Rosetta: Enabling Robust TLS Encrypted Traffic Classification in Diverse Network Environments with TCP-Aware Traffic Augmentation
Rosetta: обеспечение надежной классификации зашифрованного TLS-трафика в различных сетевых средах с помощью аугментации трафика с учетом TCP
2023-07-28
SCID: 54.1/pm4bxrnt
Discuss with AI
TCP-aware traffic augmentationTLS encrypted traffic classificationdeep learning modelsrobust TLS flow featuresself-supervised learning
Figures from the paper
Abstract (AI)
As the majority of Internet traffic is encrypted by the Transport Layer Security (TLS) protocol, recent advances leverage Deep Learning (DL) models to conduct encrypted traffic classification. We propose Rosetta to enable robust TLS encrypted traffic classification for existing DL models. It leverages TCP-aware traffic augmentation mechanisms and self-supervised learning to understand implicit TCP semantics, and hence extracts robust features of TLS flows. Extensive experiments show that Rosetta can significantly improve the classification performance of existing DL models on TLS traffic in diverse network environments.
Key Findings
1
Extensive experiments demonstrate significant classification-performance improvements for existing deep-learning models when applied to TLS traffic.
2
Rosetta improves the robustness of existing deep-learning models for TLS-encrypted traffic classification across diverse network environments.
3
Rosetta incorporates self-supervised learning to learn robust representations of encrypted TLS traffic without relying solely on labeled data.
4
The method introduces TCP-aware traffic augmentation to expose implicit TCP semantics and improve TLS-flow feature extraction.
Research Object
TLS-encrypted network traffic (TLS flows) in diverse network environments
Research Subject
Robust classification performance and feature extraction of TLS flows across diverse network environments
Publication Details
Publication Date
2023-07-28
Journal
Publisher
ISSN
Cited by
84
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest