A statistical approach to classify Skype traffic

Authors
nasrin batmani
Nooruddin Parandin
10.52547/jde.2.7.97
Abstract
Abstract- Skype is one of the most powerful and high-quality chat tools that allows its users to use of many services such as: transferring audio, sending messages, video conferencing and audio for free. Skype traffic has a lot of Internet traffic. Hence, Internet service providers need to identify traffic to do the quality of service and network management. On the other hand, Skype developers have been attempting to encrypt Skype traffic because traditional methods like port-based and deep packet inspection can not identify Skype traffic. As a result, we use statistical methods to identify these types of traffic. Hence, in this study, we use unsupervised machine learning methods, which is a statistical method, to separate the various services of Skype. The algorithms used in this work are K-Means, EM and Density-based. The results show that the EM algorithm has better performance than other algorithms. Also, by comparing the proposed strategy with previous work, results indicate that algorithms detect traffic better than other
Keywords
internet traffic
Skype
identify encrypted flow
clustering algorithms
1- Finsterbusch, M., Richter, C., Rocha, E., Muller, J. A., & Hanssgen, K. (2014). A survey of payload-based traffic classification approaches. IEEE Communications Surveys & Tutorials, 16(2), 1135-1156.
2- Korczyński, M., & Duda, A. (2012, June). Classifying service flows in the encrypted skype traffic. In Communications (ICC), 2012 IEEE International Conference on (pp. 1064-1068). IEEE.
3- Yuan, Z., Du, C., Chen, X., Wang, D., & Xue, Y. (2014, February). Skytracer: Towards fine-grained identification for skype traffic via sequence signatures. In Computing, Networking and Communications (ICNC), 2014 International Conference on (pp. 1-5). IEEE.
4- Alshammari, R., & Zincir-Heywood, A. N. (2009, July). Machine learning based encrypted traffic classification: Identifying ssh and skype. In Computational Intelligence for Security and Defense Applications, 2009. CISDA 2009. IEEE Symposium on (pp. 1-8). IEEE.
5- Branch, P. A., Heyde, A., & Armitage, G. J. (2009, June). Rapid identification of Skype traffic flows. In Proceedings of the 18th international workshop on Network and operating systems support for digital audio and video (pp. 91-96). ACM
6- Korczyński, M., & Duda, A. (2014, April). Markov chain fingerprinting to classify encrypted traffic. In Infocom, 2014 Proceedings IEEE (pp. 781-789). IEEE.