Research On High Scalable SVC Video Transmission Technology Based On IPv6

In the past few years,video streaming services have become more and more popular.Video traffic has accounted for the vast majority of Internet traffic,and is growing steadily every year.However,with the expansion of the scacle of network,the network structure and network environment have become more and more complex.Traditional streaming media technology cannot provide high-quality video transmission services in various complex network environments,and it is even more difficult to guarantee the quality of user experience(Qo E).At the same time,with the development of IPv6 technology,the application development and technology innovation for the next generation of IPv6 Internet is an inevitable trend.Therefore,in order to meet the growing demand for high-quality video services and requirements of integrating next generation network technology,it is urgent to design an efficient and feasible streaming media solution suitable for the requirements of modern video applications,so as to improve the quality of user experience.Scalable video coding technology(SVC)can encode the original video into a multi-layer stream.By adjusting the level of the output stream,it can adapt to various complex network environments.Therefore,the research work of this thesis focuses on SVC technology to implement an adaptive streaming media solution in IPv6 network,and verifies the feasibility and efficiency of the solution in various complex application environmentsFirstly,because of the defects of traditional streaming media technology and ABR scheme,a scalable streaming media service system is designed and implemented by using SVC technology combined with HTTP transport protocol.The system consists of a scalable streaming media server,a video receiving client and a management server.At each decisionmaking time,the system can dynamically calculate the layer of the SVC video stream according to the measurement data and the information from the client,and then extract and encapsulate the video stream to provide adaptive video on demand service for users.Secondly,the thesis analyzes the SVC video streaming service,and concludes that there are four main factors closely related to the quality of user experience,namely,the average quality of SVC video,the client cache state,the frequency of layer switching and the layer switching range.On this basis,a Qo E evaluation model for SVC video streaming service is designed.Thirdly,this thesis proposes a solution to the problem that SVC is lack of efficient layer adaptive selection strategy.From the perspective of user Qo E,the layer adaptive selection problem of SVC video stream is modeled as a Markov decision process(MDP)model,and the Q-learning algorithm is used to solve the optimal strategy.Finally,the thesis uses the MDP-based SVC layer adaptive selection algorithm as the implementation of the layer computing model in the scalable streaming media server-side,and deploys the scalable streaming media service system in the experimental network.Compared with other adaptive selection algorithm in the simulated IPv6 network and the actual IPv6 environment,our algorithm can provide users with higher video quality,more balanced cache and smoother handoff,so as to bring users better Qo E.In addition,experiments in the SRv6 network verify the adaptability of the system and algorithm to the next-generation IPv6 technology.