PERFORMANCE ANALYSIS OF FORWARD ERROR CORRECTION AIDED P2P LIVE-VIDEO STREAMING

Abstract

Peer-2-Peer (P2P) technologies are rapidly advancing and have already proven successful in the form of various bandwidth demanding applications such as file-transfer and video streaming. The major advantage of the P2P technologies is its built-in scalability feature, relative to traditional client- server architecture. Therefore, keeping in view the large number of users involved in live video-streaming, P2P technology can be considered as a valuable tool aiming to broadcast live-video contents to a large number of users over the internet. Although, there is lot of research going on in analyzing the performance of P2P video streaming in heterogeneous networking scenarios, but to the best of our knowledge no work is done in analyzing the impact of Forward Error Correction (FEC) on the performance of P2P live-video streaming. In this paper we analyze the performance and characteristics of P2P live-video streaming on end-user quality of experience while employing FEC on the coded video packets. Additionally, in-order to carry out the performance analysis with reference to non-error protected benchmarker system; we utilized a realistic realtime French distributed Grid facility known as Grid'5000. More specifically, the performance of the system is evaluated while considering various realistic P2P video streaming constraints, such as limited playout delay, limited upload bandwidth and video rates. We concluded that the FEC will result in noticeable performance improvement in a scenario where each peer is connected with limited number of other peers that have the required contents for download. In such a situation the portion of the stream corrupted due to channel errors may not be recovered with the help of other accessible peers. However, if each peer has more connected peers with various alternative options to download the corrupted contents, then the employment of FEC may not have very noticeable advantage on the performance of the P2P video streaming system.