Title: StatStream: providing statistical reliability guarantees in peer-to-peer live video streaming

Authors: Aravindan Raghuveer, Yingfei Dong, David H-C. Du

Addresses: Yahoo! Labs, Bangalore 560071, India. ' Department of Electrical Engineering, University of Hawaii, Hawaii 96822, USA. ' Department of Computer Science and Engineering, University of Minnesota, MN 55455, USA

Abstract: Although many large-scale P2P streaming techniques have been proposed, ensuring video reliability remains a challenging issue. We present a P2P streaming framework called StatStream that provides statistical guarantees on video quality and stability. We first propose a model to quantify video quality and stability. We then develop graph construction algorithms that ensure statistical guarantee on video quality at every peer. A key advantage of the proposed approach is that video quality and stability can be explicitly controlled. Simulations show that the proposed algorithms use the server resources efficiently while significantly improving the video stability at peers.

Keywords: peer to peer streaming; streaming reliability; P2P streaming; video streaming; live video; video reliability; video quality; video stability; simulation.

DOI: 10.1504/IJAMC.2010.030004

International Journal of Advanced Media and Communication, 2010 Vol.4 No.1, pp.4 - 30

Published online: 03 Dec 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article