EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Semi-Supervised Encrypted Malicious Traffic Detection Based on Multimodal Traffic Characteristics

Ming Liu iD, Qichao Yang, Wenqing Wang, Shengli Liu

Sensors · 10.10.2024 · 24 (20) · 6507

DOI: 10.3390/s24206507 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 10.10.2024. Verfasst von Ming Liu und 3 weiteren Autor:innen. Beteiligte Einrichtung: Information Engineering University, Zhengzhou 450001, China. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Key Research and Development Program of China, Science & Technology Commission Foundation Strengthening Project, Science & Technology Commission Foundation Strengthening Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The exponential growth of encrypted network traffic poses significant challenges for detecting malicious activities online. The scale of emerging malicious traffic is significantly smaller than that of normal traffic, and the imbalanced data distribution poses challenges for detection. However, most existing methods rely on single-category features for classification, which struggle to detect covert malicious traffic behaviors. In this paper, we introduce a novel semi-supervised approach to identify malicious traffic by leveraging multimodal traffic characteristics. By integrating the sequence and topological information inherent in the traffic, we achieve a multifaceted representation of encrypted traffic. We design two independent neural networks to learn the corresponding sequence and topological features from the traffic. This dual-feature extraction enhances the model’s robustness in detecting anomalies within encrypted traffic. The model is trained using a joint strategy that minimizes both the reconstruction error from the autoencoder and the classification loss, allowing it to effectively utilize limited labeled data alongside a large amount of unlabeled data. A confidence-estimation module enhances the classifier’s ability to detect unknown attacks. Finally, our method is evaluated on two benchmark datasets, UNSW-NB15 and CICIDS2017, under various scenarios, including different training set label ratios and the presence of unknown attacks. Our model outperforms other models by 3.49% and 5.69% in F1 score at labeling rates of 1% and 0.1%, respectively.

Schlagwörter

Stichwörter der Autor:innen

encrypted malicious traffic detectionmultimodal featuresnetwork securitysemi-supervised learning

Autor:innen und Institutionen

  • Information Engineering University, Zhengzhou 450001, China.

ORCID-Profile: Ming Liu

Förderung

  • National Key Research and Development Program of China
  • Science & Technology Commission Foundation Strengthening Project
  • Science & Technology Commission Foundation Strengthening Project

Zitieren

APA

Liu, M., Yang, Q., Wang, W., & Liu, S. (2024). Semi-Supervised Encrypted Malicious Traffic Detection Based on Multimodal Traffic Characteristics. Sensors, 24(20), 6507. https://doi.org/10.3390/s24206507

Vancouver

Liu M, Yang Q, Wang W, Liu S. Semi-Supervised Encrypted Malicious Traffic Detection Based on Multimodal Traffic Characteristics. Sensors. 2024;24(20):6507. doi:10.3390/s24206507

BibTeX
@article{Liu2024,
  author = {Liu, Ming and Yang, Qichao and Wang, Wenqing and Liu, Shengli},
  title = {Semi-Supervised Encrypted Malicious Traffic Detection Based on Multimodal Traffic Characteristics},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {20},
  pages = {6507},
  doi = {10.3390/s24206507}
}
Kostenlos starten →