QoS Aware for Peer Selection on P2P Streaming Services

Abstract

Internet Protocol-based Television (IPTV) is a mechanism for transporting a stream of video content over a network that uses the IP networking protocol. Compared to the traditional TV system, IPTV is offers much better features and services. However, most of IPTV services are based on a client-server approach that has some limitations and may not be sufficient to support the rapidly increasing demand. So the Peer-to-Peer (P2P) model, with its potential scalability and the ease of operation, has been considered as an alternative solution. Providing the quality video content in P2P IPTV services, it is agreed that to have good national selection algorithm that can improve service quality. Good peers can vastly influence the QoS of the application running on those peers. In this research work, we propose peer selection with QoS aware for P2P IPTV service by evaluating quality of video content delivered using an end-to-end measuring approach. The video time delay is calculated from buffer map of each peer to reflect the target peer quality in the system perspective. Also, in the new joining peer perspective each peer was calculated by paying attention to the path from each peer to the new joining peer. With this algorithm, the real proper and good servers can be discovered to provide a good service to a new joining peer. We have shown that our proposed scheme performs better results in terms of delay time.