The Transmission Rating Scale and its Relation to Subjective Scores

Parametric planning models provide an estimation of the perceived Quality of Experience (QoE) for a certain set of values of relevant network Quality of Service (QoS) conditions. Many of those derive from the E-model, ITU-T standard for network planning in voice transmission applications. The E-model maps the values of different network conditions into a common Transmission Rating scale, which is later used to estimate user quality scores. However, the E-model does not provide a method to obtain direct human judgments on such scale, which has also found not applicable to other applications, such as web browsing or video. In this paper, we propose a way to use the Transmission Rating for general use cases, including video quality. Be re-doing the derivation process which led to the E-model scale from the original Bellcore model, it is possible to apply it to any use case with an underlying normal distribution of subjective scores. We validate this approach by analyzing two video quality experiments which mix different judgement conditions (in particular, screen sizes and resolutions). Finally, we provide general recommendations to develop Transmission Rating scales in future parametric models.