Using of time characteristics in data flow for traffic classification

Authors:
Pavel Piskac;Jiri Novotny
Affiliations:
Institute of Computer Science, Masaryk University, Brno, Czech Republic;Institute of Computer Science, Masaryk University, Brno, Czech Republic
Venue:
AIMS'11 Proceedings of the 5th international conference on Autonomous infrastructure, management, and security: managing the dynamics of networks and services
Year:
2011

Citing 3
Cited 2

BLINC: multilevel traffic classification in the dark

Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications
Traffic classification using clustering algorithms

Proceedings of the 2006 SIGCOMM workshop on Mining network data
Early application identification

CoNEXT '06 Proceedings of the 2006 ACM CoNEXT conference

Flow information storage assessment using IPFIXcol

AIMS'12 Proceedings of the 6th IFIP WG 6.6 international autonomous infrastructure, management, and security conference on Dependable Networks and Services
Application specific processor with high level synthesized instructions (abstract only)

Proceedings of the 2014 ACM/SIGDA international symposium on Field-programmable gate arrays

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a protocol detection using statistic information about a flow extended by packet sizes and time characteristics, which consist of packet inter-arrival times. The most common way of network traffic classification is a deep packet inspection (DPI). Our approach deals with the DPI disadvantage in power consumption using aggregated IPFIX data instead of looking into packet content. According to our previous experiments, we have found that applications have their own behavioral pattern, which can be used for the applications detection. With a respect to current state of development, we mainly present the idea, the results which we have achieved so far and of our future work.