An Anomaly Detection and Analysis Method for Network Traffic Based on Correlation Coefficient Matrix

摘要

Based on TCP protocol, this paper aims at TCP flows, discusses the effects of multivariate correlation analysis on network traffic, obtains the quantitative relationship between different types of TCP packets in each time unit by correlation coefficient matrix, and finally proposes an anomaly detection and analysis method based on the correlation coefficient matrix. The experimental results show that our method can efficiently distinguish normal and abnormal traffic, and accurately detect and classify various anomaly behaviors (such as network scanning and DDoS attacks) in network traffic. The linear complexity of our method makes real-time detection and analysis practical.