Research On The Adaptive Video Streaming Collaborative Caching And Transcoding Based On Wireless Edge Technology

The rapid development of wireless communication technology and personal mobile devices has played a strong role in promoting mobile Internet services,and has also led to rapid growth of network traffic.Among them,video streaming occupies a considerable part of the network traffic.The rich video content service and advanced mobile devices support video streaming with a variety of bit rate versions,providing users with a variety of choices,but also bringing great challenges to the transmission and distribution of video streaming.Currently,cloud-based DASH system is used to solve the diverse needs of users,but the system still has some problems.Firstly,in cloud-based system,videos are stored in the cloud which is far away from the user,which will inevitably produce excessive delay.Secondly,the same video needs to be transmitted between core networks multiple times,which greatly wastes the bandwidth resources of the core network.Thirdly,videos require huge storage space,and transcoding video into different bit rate versions is also a compute-intensive task,so cloud servers that carry many areas of video storage and transcoding are under great pressure.The mobile edge computing and wireless edge caching technology allows video storage and transcoding in edge network close to the user,which helps to alleviate the pressure on the cloud servers,reduces the user response delay,and reduces the bandwidth pressure of the core network.The mobile edge computing and wireless edge caching technology gradually become a hot issue of research.In addition,the popularity of crowdsourcing services and the rapid development of mobile devices make it possible to participate in transcoding with a broad user base.In order to solve the problems existing in the current system,this thesis focuses on the caching and transcoding technology of adaptive video streams in wireless edge networks.The main content of the thesis include:(1)A wireless edge cooperative caching and processing system is proposed.This thesis considers the problem of Base Station's neighborhood and the limitation of wired bandwidth,wireless bandwidth,storage capacity and computing capacity.Then the problem is formulated as a network utility maximization problem for edge networks,and is proved to be NP-Hard through analysis.A CCP algorithm based on computing capacity utilization of Base Station is proposed for video content caching.A heuristic routing scheduling algorithm GHRS is designed for video requests,which determines the routing of video requests and the scheduling of transcoding resources.On this basis,in order to find a better solution,iteration algorithm GHRS-Iter is designed.Simulation results show that our proposed solutions achieve significant increase in terms of network utility of edge system and cache hit ratio and decrease in response delay.(2)A wireless edge cooperative crowdsourcing transcoding system is proposed.As far as we know,it is the first system to consider crowdsourcing services for video transcoding in edge networks.The problem is modeled as an auction model,and an incentive-compatible auction mechanism is designed for the two-dimensional bidding information,with the aim of maximizing social welfare.An algorithm GDA that conforms to economic attributes is designed for our scenario to select the appropriate users to participate in the transcoding tasks.Finally,extensive simulations prove that the system can make a relatively large contribution to the network utility of the edge system,and the proposed GDA algorithm has a great performance improvement compared with the traditional auction algorithm.