Title: Quality of experience characterization and provisioning in mobile cellular networks

Author: Eirini Liotou

Supervisor(s) and Committee member(s): Lazaros Merakos, Professor, National and Kapodistrian University of Athens Efstathios Hadjiefthymiades, Associate Professor, National and Kapodistrian University of Athens Athanasia Alonistioti, Assistant Professor, National and Kapodistrian University of Athens Georgios Polyzos, Professor, Athens University of Economics and Business Dimitrios Varoutas, Associate Professor, National and Kapodistrian University of Athens Christos Xenakis, Associate Professor, University of Piraeus Alexandros Kaloxylos, Assistant Professor, University of Peloponnese

URL: http://thesis.ekt.gr/thesisBookReader/id/44514#page/1/mode/2up

DOI: https://www.researchgate.net/profile/Eirini_Liotou


phd_thesis_photoTraditionally, previous generations of mobile cellular networks have been designed with Quality of Service (QoS) criteria in mind, so that they manage to meet specific service requirements. Quality of Experience (QoE) has, however, recently emerged as a concept, disrupting the design of future network generations by giving clear emphasis on the actually achieved user experience. The emergence of the QoE concept has been a result of the inevitable strong transition that the Telecom industry is currently experiencing from system-centric networks to more user-centric solutions and objectives. Mobile network operators, service providers, application developers, as well as other stakeholders involved in the service provisioning chain have been attracted by the opportunities that the integration of the QoE concept could bring to their business; indeed, the provisioned QoE constitutes a determining factor of differentiation among different stakeholders, a tendency which is expected to become even more intense in the years to come. Motivated by this boost towards user-centricity, the objective of the research conducted in this thesis is to explore the challenges and opportunities that arise in modern mobile cellular networks when QoE is considered. Such opportunities concern, first of all, the possibility to comprehend the QoE that a provider achieves when provisioning a service. This can be enabled by the implementation and integration of QoE assessment methods into the real-time operation of a network. Then, the next step is the exploitation of collected QoE-related intelligence in order to re-examine existing network-layer mechanisms (e.g., radio scheduling), or application-layer mechanisms (e.g., video streaming), as well as propose novel cross-layer approaches towards ameliorating the achieved QoE. Moreover, the opportunity emerges to propose novel algorithms that stem from the inherent idiosyncrasies of QoE, such as the non-linear impact of QoS-related parameters on QoE, as a way to further enhance the users’ QoE. In this direction, throughout this thesis, QoE estimation models and metrics are explored and exploited in order to quantify QoE and thus, to improve existing mechanisms of mobile cellular networks. The core of this thesis is the proposal of a QoE provisioning cycle that allows the control, monitoring (i.e., modeling) and management of QoE in a cellular network. Each one of these functions is further analyzed, while emphasis is given on the modeling and management operations. In terms of modeling, QoE assessment methods and QoE-related performance indicators are described and classified. Parametric quality estimation is identified as the most appealing type of QoE estimation in mobile cellular networks, thus, it is thoroughly described for widely used types of services, such as Voice over IP (VoIP) and video streaming. In terms of QoE management, novel QoE-aware mechanisms that demonstrate QoE improvements for the users are proposed, namely: a) a QoE-driven Device-to-Device (D2D) communication management scheme that enhances end-user QoE, b) a “consistent” radio scheduling algorithm that improves the end-user QoE by mitigating throughput fluctuations, and c) a context-aware HTTP Adaptive Streaming (HAS) mechanism that successfully mitigates stallings (i.e., video freezing events) in the context of bandwidth-challenging scenarios. Moreover, a programmable QoE-SDN APP into the Software-Defined Networking (SDN) architecture is introduced, which enables network feedback exposure from mobile network operators to video service providers, revealing QoE benefits for the customers of video providers and bandwidth savings for the network operators. Overall, this thesis promotes the uniting of the domain of QoE with the domain of mobile communications, as well as the collaboration of mutual-interest between mobile network operators (network layer) and service providers (application layer), presenting the high potential from such approaches for all involved stakeholders.

Green, Adaptive and Intelligent Networking Group

URL: http://gain.di.uoa.gr/%20

Bookmark the permalink.