Detection method for encrypted malicious traffic

By combining random noise with malicious traffic samples, using generator and two-way long and short-term memory network models to extract features, and building a multi-level learning framework, the limitations of encrypted malicious traffic detection in the existing technology are solved, and the complex encrypted traffic is realized Accurate identification and real-time monitoring.

CN119109709AActive Publication Date: 2024-12-10INNER MONGOLIA UNIV OF TECH
View PDF 7 Cites 4 Cited by

Patent Information

Application Number
CN202411507198.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-28
Publication Date
2024-12-10
Estimated Expiration
2044-10-28

Smart Images

  • Figure CN119109709A_ABST
    Figure CN119109709A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of encryption, in particular to a detection method for encrypted malicious traffic. The random noise vector and the malicious traffic sample are combined, diversified training data can be generated, the model can learn more features, context information can be considered and richer feature representation can be extracted at the same time by using the bidirectional model, complex modes and sequential relations in encrypted traffic can be captured, and the method has the advantages of being high in robustness, high in robustness and the like. By combining bidirectional feature extraction, forged sample generation and adversarial training, a multi-level learning framework is formed, dependence on a single algorithm is avoided, by combining time sequence features and local spatial features, the model can effectively capture dynamic changes and spatial relevance of traffic, the detection accuracy is improved, and the detection efficiency is improved. Through real-time traffic monitoring and confidence scoring, the method effectively solves the limitation of the existing encrypted malicious traffic detection, and enhances the real-time performance, dynamic adaptability and detection accuracy of the system.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Network encrypted traffic recognition method and device

    CN112163594A

  • Network malicious encrypted traffic identification method and system

    CN112949702A

  • Social network false flow black and grey production automatic mining method and system

    CN113486243A

  • Malicious traffic avoidance detection method based on generative adversarial network

    CN116707992A

  • Encrypted traffic identification method fusing one-dimensional Inception structure and ET-BERT model

    CN116996447A

Cited By

  • Malicious encrypted traffic detection method based on multi-dimensional feature representation and prompt learning

    CN119766534A

  • Generation method and device of malicious domain name sample and storage medium

    CN120378400A

  • Network intrusion detection method, system, device, medium and product

    CN120811688A

  • Attack detection method and device for encryption process of vehicle-mounted firmware update package

    CN121841752A