Topology construction and resource allocation in P2P live streaming

Research output: Chapter in Book/Report/Conference proceedingChapter

5 Scopus citations

Abstract

We identify inefficient network topologies and resource allocation mechanisms to be the key obstacles toward adopting present P2P systems as the platform of choice for delivering high-quality live multimedia content. To this end, we propose a comprehensive optimization framework that overcomes these challenges in a systematic manner. In particular, we design a delay-based network construction procedure that creates small-world topologies that provide for efficient data delivery routes and increased data sharing between peers in the network. Furthermore, we equip the nodes with a utility-based packet scheduling technique that maximizes the video quality at a receiving peer subject to available bandwidth resources while allowing for dissemination of less frequently encountered data in the network. Finally, we design an uplink sharing strategy that enables the peers to deal effectively with free-riders in the system. Through theoretical analysis and simulation experiments we demonstrate that the proposed protocols provide for substantial improvements in performance in mesh-pull based P2P live multimedia delivery. Specifically, significant gains are registered over existing solutions in terms of average video quality and decoding rate. The proposed mesh construction procedure provides further gains in performance in terms of reductions in frame-freeze and playback latency relative to the commonly employed approach of random peer neighbour selection. Corresponding gains in video quality for the media presentation are also registered due to the improved continuity of the playback experience.

Original languageEnglish (US)
Title of host publicationIntelligent Multimedia Communication
Subtitle of host publicationTechniques and Applications
EditorsChang Wen Chen, Zhu Li, Shiguo Lian
Pages217-251
Number of pages35
DOIs
StatePublished - 2010
Externally publishedYes

Publication series

NameStudies in Computational Intelligence
Volume280
ISSN (Print)1860-949X

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Keywords

  • Distributed control
  • IPTV
  • Multimedia communications
  • Packet scheduling
  • Peer-to-peer networks
  • Resource allocation
  • Video streaming

Fingerprint

Dive into the research topics of 'Topology construction and resource allocation in P2P live streaming'. Together they form a unique fingerprint.

Cite this