Usefulness of Sensor Fusion for Security Incident Analysis

View Sample PDF

Author(s): Ciza Thomas (College of Engineering, India)and N. Balakrishnan (Indian Institute of Science, India)
Copyright: 2012
Pages: 16
Source title: Situational Awareness in Computer Network Defense: Principles, Methods and Applications
Source Author(s)/Editor(s): Cyril Onwubiko (Research Series Ltd, UK)and Thomas Owens (Brunel University, UK)
DOI: 10.4018/978-1-4666-0104-8.ch010

Keywords: Information Science Reference / Information Security & Privacy / IT Security and Ethics / Security & Forensics

Purchase

View Usefulness of Sensor Fusion for Security Incident Analysis on the publisher's website for pricing and purchasing information.

Abstract

Intrusion Detection Systems form an important component of network defense. Because of the heterogeneity of the attacks, it has not been possible to make a single Intrusion Detection System that is capable of detecting all types of attacks with acceptable levels of accuracy. In this chapter, the distinct advantage of sensor fusion over individual IDSs is proved. The detection rate and the false positive rate quantify the performance benefit obtained through the fixing of threshold bounds. Also, the more independent and distinct the attack space is for the individual IDSs, the better the fusion of Intrusion Detection Systems performs. A simple theoretical model is initially illustrated and later supplemented with experimental evaluation. The chapter demonstrates that the proposed fusion technique is more flexible and also outperforms other existing fusion techniques such as OR, AND, SVM, and ANN, using the real-world network traffic embedded with attacks.

The IRMA Community

Research IRM

Usefulness of Sensor Fusion for Security Incident Analysis

Purchase

Abstract

Related Content

IRMA Sponsors