Encrypted Traffic Classification
AI-powered methods for classifying and analyzing encrypted network traffic
Over 90% of internet traffic is now encrypted. While encryption protects privacy, it also makes traditional network monitoring impossible. We develop AI systems that classify encrypted traffic without breaking encryption—enabling network security and management while preserving user privacy.
Key Challenges
Modern encryption protocols (TLS 1.3, ESNI, DoH, HTTP/3) hide traffic characteristics completely. Network operators need to identify traffic types for security monitoring, quality of service, and capacity planning, but cannot compromise encryption or user privacy.
Our Approach
Generative AI for Data Synthesis
Using GANs to create realistic encrypted traffic datasets. Addresses data scarcity in encrypted traffic research.
Post-Quantum Traffic Classification
First system (PQClass) to classify traffic encrypted with post-quantum algorithms. Prepares networks for quantum-resistant cryptography.
Spectral Analysis
Frequency-domain analysis identifies application patterns in encrypted communications without decryption.
Advanced Data Augmentation
Novel techniques that improve model performance across diverse network conditions.
Zero-Day Detection
SimCSE-based contrastive learning detects previously unseen attack patterns in encrypted traffic.
Impact
Our methods achieve high accuracy while fully respecting encryption and privacy. Applications include real-time threat detection, network optimization, and the world's first post-quantum traffic classification system.
Related Publications
2026
- Quality of Experience Prediction for First Person Shooter Online Gaming: The Case Study of Call of Duty2026Proceedings of the IEEE Consumer Communications & Networking Conference (CCNC 2026)
2025
- Enhancing Encrypted Internet Traffic Classification Through Advanced Data Augmentation Techniques2025Proceedings of the IEEE International Conference on Communications
- PQClass: Classification of Post-Quantum Encryption Applications in Internet Traffic2025Proceedings of the IEEE International Conference on Communications
- A Classification-by-Retrieval Framework for Few-Shot Anomaly Detection to Detect API Injection2025Computers & Security
2024
- The Art of Time-Bending: Data Augmentation and Early Prediction for Efficient Traffic Classification2024Expert Systems with Applications
- CBR–Boosting Adaptive Classification By Retrieval of Encrypted Network Traffic with Out-of-Distribution2024arXiv preprint arXiv:2403.11206
- Enhancing Encrypted Internet Traffic Classification Through Advanced Data Augmentation Techniques2024arXiv preprint arXiv:2407.16539
- OSF-EIMTC: An Open-Source Framework for Standardized Encrypted Internet Traffic Classification2024Computer Communications
- Hidden in Time, Revealed in Frequency: Spectral Features and Multiresolution Analysis for Encrypted Internet Traffic Classification20242024 IEEE 21st Consumer Communications & Networking Conference (CCNC)
- Revolutionizing Our Way to Better Classifiers: Leveraging Synthetic Data with Generative Models for Encrypted Network Traffic Classification2024Available at SSRN 4654236
2023
- When a RF Beats a CNN and GRU, Together—A Comparison of Deep Learning and Classical Machine Learning Approaches for Encrypted Malware Traffic Classification2023Computers & Security
2022
- SimCSE for Encrypted Traffic Detection and Zero-Day Attack Detection2022IEEE Access
- MalDIST: From Encrypted Traffic Classification to Malware Traffic Detection and Classification20222022 IEEE 19th annual consumer communications & networking conference (CCNC)
2021
- PCL: Packet Classification with Limited Knowledge2021IEEE INFOCOM 2021-IEEE Conference on Computer Communications
2020
- Encrypted Video Traffic Clustering Demystified2020Computers & Security
- Robust Machine Learning for Encrypted Traffic Classification2020CoRR
2016
- Robust Machine Learning for Encrypted Traffic Classification2016arXiv preprint arXiv:1603.04865