An early recognition algorithm for BitTorrent traffic based on improved K-means

Abstract

In response to the deficiencies of BitTorrent, the concept of density radius was proposed, and the distance from the maximum point of radius density to cluster center as a cluster radius was taken to solve the too large cluster radius resulted from the discrete points and to reduce the false positive rate of early recognition algorithms. Simulation results show that in the actual network environment, the improved algorithm, compared with K-means, will reduce the false positive rate of early identification algorithm from 6.3% to 0.9% and has a higher operational efficiency.

Keywords

traffic identification / early recognition algorithm / cluster radius / false positive/negative rate

References

[1] SenS., WangJ.. Analyzing peer-to-peer trafficacross large networks [J]. IEEE/ACM Transactions on Networking, 2004, 12(2): 219-232

[2] Ernesto. Bittorrent continues to dominate internet traffic [EB/OL]. 2009-04-20. http://torrentfreak.com/bittorrent-dominates-internet-traffic-070901/.

[3] KaragiannisT., BroidoA., BrownleeN., ClaffyK., FaloutsosM.. Is P2P dying or just hiding? [C]. Proceeding of Global Telecommunications Conference 2004, 2004, New York, ACM Press: 1532-1538

[4] LiJ.-t., JiangY.-ling.. Survey of P2P traffic identification and engineering technology [J]. Telecommunications Science, 2005, 21(3): 57-61

[5] MadhukarA., WilliamsonC.. A longitudinal study of P2P traffic classification [C]. Proceedings of the International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, 2006, Heidelberg, Springer: 179-188

[6] SenS., SpatscheckO., WangD.. Accurate, scalable in-network identification of P2P traffic using application signatures [C]. Proceedings of the 13th International Conference on World Wide Web, 2004, New York, ACM Press: 512-521

[7] LiuG., FangB.-x., HuM.-z., ZhangH.-li.. Capture method and self-similarity assess of bittorrent traffic [J]. Application Research of Computers, 2006, 23(5): 205-206

[8] DuL., YuJ.-x., WangX.-jing.. A coarse-grained differentiated routing algorithm inmulti-protocol label switching traffic engineering [J]. Journal of Central South University: Science and Technology, 2010, 17(6): 1258-1263

[9] ZanderS., NguyenT., ArmitageG.. Automated traffic classification and application identification using machine learning [C]. Proceedings of the IEEE Conference on Local Computer Networks 30th Anniversary, 2005, Sydney, IEEE Society: 250-257

[10] EkmanJ., ArlittM., MahantiA.. Traffic classification using clustering algorithms [C]. Proceedings of the Special Interest Group on Data Communication, 2006, New York, ACM Press: 281-286

[11] KaragiannisT., PapagiannakiD., FaloutsosM.. Blinc: Multilevel traffic classification in the dark [C]. Proceedings of the 2005 SIGCOMM Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, 2005, New York, ACM Press: 229-240

[12] McgregorA., HallM., LorierP., BrunskillJ.. Flow clustering using machine learning techniques [C]. Proceedings of the Personalised Ambient Monitoring, 2005, Heidelberg, Springer: 205-214

[13] JUNIOR G P S, MAIA J E B, HOLANDA R, DE SOUSA J N. P2P Traffic identification using cluster analysis [C]// Proceedings of the International Symposium on Global Information Infrastructure. New York, 2007: 128–133.

[14] LuG., ZhangH.-L., YeLin.. P2P traffic identification [J]. Journal of Software, 2011, 22(6): 1281-1298

[15] BernailleL., TeixeiraR.. Early recognition of encrypted applications [C]. Proceedings of the International Symposium on Passive and Active Measurement, 2007, Heidelberg, Springer: 165-175

[16] BernailleL., TeixeiraR., SalamatianK.. Early application identification [C]. Proceedings of the 2006 ACM CoNEXT Conference, 2006, New York, ACM Press: 70-82

[17] BernailleL., TeixeiraR., AkodkenouI., SouleA., SalamatianK.. Traffic classification on the fly [C]. Proceedings of the Special Interest Group on Data Communication, 2006, New York, ACM Press: 23-26