Streaming Video Identification for Caching and Sharing

Abstract

Video streaming is the major network traffic in the present days. Thus understanding the video traffic and optimizing the video traffic can help controlling the bandwidth utilization. In this paper, we present a method to identify videos of common interest so that bandwidth-saving methods can be developed. The paper introduces a way to recreate the initial part of the video, from which frames are extracted. These frames that come from various different video traces are compared to finalize whether two different traces of video traffic belong to the same video content or not. The analyzed result is then utilized by the router to manage the caching and sharing of these contents for multiple users in a LAN. Further, multicast is proposed to be used inside the LAN.