A Fast SVC-Based Channel-Recommendation System for an IPTV on a Cloud and P2P Hybrid Platform

摘要

In this paper, we propose a fast scalable video coding (SVC (-based channel-recommendation system for IPTV on a Cloud and peer-to-peer (P2P) hybrid platform. When a user switches channels, the system redirects the client to the cloud network and delivers the base layer of SVC streams to the client. The system provides a multichannel preview window with a small resolution and a fast channel-switching mechanism without additional traffic. After a user has selected a channel, the system redirects the client to the P2P network and sends the necessary enhancement layer streams, so that the client can receive high-quality video. Because of the fact that the recommendation system is known as an effective approach for enhancing the efficiency of channel previews, we propose a novel recommendation system based on the feedback loser tree (FLT) algorithm. The FLT algorithm can be trained by the user's historical log, and attempt to find the user's preferred channels quickly. Our experimental results indicate that the proposed platform can obtain a higher peak signal to noise ratio quality than the original P2P networks, and the proposed system can help users find their favorite channels in only 2 to 5 switching pages. The performance of the proposed system is ～75% better than that of the high-performance multichannel preview system.