基于网络流量的移动恶意应用实时检测方法及装置
By using flow spectrum transformation based on network traffic and generative adversarial networks, the problem of the inability to detect encrypted traffic in real time in existing technologies is solved, enabling real-time detection and identification of malicious traffic while reducing computational and storage overhead.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INSTITUTE OF INFORMATION ENGINEERING CHINESE ACADEMY OF SCIENCES
- Filing Date
- 2023-06-28
- Publication Date
- 2026-07-17
AI Technical Summary
Existing deep packet inspection and deep learning methods cannot effectively detect encrypted traffic and cannot achieve real-time detection and alerting of malicious traffic. Traditional machine learning, which relies on feature engineering, is difficult to quickly identify malicious traffic in high-bandwidth networks.
By collecting network traffic data packets, extracting the quintuple, length, timestamp, and transmission protocol type, generating traffic path signatures, and using flow spectrum transformation and generative adversarial networks to generate new feature vectors, performing cluster analysis, delineating benign traffic boundaries, and achieving real-time malicious traffic detection.
It enables real-time detection of malicious traffic, reduces computational and storage overhead, and improves the ability to identify encrypted traffic.
Smart Images
Figure CN116996261B_ABST