Parameter estimation for sums of correlated gamma random variables. Application to anomaly detection in Internet Traffic

Chatelain Florent 1 Pierre Borgnat 2 Jean-Yves Tourneret 1 Patrice Abry 2
1 IRIT - Institut de recherche en informatique de Toulouse
2 Phys-ENS - Laboratoire de Physique de l'ENS Lyon
Abstract : A new family of distributions, constructed by summing correlated gamma random variables, is studied. First, a simple closed-form expression for their density is derived. Second, the three parameters characterizing such a density are estimated by using the maximum likelihood (ML) principle. Numerical simulation are conducted to compare the performance of the ML estimator against those of the conventional estimator of moments. Finally, a multiresolution multivariate gamma based modeling of Internet traffic illustrate the potential interest of the proposed distributions for the detection of anomalies. Aggregated times series of IP packet counts are split into adjacent non overlapping time blocks. The distribution of these series are modeled by the proposed multivariate gamma based distributions, over a collection of different aggregation levels. The anomaly detection strategy is based on tracking changes along time of the corresponding multiresolution parameters.
Keywords : Multivariate gamma distributions Maximum likelihood estimator Internet traffic Anomaly detection
Document type :
Conference papers
Complete list of metadatas
https://hal-ens-lyon.archives-ouvertes.fr/ensl-00290724
Contributor : Pierre Borgnat <>
Submitted on : Thursday, June 26, 2008 - 11:49:33 AM
Last modification on : Thursday, October 17, 2019 - 8:52:52 AM
Long-term archiving on : Friday, May 28, 2010 - 10:52:11 PM

File

08icassp_chatelain.pdf
Publisher files allowed on an open archive

Identifiers

  • HAL Id : ensl-00290724, version 1

Collections

ENS-LYON | UNIV-TLSE2 | SMS | UNIV-LYON1 | UNIV-TLSE3 | UT1-CAPITOLE

Citation

Chatelain Florent, Pierre Borgnat, Jean-Yves Tourneret, Patrice Abry. Parameter estimation for sums of correlated gamma random variables. Application to anomaly detection in Internet Traffic. IEEE Int. Conf. on Acoust., Speech and Signal Proc. ICASSP-08, IEEE, Mar 2008, Las Vegas, United States. ⟨ensl-00290724⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

279

Files downloads

1358

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD