MIETT: Multi-Instance Encrypted Traffic Transformer for Encrypted Traffic Classification
MIETT: трансформер зашифрованного трафика с множественными экземплярами для классификации зашифрованного трафика
2025-04-11
SCID: 54.1/8vpgm4fs
Discuss with AI
Flow Contrastive Learning (FCL)Multi-Instance Encrypted Traffic Transformer (MIETT)Packet Relative Position Prediction (PRPP)Two-Level Attention (TLA)encrypted traffic classification
Figures from the paper
Abstract (AI)
Network traffic includes data transmitted across a network, such as web browsing and file transfers, and is organized into packets (small units of data) and flows (sequences of packets exchanged between two endpoints). Classifying encrypted traffic is essential for detecting security threats and optimizing network management. Recent advancements have highlighted the superiority of foundation models in this task, particularly for their ability to leverage large amounts of unlabeled data and demonstrate strong generalization to unseen data. However, existing methods that focus on token-level relationships fail to capture broader flow patterns, as tokens, defined as sequences of hexadecimal digits, typically carry limited semantic information in encrypted traffic. These flow patterns, which are crucial for traffic classification, arise from the interactions between packets within a flow, not just their internal structure. To address this limitation, we propose a Multi-Instance Encrypted Traffic Transformer (MIETT), which adopts a multi-instance approach where each packet is treated as a distinct instance within a larger bag representing the entire flow. This enables the model to capture both token-level and packet-level relationships more effectively through Two-Level Attention (TLA) layers, improving the model's ability to learn complex packet dynamics and flow patterns. We further enhance the model's understanding of temporal and flow-specific dynamics by introducing two novel pre-training tasks: Packet Relative Position Prediction (PRPP) and Flow Contrastive Learning (FCL). After fine-tuning, MIETT achieves state-of-the-art (SOTA) performance across five datasets, demonstrating its effectiveness in classifying encrypted traffic and understanding complex network behaviors.
Key Findings
1
After fine-tuning, MIETT achieves state-of-the-art (SOTA) performance across five datasets for encrypted traffic classification.
2
Introduces Two-Level Attention (TLA) layers to model both token-level and packet-level interactions, improving learning of complex packet dynamics and flow patterns.
3
Introduces two novel pre-training tasks—Packet Relative Position Prediction (PRPP) and Flow Contrastive Learning (FCL)—to enhance temporal and flow-specific representation learning.
4
Proposes MIETT, a Multi-Instance Encrypted Traffic Transformer that treats each packet as an instance within a flow-level bag to capture packet- and flow-level relationships.
Research Object
Encrypted network traffic flows (modeled as bags of packets for multi-instance learning)
Research Subject
Classification of encrypted traffic by learning token-level and packet-level relationships and flow patterns using a Multi-Instance Encrypted Traffic Transformer (MIETT) with Two-Level Attention and pre-training tasks (Packet Relative Position Prediction and Flow Contrastive Learning)
Publication Details
Publication Date
2025-04-11
Journal
Publisher
ISSN
Cited by
27
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai5
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale2020
ET-BERT: A Contextualized Datagram Representation with Pre-training Transformers for Encrypted Traffic Classification2022
FS-Net: A Flow Sequence Network For Encrypted Traffic Classification2019
FlowPrint: Semi-Supervised Mobile-App Fingerprinting on Encrypted Network Traffic2020
PERT: Payload Encoding Representation from Transformer for Encrypted Traffic Classification2020