Multimedia Data Flow Traffic Classification Using Intelligent Models Based on Traffic Patterns

Nowadays, there is high interest in modeling the type of multimedia traffic with the purpose of estimating the network resources required to guarantee the quality delivered to the user. In this work we propose a multimedia traffic classification model based on patterns that allows us to differentiate the type of traffic by using video streaming and network characteristics as input parameters. We show that there is low correlation between network parameters and the delivered video quality. Because of this, in addition to network parameters, we also add video streaming parameters in order to improve the efficiency of our system. Finally, it should be noted that, based on the objective video quality received by the user, we have extracted traffic patterns that we use to perform the development of the classification model.