P000081R1
OPTIMIZING PATTERN-PUSH SCHEDULING FOR PEER-TO-PEER LIVE STREAMING
*Yucai Feng (National Engineering Research Center of Digital Life, State-Province Joint Laboratory of Digital Home Interactive Applications, School of Information Science & Technology, Sun Yat-sen University)
Guifeng Zheng (School of Software, Sun Yat-sen University)
Xiaonan Luo (National Engineering Research Center of Digital Life, State-Province Joint Laboratory of Digital Home Interactive Applications, School of Information Science & Technology)
Zheqi Li (Sun Yat-sen University)
In peer-to-peer (P2P) live streaming, packets lose inevitably in the data transmission because of peer failure or transmission error. This paper proposes a simple and efficient P2P streaming scheduling algorithm which focuses on lost packet repairing. We design a bitmap to record the information of packets that are not in request queue. When packet loses, peer requests for it referring to the bitmap. We get the bitmap through pattern and packet ID and transmit it on idle bandwidth. Simulation results with NS2 show that the algorithm reduces efficiently packet loss and packet arrival delay, and increases bandwidth utilization as compared with other typical scheduling algorithms.