Towards Systematic Traffic Annotation

Romain Fontugne (1, 2) , Pierre Borgnat (3) , Patrice Abry (3) , Kensuke Fukuda (1)

1 NII - National Institute of Informatics
2 The Graduate University for Advanced Studies
3 Phys-ENS - Laboratoire de Physique de l'ENS Lyon

Maintaining Internet network resources available and se- cured is an unmet challenge. Hence, tracffic classication and anomaly detection received much attention in the last few years, and several algorithms have been proposed for backbone traces. However, the evaluation of these methods usually lacks rigor, leading to hasty conclusions. A systematic approach permits to maintain updated database in which recent traces are regularly added, and labels are improved with upcoming algorithms. In this paper we discuss the difficulties faced in comparing events provided by distinct algorithms, and propose a methodology to achieve it.

Mots clés: Traffic classification; Anomaly detection; Benchmarking

Domaines: Réseaux et télécommunications [cs.NI]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Romain Fontugne, Pierre Borgnat, Patrice Abry, Kensuke Fukuda. Towards Systematic Traffic Annotation. CoNEXT'09 Student Workshop, Dec 2009, Roma, Italy. ⟨ensl-00475933⟩