License: Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported license (CC BY-NC-ND 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/OASIcs.KiVS.2011.61
URN: urn:nbn:de:0030-drops-29582
Go to the corresponding OASIcs Volume Portal

Münz, Gerhard ; Heckmüller, Stephan ; Braun, Lothar ; Carle, Georg

Improving Markov-based TCP Traffic Classification

6.pdf (1 MB)


This paper presents an improved variant of our Markov-based TCP traffic classifier and demonstrates its performance using traffic captured in a university network. Payload length, flow direction, and position of the first data packets of a TCP connection are reflected in the states of the Markov models. In addition, we integrate a new "end of connection" state to further improve the classification accuracy. Using 10-fold cross validation, we identify appropriate settings for the payload length intervals and the number of data packets considered in the models. Finally, we discuss the classification results for the different applications.

BibTeX - Entry

  author =	{Gerhard M{\"u}nz and Stephan Heckm{\"u}ller and Lothar Braun and Georg Carle},
  title =	{{Improving Markov-based TCP Traffic Classification}},
  booktitle =	{17th GI/ITG Conference on Communication in Distributed Systems (KiVS 2011) },
  pages =	{61--72},
  series =	{OpenAccess Series in Informatics (OASIcs)},
  ISBN =	{978-3-939897-27-9},
  ISSN =	{2190-6807},
  year =	{2011},
  volume =	{17},
  editor =	{Norbert Luttenberger and Hagen Peters},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum fuer Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{},
  URN =		{urn:nbn:de:0030-drops-29582},
  doi =		{10.4230/OASIcs.KiVS.2011.61},
  annote =	{Keywords: Markov model, TCP Traffic Classification, network}

Keywords: Markov model, TCP Traffic Classification, network
Collection: 17th GI/ITG Conference on Communication in Distributed Systems (KiVS 2011)
Issue Date: 2011
Date of publication: 25.02.2011

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI