An Efficient Source Allocation Approach for QoS Support in P2PTV Systems

Due to the increasing volume of multimedia traffic on the Internet nowadays, the traditional client-server architecture may face scalability and deployment cost problems when the number of subscribers grows large. To tackle these problems, more and more media streaming services, such as IPTV are deployed based on Peer-to-Peer (P2P) architecture. According to the current Internet infrastructure, most of subscribers are behind Network Address Translator (NAT) and using asymmetric access link, so they can not directly share their data with other NATed peers. This situation can easily lead to insufficient upload bandwidth and result in video stuck. Further, the improper selection of peers to download video data may also slow down the video delivery. Both above situations degrade the quality of services of P2PTV system. In this article, we propose approaches to deal with peer grouping and source peer selection, the former is to restrict connection to certain peer groups for improving the share rate of NATed peers, and the latter is to probe partner resource status for optimizing data scheduling. Simulation results show that our proposed approach is able to increase the share rate of NATed peers as well as to reduce video downloading time significantly.