Eigen-Value Based Multivariate Selfsimilarity Analysis in Internet Traffic: A Case Study
Internet traffic modeling and analysis is critical for network design and for cybersecurity. Internet time series are well characterized by scalefree temporal dynamics. However, scalefree analysis remained so far univariate, applied independently to directional counts of either bytes or packets while chal-lenges in cybersecurity naturally call for multivariate analysis. Elaborating on recent theoretical developments on eigenvalue-based multivariate self-similarity analysis, this work provides evidence, for the first time, of multivariate self-similarity in 17 years of Internet traffic data from the MAWI repository. It discusses the potential use of multivariate self-similarity for regular background traffic characterization and anomaly detection.
Authors
- Herwig Wendt (ORCID: https://orcid.org/0000-0001-6930-3083)
- Patrice Abry (ORCID: https://orcid.org/0000-0002-7096-8290)
- Kenjiro Cho (ORCID: https://orcid.org/0000-0001-7575-186X)
- Romain Fontugne (ORCID: https://orcid.org/0000-0002-9816-5625)
- Gustavo Didier (ORCID: https://orcid.org/0000-0003-1363-5868)
- Kensuke Fukuda (ORCID: https://orcid.org/0000-0001-8372-2807)
Institutions
- École Normale Supérieure de Lyon (FR)
- Tulane University (US)
- Centre National de la Recherche Scientifique (FR)
- Université Toulouse III - Paul Sabatier (FR)
- National Institute of Informatics (JP)
- Université Toulouse-I-Capitole (FR)
- Laboratoire de Physique de l'ENS de Lyon (FR)
- Institut de Recherche en Informatique de Toulouse (FR)
- IIJ Research Laboratory (JP)
- Université Toulouse - Jean Jaurès (FR)
- Institut Polytechnique de Bordeaux (FR)
Publication Details
- Published
- 2025-09-08
- DOI
- https://doi.org/10.23919/eusipco63237.2025.11226771
- Primary Topic
- Network Traffic and Congestion Control
- Type
- article
- Field-Weighted Citation Impact
- 0.00