MaMPF: Encrypted Traffic Classification Based on Multi-Attribute Markov Probability Fingerprints
MaMPF: классификация зашифрованного трафика на основе многoатрибутных вероятностных отпечатков Маркова
2018-06-01
SCID: 54.1/8hv7wn4j
Discuss with AI
Markov modelsMulti-attribute Markov Probability FingerprintsSSL/TLS trafficencrypted traffic classificationlength block sequence
Figures from the paper
Abstract (AI)
With the explosion of network applications, network anomaly detection and security management face a big challenge, of which the first and a fundamental step is traffic classification. However, for the sake of user privacy, encrypted communication protocols, e.g. the SSL/TLS protocol, are extensively used, which results in the ineffectiveness of traditional rule-based classification methods. Existing methods cannot have a satisfactory accuracy of encrypted traffic classification because of insufficient distinguishable characteristics. In this paper, we propose the Multi-attribute Markov Probability Fingerprints (MaMPF), for encrypted traffic classification. The key idea behind MaMPF is to consider multi-attributes, which includes a critical feature, namely “length block sequence” that captures the time-series packet lengths effectively using power-law distributions and relative occurrence probabilities of all considered applications. Based on the message type and length block sequences, Markov models are trained and the probabilities of all the applications are concatenated as the fingerprints for classification. MaMPF achieves 96.4% TPR and 0.2% FPR performance on a real-world dataset from campus network (including 950,000+ encrypted traffic flows and covering 18 applications), and outperforms the state-of-the-art methods.
Key Findings
1
MaMPF introduces multi-attribute Markov probability fingerprints for classifying encrypted network traffic when traditional rule-based methods are ineffective.
2
MaMPF outperforms existing state-of-the-art encrypted traffic classification methods on the evaluated dataset.
3
Markov models trained on message types and length block sequences concatenate application probabilities into fingerprints for traffic classification.
4
On a real-world campus-network dataset containing more than 950,000 encrypted flows across 18 applications, MaMPF achieves 96.4% TPR and 0.2% FPR.
5
The method uses length block sequences to capture time-series packet lengths through power-law distributions and relative application-specific occurrence probabilities.
Research Object
Encrypted network traffic flows (SSL/TLS) from a campus network covering multiple applications
Research Subject
application-level encrypted traffic classification based on multi-attribute Markov probability fingerprints, particularly using packet-length sequences and message types
Publication Details
Publication Date
2018-06-01
Journal
Publisher
ISSN
Cited by
124
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai2
Cited by5
Network traffic classification: Techniques, datasets, and challenges2022
A Novel Multimodal Deep Learning Framework for Encrypted Traffic Classification2022
Seeing Traffic Paths: Encrypted Traffic Classification With Path Signature Features2022
Automatic Mobile App Identification From Encrypted Traffic With Hybrid Neural Networks2020
Incremental Learning for Mobile Encrypted Traffic Classification2021