An efficient hot channel identification scheme for IPTV channel navigation

Internet protocol television (IPTV), which is rapidly growing in popularity, can provide hundreds of channels; but the very number of channels, and the delay in switching between them due to network bandwidth limitations, can make it difficult for viewers to find the content they want. To address these problems, this paper presents a novel scheme that identifies ´hot´ channels, those likely to be watched in the near future by a target viewer. Then, the scheme actively maintains a list of hot channels, and a second list of cold channels. Channels are allocated to one of the lists based on their recent popularity across all IPTV viewers and the target viewer´s own viewing history. The hot list is searched first during channel navigation, and hot channels are also prefetched in low resolution to support a preview mode; this combination usually reduces zapping time substantially, even if network bandwidth is limited. A prototype architecture of the proposed scheme is developed. Experimental results show that the proposed scheme significantly improves channel seek time compared to adjacency-based prefetching, popularity-based prefetching, popularity-based arrangement, or a simple numerical ordering.