A Dynamic Intrusion Detection System Based on Multivariate Hotelling's T2 Statistics Approach for Network...

Area of Science:

• Computer Science
• Network Security
• Statistical Analysis

Background:

• Growing communication demands necessitate robust network security.
• Effective network intrusion detection is crucial for secure data transfer.
• Existing models require continuous improvement for efficiency and accuracy.

Purpose of the Study:

• To develop an efficient and dynamic network intrusion detection model.
• To apply Hotelling's T(2) multivariate statistical analysis for intrusion detection.
• To enhance network security protocols for reliable communication.

Main Methods:

• Utilized Hotelling's T(2) method, a multivariate statistical technique.
• Incorporated preprocessing, statistical analysis, and attack detection components.
• Generated profiles using T-square distance metrics and the central limit theorem for thresholding.

Main Results:

• Classified network traffic profiles as normal or attack types with high accuracy.
• Achieved very high detection rates across all attack classes.
• Demonstrated significantly low false alarm rates compared to existing models.

Conclusions:

• The multivariate Hotelling's T(2) model offers superior performance for network intrusion detection.
• This approach provides a more accurate and reliable security solution.
• The model effectively addresses the need for enhanced network security in modern communication systems.