Abstract
By distinguishing long and short TCP flows, we address in this paper the problem of efficiently computing the characteristics of long flows. Instead of using time consuming off-line flow classification procedures, we investigate how flow characteristics could directly be inferred from traffic measurements by means of digital signal processing techniques. The proposed approach consists of classifying on the fly packets according to their size in order to construct two signals, one associated with short flows and the other with long flows. Since these two signals have intertwined spectral characteristics, we use a blind source separation technique in order to reconstruct the original spectral densities of short and long flow sources. The method is applied to a real traffic trace captured on a link of the France Telecom IP backbone network and proves efficient to recover the characteristics of long and short flows.
This work has been partially supported by the RNRT project Metropolis
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Barakat, C., Thiran, P., Iannaccone, G., Diot, C., Owezarski, P.: A flow-based model for Internet backbone traffic. In: Proc. ACM SIGCOMM Internet Measurement Workshop, Marseille (2002)
Fredj, S.B., Bonald, T., Proutiere, A., Regnie, G., Roberts, J.: Statistical bandwidth sharing: A study of congestion at flow level. In: Proc. ACM Sigcomm. (2001)
Olivier, P., Benameur, N.: Flow level IP traffic characterization. In: Proc. ITC’17, Salvador de Bahia, Brasil (2001)
Papagiannaki, K., Taft, N., Bhattachayya, S., Thiran, P., Salamatian, K., Diot, C.: On the feasibility of identifying elephants in Internet backbone traffic. Technical Report TR01-ATL-110918, Sprint Labs, Sprint ATL (2001)
Pham, D.: Blind separation of instantaneous mixture of sources via the Gaussian mutual information criterion. Technical report, Laboratory of Modeling and Computation, IMAG, C.N.R.S, France (2001)
Pham, D.: Mutual information approach to blind separation for stationnary sources. In: Proc. ICA 1999 Conference, Ausois, pp. 215–220 (1999)
Ben-Azzouna, N., Clérot, F., Fricker, C., Guillemin, F.: A flow-based approach to modeling ADSL traffic on an IP backbone link. Annals of Telecommunications (2004)
Claffy, K., Miller, G., Thompson, K.: The nature of the beast: Recent traffic measurement from an Internet backbone. In: Proc. of Inet. (1998)
Leland, W., Taqqu, M., Willinger, W., Wilson, D.: On the self-similar nature of Ethernet traffic. IEEE/ACM Trans. Net., 1–15 (1994)
Zhang, Z.L., Ribeiro, V., Moon, S., Diot, C.: Small time scaling behavior of Internet backbone traffic: An empirical study. In: Proc. Infocom 2003 (2003)
Ben-Azzouna, N., Guillemin, F.: Analysis of ADSL traffic on an IP backbone link. In: Proc. Globecom 2003, San Francisco, CA (2003)
Wornell, G.: Signal processing with fractals. A wavelet-based appoach. Prentice Hall Signal Processing Series (1995)
Cover, T., Thomas, J.: Elements of Information Theory. Wiley-Interscience Publication,Hoboken (1991)
Castanié, F.F., ed.: Analyse spectrale. Hermès Science Publications,China (2003)
Priestley, M.: Spectral Analysis and Time Series. Academic Press, London (1981)
Kunt, M.: Traitement Numérique des Signaux. Presses Polytechniques Romandes, Lausanne (1984)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Saddi, W., Ben Azzouna, N., Guillemin, F. (2004). IP Traffic Classification via Blind Source Separation Based on Jacobi Algorithm. In: Freire, M.M., Chemouil, P., Lorenz, P., Gravey, A. (eds) Universal Multiservice Networks. ECUMN 2004. Lecture Notes in Computer Science, vol 3262. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-30197-4_29
Download citation
DOI: https://doi.org/10.1007/978-3-540-30197-4_29
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-23551-4
Online ISBN: 978-3-540-30197-4
eBook Packages: Springer Book Archive