Application Of Scalable Video Coding Technology Of P2p Streaming Media Subdivision Algorithm Research

Scalable video coding (SVC) is a new video codec technology, with whichdifferent video qualities of the encoded video are optional. The image quality can beselected from three dimensions, which are resolution, frame rate and image detail.With the corresponding quality of the video data extracted, video can played in thedegree of quality.With the rapid development of mobile network, mobile devices accounted for anincreasing proportion in the Internet. With the development of IPTV, high definitionTVs as network terminals are participated in the network, the current network is turnedinto heterogeneous network. In heterogeneous network, hardware conditions ofnetwork terminals subject to a variety of processor performance, display size andnetwork bandwidth restrictions. The video quality of streaming media of each networkterminals is in big difference. Scalable video coding is a perfect selection to meet theviewing demands of video of the different terminals in heterogeneous networks.P2P network transmission is mature now, which has low cost advantages and canimprove the bandwidth utilization of network terminal. In this paper, we research themethod of application of the SVC codec in P2P streaming system, mainly forsegmentation algorithm design of P2P transmission and encapsulation methods ofvideo.In order to retain the characteristics of the SVC video quality scalable during theP2P transmission, SVC video data is separately placed according to the difference ofquality. Such an approach may result in that data of different quality in playback is notsynchronized. We did a large number of video data analyses in the second chapter, andquantified the not synchronized seriousness by sampling simulation.How to solve the data synchronization is demanded to consider when designingthe segmentation algorithm. The key to solve the problem is to make sure each blockcarries playback time information. On the basis, two segmentation algorithms aredesigned. Through characteristics analysis and experiment, we analyzed their advantages and disadvantages of the two segmentation algorithms.In order to satisfy that video can be played on most the platform, MP4media fileformat is selected. SVC encapsulation method in the ISO standard is very complex,and will raise a large number of metadata, but also not be able to meet the requirementsof the P2P transmission. So we decided to design a new encapsulation method. Theencapsulation method designed in this article is match to the segmentation algorithm.This method is not very complex to implement, but in the playback data scheduling,some extra computing is needed.