Periodic Reporting for period 2 - QoE-Net (innovative Quality Of Experience maNagement in Emerging mulTimedia services)
Berichtszeitraum: 2017-01-01 bis 2018-12-31
Based on these considerations, the QoE-Net project has built and has managed an inter-disciplinary training network in the core disciplines involved in the research and development of Quality of Experience models, optimization approaches, control and management schemes for new and emerging multimedia services, with emphasis on the following: mobile gaming, social TV and web services. Indeed, the QoE, unlike the traditional QoS (Quality of Service) concept, reflects directly to an end-user’s perceived experience for a service, and thus requires knowledge from fields such as: psychology, visual and performing arts, communications and networking, signal processing and media technology, as well as customer behaviour research and business studies; it can consequently only be addressed with a consortium of high-quality academic and research partners, with interlink with industrial partners and standardization bodies, as the one of QoE-Net.
As the project itself is a training network, research training is a core element of the activities. First of all, ESRs have been trained through participation in tutorials and seminars. For this purpose, Online Short Courses are organized for the ESRs, which are coordinated by the ESRs themselves. A total of 17 of these courses have been delivered via an online video conferencing platform. The topics covered by these courses include standardization, perception, quality assessment, QoE-aware solutions, display technologies, video coding, usability, business aspects, etc. The expertise of the ESRs is also broadened by transferable skills training events. A total of 15 events have taken place in difference European countries. These events focused on the development of scientific paper writing skills and oral presentation skills – which are essential for the dissemination of research results – standardization, patenting, project proposal writing and business planning, seminars and a summer school on QoE modelling and management, training ESRs in interdisciplinary methods, and others.
• It often happens that QoE models used by service providers for monitoring purposes rely on several influencing factors that are not at the hands of a single provider. In particular, this scenario applies to the over-the-top (OTT) and Internet service providers (ISPs), which act with complementary roles in service delivery over the Internet. To deal with this issue, the project has highlighted the importance of a possible OTT-ISP collaboration for a joint service management in terms of technical and economic aspects and proposed a potential collaboration and information exchange among them for an improvement of the final quality perception.
• The application of Software Defined Network (SDN) and Network Function Virtualization (NFV) technologies for QoE management has been studied during project, including the case of 5G networks. The relevant advantage is the exploitation of the virtualization features of the network nodes and devices to flexibly deploy monitoring and control functions in the different points of the network according to the varying load of the running applications and status of the network.
• Adaptive video streaming services such as those provided by YouTube, Netflix and Twitch have increased exponentially in recent years. These services are expected to dominate the consumer Internet traffic in the near future. It is important and crucial to have a comprehensive understanding on how video streaming quality (or QoE) is affected by application and networking parameters, and how human perception system reacts to a variety of video adaptation mechanisms or impairments (e.g. resolution switching). A systematic study and analysis on QoE assessment for video streaming services have been carried out, and novel QoE management schemes to enhance QoE for video streaming services have been proposed and developed in this project. Other than publications, some contributions to ITU-T Study Group 12, Question 14 have been made in order to develop a standardized quality prediction model for adaptive video streaming which reached the consent by November 2017 and became an ITU-T recommendation, named P.1203. Three patent applications for Deutsche Telekom AG have been also filed together with TU Berlin and TU Ilmenau.
• With respect to gaming service, the influence factors that have an impact on gaming QoE (including mobile gaming) have been identified, listed and proposed to ITU-T SG 12 that was finalized in the ITU meeting in September, 2017. Moreover, the proposal of subjective evaluation methods for gaming quality reached the consent at the ITU-T Study Group 12 in May 2018 and became an ITU-T recommendation. The efforts w.r.t. mobile gaming were not only limited to these two recommendation documents but also includes several publications and some other activities in order to identify the impact of influencing factors on quality.