Parameter estimation for sums of correlated gamma random variables. Application to anomaly detection in Internet Traffic - ENS de Lyon - École normale supérieure de Lyon Accéder directement au contenu
Communication Dans Un Congrès Année : 2008

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

Résumé

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.
Fichier principal
Vignette du fichier
08icassp_chatelain.pdf (161.49 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

ensl-00290724 , version 1 (26-06-2008)

Identifiants

  • HAL Id : ensl-00290724 , version 1

Citer

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⟩
169 Consultations
1172 Téléchargements

Partager

Gmail Facebook X LinkedIn More