Section: New Results
Quality of Experience
Participants : Gerardo Rubino, Adlen Ksentini, Yassine Hadjadj-Aoul, Sofiene Jelassi, Sebastián Basterrech.
We continue the development of the PSQA technology (Pseudo-Subjective Quality Assessment) in the area of Quality of Experience (QoE). PSQA is today a stable technology allowing to build measuring modules capable of quantifying the quality of a video or an audio sequence, as perceived by the user, when received through an IP network. It provides an accurate and efficiently computed evaluation of quality. Accuracy means that PSQA gives values close to those than can be obtained from a panel of human observers, under a controlled subjective testing experiment, following an appropriate standard (which depends on the type of sequence or application). Efficiency means that our measuring tool can work in real time, if necessary. Observe that perceived quality is the main component of QoE. PSQA works by analyzing the networking environment of the communication and some the technical characteristics of the latter. It works without any need to the original sequence (as such, it belongs to the family of no-reference techniques).
It must be pointed out that a PSQA measuring or monitoring module is network dependent and application dependent. Basically, for each specific networking technology, application, service, the module must be built from scratch. But once built, it works automatically and efficiently, allowing if necessary its use in real time.
At the heart of the PSQA approach there is the statistical learning process necessary to develop measuring modules. So far we have been using Random Neural Networks (RNNs) as our learning tool (see  for a general description), but recently, we have started to explore other approaches. For instance, in the last ten years a new computational paradigm was presented under the name of Reservoir Computing (RC)  covering the main limitations in training time for recurrent neural networks while introducing no significant disadvantages. Two RC models have been developed independently and simultaneously under the name of Liquid State Machine (LSM)  and Echo State Networks (ESN)  and constitute today one of the basic paradigms for Recurrent Neural Networks modeling  . The main characteristic of the RC model is that it separates two parts: a static sub-structure called reservoir which involves the use of cycles in order to provide dynamic memory in the network, and a parametric part composed of a function such as a multiple linear regression or a classical single layer network. The reservoir can be seen as a dynamical system that expand the input stream in a space of states. The learning part of the model is the parametric one. In  we propose a new learning tool which merges the capabilities of Random Neural Networks (RNNs) with those of Reservoir Computing Models (RCMs). We keep some of the nice features of RNNs with the ability of RCMs in predicting time series values. Our tool is called Echo State Queueing Network. In the paper, we illustrate its performances in predicting, in particular, Internet traffic. We also worked on the bottleneck of the PSQA building process, from the time consuming point of view, the subjective test sessions. We proposed in  and  new PSQA modules for VoIP and SVC video, respectively. In  , we used PESQ for replacing the subjective test in the training step of PSQA. This module is dedicated to iLBC and Speex codecs. Whereas in  , we used VQM tool to evaluate the SVC video sequences to train PSQA.
In  , a general presentation of our approach in Dionysos was given, together with some guidelines in looking for extensions able to deal with the evaluation of generic applications or services over the Internet.
We presented a tutorial on Quality of Experience in Qest'2012  , based on our past research results in evaluating the perceptual quality in voice or video applications, and on the current work performed in the QuEEN project.
Our perceptual quality work is being extended to investigate the quality of user experience including a large scope that involves human and technology factors. This work is conducted in the context of the Celtic-QuEEN project where a complete QoE monitoring platform is being designed. In Qest'2012  , we presented a tutorial on Quality of Experience based on our past research results in evaluating the perceptual quality in voice or video applications, and on the current work performed in QuEEN.
On the other hand, we continue our study of quality of temporally interrupted VoIP service frequently observed over wireless and data networks. A flagship paper regarding the perception of interruptions in the context of VoIP service is published in  . In  we presented a detailed state-or-the-art in the area.