Class-of-service mapping for QoS: a statistical signature-based approach to IP traffic classification

By Matthew Roughan, Subhabrata Sen, Oliver Spatscheck, Nick G. Duffield. ACM/SIGCOMM Internet Measurement Conference, 2004

584 citations recorded by the source corpus.

Subfields: Computer Networks and Communications

View Paper: https://www.semanticscholar.org/paper/d10ffbd7675f3373d62db9ff087b22c42991383e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

Citations in the corpus

Early application identification (2006, background): “As a result, the applicability of port-based analysis is increasingly limited [14, 21, 17, 15].”
Traffic classification using clustering algorithms (2006, methodology): “use nearest neighbor and linear discriminate analysis [14].”
Internet inter-domain traffic (2010, background): “Unfortunately, port numbers alone provide a severely limited mechanism for classifying applications [33].”
Offline/realtime traffic classification using semi-supervised learning (2007, methodology): “[27] classified flows into four predetermined traffic classes (interactive, bulk data transfer, streaming, and transactional) using the Nearest Neighbor and the Linear Discriminate Analysis classification techniques.”
Traffic classification through simple statistical fingerprinting (2007, methodology): “[11] shows that a useful set of features allowing discrimination between traffic classes can be located at different levels (single packet, flow, connection and so on) and they can be successfully exploited by Nearest Neighbor and Linear Discriminant Analysis algorithms.”
A survey of techniques for internet traffic classification using machine learning (2008, methodology): “[18] proposed to use the nearest neighbours (NN), linear discriminate analysis (LDA) and Quadratic Discriminant Analysis (QDA) ML algorithms to map different network applications to predetermined QoS traffic classes.”
Internet traffic classification demystified: myths, caveats, and the best practices (2008, methodology): “Substantial attention has been invested in data mining techniques and machine learning algorithms using flow features for traffic classification [29, 31, 36, 9, 16, 17, 7, 14, 42, 20, 39, 27].”
Network Traffic Classification Using Correlation Information (2013, methodology): “, k-Nearest Neighbor (k-NN) [13], usually require no training phase and make classification decision based on closest training samples in the feature space [14].”
Automated traffic classification and application identification using machine learning (2005, methodology): “In [12] the authors use nearest neighbour (NN) and linear discriminate analysis (LDA) to successfully map applications to different QoS classes using up to four attributes.”
A Survey on Internet Traffic Identification (2009, background): “One research study performs a statistical identi ﬁ cation of application type based on traf ﬁ c attributes and labels it to a speci ﬁ c class of service in the network [52].”
Internet traffic classification using bayesian analysis techniques (2005, unknown): “[10] perform classification of traffic flows into a small number of classes suitable for Quality of Service applications.”
Support Vector Machines for TCP traffic classification (2009, background): “Roughan et al. [6] show that a useful set of features allowing discrimination between traffic classes can be located at different levels such as single packet, flow, connection and so on.”
FS-Net: A Flow Sequence Network For Encrypted Traffic Classification (2019, methodology): “(2004) used statistical signatures to classify P2P application traffic [19].”
Bayesian Neural Networks for Internet Traffic Classification (2007, background): “, [9] and [24]), shows that, even with the removal of port and IP address, data-anonymization may not hide the application in use.”