Grigoriou, Elisavet; Atzori, Luigi; Saoulidis, Theocharis; Pilloni, Virginia; Chatzimisios, Periklis
An agent-based QoE monitoring strategy for LTE networks (Inproceedings)
In: ICC2018, 2018.
The new generation of Long Term Evolution (LTE) networks provides ubiquitous broadband access to mobile devices matching land communications in quality and speed. However, to optimize network resource usage in a dynamic environment network operators need models and strategies to constantly assess and manage the end-user's Quality of Experience (QoE). Given the importance of these activities, in the current paper, we focus on quality monitoring and the usage of QoE-agents in an LTE-Advanced Pro network. Specifically, we identify the location and the operation of the QoE-Agents based on the accuracy of the measurements and the load in the network by considering the frequency of the measurements and the running applications. Emulations have been also carried out to evaluate two considered scenarios with different network conditions as well as with different quality sampling rates and application configurations. The preliminary results have shown that the proposed strategy results in acceptable errors from our measurements, low CPU utilization and acceptable memory utilization.
Grigoriou, Elisavet; Atzori, Luigi; Pilloni, Virginia
In: Globecom2017, 2017.
In this paper, we illustrate a Software Defined Network (SDN)-based architecture for Quality of Experience (QoE) management. This approach solves two of the major problems of current networking technologies which are related to the limitations in scalability and flexibility. Its 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 SDN control functions. As a result the QoE monitoring and management is deployed at the application layer on top of the controller. In order to evaluate the proposed framework and architecture, a platform has been developed, which is called QoE-MoMa (QoE-Monitoring and Management) platform, making use of the Opendaylight solution and Mininet emulation environment. To evaluate QoE-MoMa, we focused on the video streaming service, whose final quality has been evaluated using the estimated MOS (eMOS) model that mostly considers rebuffering events, duration of the rebuffering, switch quality rates, video resolution, and quantization parameter. The results show the efficiency of the proposed approach observing that higher QoE level is achieved if we consider application and network parameters. In conclusion, we consider that QoE-MoMa is useful as a QoE monitoring and management tool for a variety of services and can be deployed on a real network conveniently.