Anomaly Detection

Network-Based Anomaly Detection Using Discriminant Analysis

ABSTRACT

Anomaly-based Intrusion Detection Systems (IDS) can be a valuable tool for detecting novel network attacks. This paper analyzes the use of linear and non-linear discriminant analysis on packet header information from Transport and Internet layers of the TCP/IP model to classify packets as normal or abnormal. By training on normal traffic for a particular service (web and secure shell) and known attacks, the classifier can automatically identify differences between packets that may be used to classify future unknown traffic.

Journal of Information Warfare

The definitive publication for the best and latest research and analysis on information warfare, information operations, and cyber crime. Available in traditional hard copy or online.

Keywords

Quill Logo

The definitive publication for the best and latest research and analysis on information warfare, information operations, and cyber crime. Available in traditional hard copy or online.

SUBSCRIBE NOW

Get in touch

  • Journal of Information Warfare
    114 Ballard Street
    Yorktown, VA
    23690
  • 757.234.6664
  • jiw@gbpts.com