2010 · 99 citations · 10 references
ComputingComputer ScienceMachine LearningPort EngineeringPrivacy EngineeringNetwork Traffic AnalysisPattern RecognitionTraffic EngineeringTransport LayerAccuracy-computational Cost TradeoffClassification AccuracyClassification Accuracy AssessmentDeep Packet InspectionPacket ClassificationPacket InspectionTraffic ClassificationTraffic Inspection
Traffic classification approaches based on deep packet inspection (DPI) are considered very accurate, however, two major drawbacks are their invasiveness with respect to users privacy, and their significant computational cost. Both are a consequence of the amount of per-flow payload data - we refer to it as "deepness" - typically inspected by such algorithms. At the opposite side, the fastest and least data-eager traffic classification approach is based on transport-level ports, even though today it is mostly considered inaccurate. In this paper we propose a novel approach to traffic classification - named PortLoad - that takes the advantages of both worlds: the speed, simplicity and reduced invasiveness of port-based approaches, on a side, and the classification accuracy of DPI on the other one.
10
Transport layer identification of P2P traffic
Thomas Karagiannis, Andre Broido, Michalis Faloutsos et al. · 2004 · 713 citations
Computing, Information Technology, Digital Communication +16
Algorithms for packet classification
Pravir Singh Gupta, Nick McKeown · IEEE Network · 2001 · 653 citations
Thomas Karagiannis, Andre Broido, Nevil Brownlee et al. · 2005 · 507 citations