborative approach in optical domain and wireless domain.Furthermore,user experience for video streaming,bandwidth provisioning and caching strategy are jointly optimized to improve QoE,which utilize the methods of dynamic caching in optical domain and buffer level-aware bandwidth configuration in wireless domain.The results show that the proposed mechanism enhances the quality of user experience and effectively improves the cache hit rate.Next,a video distribution strategy based on end-edge-cloud collaboration is proposed to alleviate the contradiction between user experience and network resources.Considering the zero buffer interruption time and the spectrum efficiency of the network during cluster handover,the algorithm of alliance formation based on merging andsplitting rules is used to construct virtual passive optical network to avoid the interruption of wireless link transmission.Furthermore,tolerable video interruption latency and version transition latency are analyzed to optimize quality adaptation and wireless bandwidth allocation.Finally,according to the bandwidth-hit ratio,the user priority and the link rate matching relationship in the queue model,the fronthaul/backhaul bandwidth and cache resources are jointly configured to ensure the reliability and high efficiency of service transmission.The results show that the proposed mechanism can effectively enhance the quality of user experience and avoid network congestion.Finally,the research works are concluded and the future works are presented.