A Software Deep Packet Inspection System for Network Traffic Analysis and Anomaly Detection.

Area of Science:

• Computer Science
• Network Security
• Data Analysis

Background:

• Network anomalies pose significant threats to system integrity and data security.
• Existing anomaly detection methods may lack efficiency or the ability to adapt to diverse traffic types.

Purpose of the Study:

• To propose a new method for detecting network anomalies by analyzing the Hurst parameter of network traffic.
• To develop algorithms for a Deep Packet Inspection (DPI) system that enhances network security and Quality of Service (QoS).

Main Methods:

• Evaluating the Hurst (H) parameter of network traffic using the rescaled range (RS) method.
• Defining criteria for anomaly detection and prevention using the Three Sigma Rule and Hurst parameter.
• Developing algorithms for DPI, protocol detection, statistical analysis, and flow regulation for QoS.

Main Results:

• The proposed method demonstrates low computational time and short monitoring duration.
• The developed DPI system components effectively identify non-standard factors and dependencies, improving intrusion detection systems.
• Comparison with SolarWinds Deep Packet Inspection shows potential for enhanced anomaly detection and prevention.

Conclusions:

• The Hurst parameter-based method provides an efficient and adaptable approach to network anomaly detection.
• The developed DPI system enhances information security, particularly for Internet of Things (IoT) communication infrastructures.
• The system reduces risks associated with network vulnerabilities and improves overall cybersecurity posture.