Abstract
P2P networks are a technology able to deliver real time and video-on-demand services over IP networks. Layered video coding techniques are being introduced due to their ability to deliver streams at different scales (temporal, spatial and SNR) that solve the heterogeneity problem. This eases transmission in the case of limited bandwidth, as the devices can pick and decode the minimum bit rate base layer. Existing work examines layered video in client-server scenarios. In contrast, this paper analyzes scalable coding H.264/SVC over P2P networks based on an SNR-temporal Codec. Due to the interdependency between the different SVC layers, issues of reliability and quality of experience arise unless proper measures are taken to protect the base layer. The authors explore the effectiveness of a combination of P2P strategies, for example, hybrid P2P architecture, P2P locality, and P2P redundancy, to assess the viability and benefits of scalable video coding over P2P. The resulting performance is compared with a state-of-the-art P2P TV platform.