Personalized service degradation on ott applications

Abstract

OTT applications are known by their large consumption of network resources for their correct operation. In the mobile networks scope, where operators offer users data plans with limited consumption, service degradation is a measure implemented in a generalized way in order to apply limits to the amount of information that can be transferred by the users over a period of time when a user exceeds his/her established consumption limit, in order to save resources and ensure the correct performance of the network. Therefore, the behavior and preferences presented by the user in the consumption of OTT applications are never considered when this mechanism is implemented.

Considering the previous background, this research project aims personalizing the service degradation policies in mobile Internet based on the OTT applications consumption of the users.

In order to accomplish this objective an identification of the QoS parameters related to the service degradation and the parameters that describe the user consumption behavior will be identified. A selection of classification techniques aimed at characterizing the users OTT consumption behavior will be performed and a set of personalized service degradation policies considering the user behavior will be proposed.

Some of the obtained results in the development of this research project are: a group of datasets that enable the classification of the users based on their consumption behavior; a set of parameters that must be considered in order to propose QoS policies and to characterize the user consumption behavior; the implementation of a classifier that enables the identification of the consumption trends of OTT applications by users and the proposal of personalized service degradation policies considering the consumption trends of OTT applications of users.