Estimating and Synthesizing QoE Based on QoS Measurement for Improving Multimedia Services on Cellular Networks Using ANN Method

Abstract

A quarter of the world population uses the smartphones to access the Internet and various types of multimedia services on the cellular networks, leading to the importance of focusing on user-centric approach, based on the Quality of Experience (QoE) metric, to measure the business success for the Mobile Network Operator (MNO). Although, the quality of the network can be improved with the help of Quality of Service (QoS), but it does not indicate the user satisfaction. Therefore, it is of vital importance to use the QoE along with the QoS parameters to evaluate and improve the quality of the network. In this paper, we propose the QoE modelling, based on the QoS parameters, by using the Artificial Neural Network (ANN) method to evaluate and synthesize the QoE in the actual environment, with the help of Drive Tests. The relationship between the QoS parameters and Opinion Score (OS) has been analyzed and investigated, prior to the selection of QoS parameters for the creation of QoE model and the process of parameter synthesis. The datasets of QoS parameters and OS have been collected with the help of end devices and the subjective evaluation method from the group of defined users, for each of the following multimedia services, i.e., YouTube, Facebook, Line, and the Web browser, in the real environment. The human behavior can be efficiently learned by using the properties of ANN from the collected datasets, to generate the QoE model along with the estimation of QoE score, instead of using the real humans. The results obtained from the process of parameter synthesis have been used as the main guiding paradigm for improving the performance of the networks in terms of user-centric approach based on the QoS parameters.