A novel frame-type aware scheduling algorithmin mesh-based P2P video streaming

Abstract

Perceptual visual quality guarantee is of vital concern for the P2P streaming systems. One of the challenging issues for assuring overall video quality in mesh-based P2P streaming is to design an effective media chunk scheduler. To the best of our knowledge, there is still a lack of studies on integrating video characteristics into P2P media chunk scheduler. Video packets do not contribute evenly to the video quality at a receiving peer. The lack of more important packets will degrade the received video quality significantly. In this paper, we propose a new scheduling algorithm, which tends to request the most important video packets from senders that can deliver them in a shorter time. We consider two features of the sender peers in order to choose the best supplier for the most important part of a GoP: queuing time and transmission time. Our proposed algorithm takes both peers' features and video characteristic into account simultaneously, which, based on the simulation results, leads to a substantial improvement both in the playback quality and load balance distribution across peers.