Improving Anomaly Detection Methods for Intrusion Detection in MANETS

Please enable javascript in your browser.

Javanmard,Fahime | 2012

762 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 43513 (19)
University: Sharif University of Technology
Department: Computer Engineering
Advisor(s): Hemmatyar, Ali Mohammad Afshin
Abstract:
In recent decades, Securing mobile ad hoc networks has attracted much attention. Today, several security tools, such as intrusion detection systems are used in the network. Methods based IDS works on pattern recognition and anomaly detection are divided into two categories. Pattern recognition methods based on known attack patterns work with high detection rate, but do not have the ability to detect new attacks. Anomaly detection techniques have the ability to detect new attacks, but they have high false alarm rate.
In this thesis, an anomaly detection system based on artificial immune designed, implemented and evaluated.For example, an anomaly detection methods such cases, a variety of attacks, memory being capable of self-regulation and learning algorithms of artificial immune noted. The proposed system combines network algorithm based on the immune danger theory is used. The data collected is used by the NS2 simulator. To evaluate the proposed system is implemented with two attacking the black hole and wormhole. Evaluation results show that the proposed algorithm and the accuracy of the detection rate is satisfactory.
Keywords:
Mobile Ad Hoc Network ; Intrusion Detection System ; Artificial Immune System (AIS)

No TOC