By Amit Eshet, Mark Lutsker, Uzi Cohen, ARRIS
The IP Video service offered by MSOs is about to enter its 2nd phase, extending from the 2nd screens to the big screen. With this transition the expectations and requirements for QoE will be going up to be at par with the legacy video QAM service. IP Video introduces new challenges when it comes to QoE monitoring. One of the most important paradigm shifts is an extremely wide range of screens and consumption habits: with screens that are few inches in size to ultra HD TVs with 80” screen; with a laid back to fully engaged experience. To make things more interesting, in some cases the format of the content itself would dramatically vary (sub-VGA to Ultra HD) while in others the exact same content, say HD, may be viewed both on a tablet and on a big-screen TV. Other critical game changers when it comes to IP video QoE monitoring are WiFi, OTT delivery of video over best effort networks which in some cases are not even owned by the operator (e.g., OTT, off-net), and of course the fact that the decoding device itself may be CE with, at best, a limited ability of the operator to control and guarantee QoE. And of course, on top of all this, operators are rightfully looking for a single QoE Monitoring solution applicable to all screens and all use cases. In this paper we will start by discussing the differences between QoE and QoS and between QoE and video quality. We will then compare different methodologies for video quality and QoE monitoring, including full-reference vs. reduced-reference, vs. no-reference; compressed vs. pixel domain; statistical vs. exhaustive. We will conclude with a review of alternatives for embedding QoE probes in the end-to-end IP Video architecture and their ability