Intrusion Detection Approach Based on DNA Signature

Abstract

Intrusion-detection systems (IDSs) aim at detecting attacks against computer systems and networks or, in general, against information systems. Most of the diseases in human body are discovered through Deoxyribonucleic Acid (DNA) investigations. In this paper, the DNA sequence is utilized for intrusion detection by proposing an approach to detect attacks in network. The proposed approach is a misuse intrusion detection that consists of three stages. First, a DNA sequence for a network traffic taken from Knowledge Discovery and Data mining (KDD Cup 99) is generated. Then, Teiresias algorithm, which is used to detect sequences in human DNA and assist researchers in decoding the human genome, is used to discover the Shortest Tandem Repeat (STR) sequence and its position (i.e., pattern or keys) in the network traffic. Finally, the Horspool algorithm is applied as a classification process to determine whether the network traffic is attack or normal. The performance of the proposed approach in terms of detection rate, accuracy, and false alarm rate are measured, showing the results are reasonable and accepted.