Design of cross-layer model for QoE enhancement in P2P live applications

Recently, IPTV service has attracted significant attentions to both of the ISPs and normal users,and the quality of experience (QoE) from the users point of view is the most important evaluation for IPTV service provider. A reasonable peer selection mechanism of P2P based IPTV service can provide the end-users with a better QoE in a limited network resource environment. Therefore, the cross-layer parameters relationships modeling exhibits its significance for peer choosing decision. In order to gain insights into tree-push P2P IPTV systems, we have undertaken an in-depth measurement study of application layer multicast (ALM) tree structure. Great quantity of packet traces for cross-layer metrics in various measurement scenarios, including both CERNET and NSFCnet, are collected and analyzed based on which a novel peer selection algorithm is expressed. Finally, we figure out the priority of important P2P application factors for the development and deployment of future P2P IPTV systems.