An empirical analysis of network coding in peer-to-peer video-on-demand systems

Abstract

In recent years, Peer-to-Peer (P2P) multimedia streaming has become an alternative to cable/satellite TV services. Many P2P streaming applications further provide users with DVD-like operations: play, pause, chapter selection, fast forward, and rewind. Such a real-time interactive multimedia streaming, commonly referred to as P2P Video-on­Demand (VoD), poses unique challenges in providing smooth playback and seamless interaction over the Internet. To improve the streaming quality, recent research employs network coding as the key enabling technology. However, the practicality and imple­mentation challenges of network coding received very little attention. In this thesis, we present a practical implementation of network coding in a P2P VoD system, based on which we identify the actual performance gain and design pitfalls when incorporating network coding in a P2P VoD system. Our performance analysis shows that, unlike P2P live streaming, directly applying network coding to a P2P VoD system does not necessar­ily lead to an immediate improvement in playback quality. Based on analysis, we propose Coded VoD, a new approach for P2P VoD streaming, to exploit the content sharing op­portunities in a P2P VoD session. It achieves smoother playback and shorter response time for DVD-like operations by utilizing a network-coding-based content prefetching mechanism.