On-Demand P2P Video Streaming: Integration of Video Coding and Network Application Design

Polytechnic University Of New York, Brooklyn NY

Investigators

Abstract

Existing video streaming networks (VSN) rely on a few dedicated servers to store and distribute videos. Such architecture is very costly, both in terms of server cost and Internet-connection cost. Failures of a single server can also cause the collapse of the entire system. This research project investigates a peer-to-peer based VSN that has the promise to provide low-cost and yet high quality video on-demand service. The system encodes each video into multiple equally important sub-streams (called descriptions) and places each sub-stream on a different peer. When a client wants to see a video, multiple peers act as servers, each sending a different sub-stream of the video to the client. When a serving peer disconnects in the middle of a streaming session, the system looks for a replacement peer that stores the same video sub-stream and has sufficient surplus uplink bandwidth. The video coder is designed such that there is only a modest degradation in video quality in the period before a replacement peer is located. The system design has five interacting components: 1) multiple description video coding; 2) sub-stream placement; 3) admission control; 4) sub-stream server selection; and 5) sub-stream delivery. Both analytical studies and experimentations are conducted to examine the system performance and its gain over infrastructure-based systems. Successful development of the system will enable people to, at very low cost, search and view on-demand enormous libraries of digital video content for the purpose of education, healthcare, entertainment, etc., thus benefiting the society as a whole.