Section: New Results

Quality of Experience

Participants : Yassine Hadjadj-Aoul, Adlen Ksentini, Gerardo Rubino, Bruno Sericola, Pantelis Frangoudis, César Viho, Quang Pham Tran Anh.

PSQA. We continue the development of the PSQA technology (Pseudo-Subjective Quality Assessment) in the area of Quality of Experience (QoE). PSQA is today a mature 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 that 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. Observe that perceived quality is, in general, the main component of QoE when the application or service involves video and audio, or voice. 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, and for any application or service, the module must be built from scratch. But once built, it works automatically and efficiently, allowing if necessary its use in real time, typically for controlling purposes.

QoE and SLA. On the applications side, we focused this year on using QoE estimates to drive service/application-level decisions. As a first use case, we proposed a multi-objective optimization framework for the problem of optimally selecting among a set of available hosting and network connectivity Service-Level Agreements (SLAs) for the migration of enterprise communication services (such as teleconferencing) to the Cloud [59] . Our framework captures the tradeoff between user experience and deployment cost, and offers a service provider the opportunity to weight these two conflicting criteria based on its preferences. Our approach is generic and can be applied to various application settings by appropriately selecting application-specific user experience models. For example, for enterprise voice teleconferencing we used the E-model for estimating user experience under a specific selection of hosting and network SLAs and a specific amount of resources (virtual machines) to deploy.

QoE and collaborative projects. We then considered QoE-aware content delivery, targeting in particular an environment where web and multimedia content is disseminated by over-the-top (OTT) providers, but assuming a level of cooperation between the content provider and the ISP (a trend which has started to become commonplace) [46] . We built on the outcome of our prior work (P.A. Frangoudis, A. Ksentini, Y. Hadjadj-Aoul, and G. Boime, “PTPv2-based network load estimation,” Proc. IEEE ISPCS 2013. (This work was carried out in the context of the FUI project IPChronos, see  7.10 .)), where we designed and implemented a network load estimation methodology and tool which operates by observing the delay behavior of the Precision Time Protocol (PTP) for network clock synchronization. After quantitatively establishing the link between network load and user experience, we proposed an architecture for OTT content delivery where user requests are redirected to the data centers expected to offer optimal QoE, taking into account, among others, information about network load in the media path offered by our load estimation service (LES) in real time. In the same context, we developed a demonstrator where the LES is integrated as an additional network probe with the QoE monitoring architecture developed in the Celtic QuEEN project (see  8.2.1.1 ). Using a simple video QoE model which takes into account network load and video information (quality/resolution, bitrate), we implemented (Our video adaptation scheme is implemented in the VLC open-source media player.) an adaptation scheme for DASH video delivery which switches among video qualities based on QoE estimates received by the QuEEN software agent.

QoE and PTPv2. In [46] , we make the case for an alternative use of the PTPv2 protocol: Adopting a learning approach, we observe its delay behavior during the protocol message exchange, derive models of its dependence on network load and build a real-time load estimation service. Then, as an application scenario of this service, we turn our attention to the provision of Over-the-Top (OTT) services. In such an environment, and assuming a level of cooperation between the ISP and the OTT provider, we demonstrate how our service can be used for estimating the QoE for web applications. To this end, we establish quantitatively the link between network load and user experience using a state-of-the-art web QoE monitoring framework, and show how our PTPv2-based load estimation scheme can integrated in an OTT service architecture and be utilized for load-aware, QoE-optimized content delivery decisions.

QoE and reneging. We consider in [45] an important Quality of Experience (QoE) indicator in mobile networks that is reneging of users due to impatience. We specifically consider a cell under heavy load conditions and compute the reneging probability by using a fluid limit analysis. By solving the fixed point equation, we obtain a new QoE perturbation metric quantifying the impact of reneging on the performance of the system. This metric is then used to devise a new pricing scheme accounting for reneging. We specifically propose several flavors of this pricing around the idea of having a flat rate for accessing the network and an elastic price related to the level of QoE perturbation induced by the communications.

QoE-aware OLSR for Video Streaming over Wireless Multihop Networks. Multi-hop environments can impact significantly ad-hoc network performance. In [57] , we propose a routing algorithm based on optimized link state routing (OLSR), aimed at guaranteeing the quality of experience (QoE) of users in these types of networks. PSQA (see above in this same section) is used to estimate a mean opinion score (MOS), and then this MOS value is exploited by the source for selecting the appropriate path in the network. Moreover, an event- triggered based on the MOS value is used to provide more relevant information in selecting the best path by the source. The performance of this proposed mechanism was validated through intensive simulation under different scenarios. The results in [57] show that the proposed scheme outperforms other OLSR-based routing protocols particularly in a heavy load and high mobility scenario.

QoE-Aware Routing for Video Streaming over VANETs. In-vehicle multimedia applications are gaining interest since recent years. However, the high loss rate caused by high mobility in vehicular networks (VANETs) imposes several challenges in multimedia transmission. Moreover, in the context of multimedia, the quality of service (QoS)-based approaches assess the quality of streaming services through network-oriented metrics while the concept of quality of experience (QoE) is built upon the perception of users. In [58] , a QoE-based routing protocol for video streaming over VANETs is proposed. By taking the mean opinion score (MOS) into account for path selection, good performance levels can be achieved, as shown by our simulation results.