Personalized Service Degradation Policies on OTT Applications Based on the Consumption Behavior of Users

Abstract

The proliferation of IP-based telecommunication networks has facilitated the decoupling of application and network layers. This kind of systems allows that Over the Top (OTT) providers deliver their content and applications directly to end users, but at the same time, the OTT applications have generated a growing impact on mobile data traffic and data revenues. In the mobile network’s scope, where the Telcos offer users data plans with limited consumption, service degradation is a measure implemented in a generalized way to apply limits to the amount of data that can be transferred by the users over a period. Currently, when a user exceeds his/her established consumption limit, the Telcos, to save resources and ensure the correct performance of the network, restrict the bandwidth according to user consumption. The vast majority of approaches have not considered the consumption behavior of users to propose a set of personalized service degradation policies that benefit the Telcos but take into consideration the users’ behavior. This paper proposes personalized service degradation policies, from the identification of different OTT services applying statistical analysis and deep packet inspection, and a classification of users, according to their consumption behavior and machine learning algorithms.