The Design of VCR-based P2P VoD Streaming Systems

碩士 === 雲林科技大學 === 資訊工程研究所 === 98 === Peer-to-Peer (P2P) network structure is one of possible solutions to efficiently support large-scale Video-on-Demand (VoD) system. Due to the asynchronous interactive behaviors and dynamic join/leave of users, it is a challenge to support full digital video casse...

Full description

Bibliographic Details
Main Authors: Sih-Ping Huang, 黃思萍
Other Authors: Ching-Lung Chang
Format: Others
Language:zh-TW
Published: 2010
Online Access:http://ndltd.ncl.edu.tw/handle/41511631291818093715
Description
Summary:碩士 === 雲林科技大學 === 資訊工程研究所 === 98 === Peer-to-Peer (P2P) network structure is one of possible solutions to efficiently support large-scale Video-on-Demand (VoD) system. Due to the asynchronous interactive behaviors and dynamic join/leave of users, it is a challenge to support full digital video cassette recording (VCR) functionality (such as Random access, Fast-forward, Fast-backward, etc.) in P2P video streaming system. Many methods currently proposed for P2P-based video streaming are difficult to support interactive video streaming operations, which incur high cost for video source searching and response latency. In this paper, we propose an Interleaving Video Frame Distribution (IVFD) scheme that has high VCR operation performance in P2P environment. Different from the current proposed which distributes the continuous video segments to the peer users, the IVFD distributes discontiguous video frames from different Group of Picture (GOP) to the peers. Based on the IVFD, the client peer would simultaneously connect to the server peers that hold the required video frames, instead of only making connection with single server peer. Under the IVFD, the client peer doesn’t need to do video data search and to make new connection with new serve peer during interactive VCR operations. It has zero latency when the client peer does operations among Random access, Fast-forward, and Fast-backward. Simulation results reveal that the proposed scheme has a great performance results in system capacity, response latency, and operation overhead.