Development of Fuzzy Relationships between Different Quality of Experience Parameters

Abstract

User psycho-acoustic and visual perception of telecommunication service quality depends on a wide palette of objective and subjective parameters. This is especially relevant in the process of evaluation, assessment or prediction of user Quality of Experience (QoE) for specific service, because these processes require the revelation of the correlations between different parameters. Using the User Datagram Protocol (UDP) based video streaming service as an example, in this paper we investigate the impact of three objective parameters (the packet loss rate, the number and the duration of packet loss occurrences) on user perception by conducting subjective quality tests. We then use Fuzzy C-Means (FCM) clustering approach to cluster test subjects' ratings as well as to define degrees of membership of each rating to specific fuzzy cluster. Based on the obtained degrees of membership, the fuzzy relationships between different objective and subjective parameters are approximated using the normal distributions.