Share by e-mail

Return to the home page of Tomasz Bujlow


A Method for Evaluation of Quality of Service in Computer Networks

Tomasz Bujlow, Sara Ligaard Nørgaard Hald, Tahir Riaz, and Jens Myrup Pedersen

ICACT Transactions on the Advanced Communications Technology (ICACT-TACT), ISSN 2288-0003 (Online), vol. 1, no. 2, pp. 17 - 25, Global IT Research Institute (GiRI), July 2012.

  Download this publication in PDF (author's version)


Abstract

Monitoring of the Quality of Service (QoS) in high-speed Internet infrastructures is a challenging task. However, precise assessments must take into account the fact that the requirements for the given quality level are service-dependent. The backbone QoS monitoring and analysis requires processing of large amounts of data and the knowledge about the kinds of applications, which generate the traffic. To overcome the drawbacks of existing methods for traffic classification, we proposed and evaluated a centralized solution based on the C5.0 Machine Learning Algorithm (MLA) and decision rules. The first task was to collect and to provide to C5.0 high-quality training data divided into groups, which correspond to different types of applications. It was found that the currently existing means of collecting data (classification by ports, Deep Packet Inspection, statistical classification, public data sources) are not sufficient and they do not comply with the required standards. We developed a new system to collect the training data, in which the major role is performed by volunteers. Client applications installed on volunteers' computers collect the detailed data about each flow passing through the network interface, together with the application name taken from the description of system sockets. This paper proposes a new method for measuring the level of Quality of Service in broadband networks. It is based on our Volunteer-Based System to collect the training data, Machine Learning Algorithms to generate the classification rules and the application-specific rules for assessing the QoS level. We combine both passive and active monitoring technologies. The paper evaluates different possibilities of the implementation, presents the current implementation of the particular parts of the system, their initial runs and the obtained results, highlighting parts relevant from the QoS point of view.


Return to the home page of Tomasz Bujlow