Intrusion Detection System (P-0054) (MFN 1820)

Abstract

As we increasingly rely on information infrastructures to support critical operations in different areas, intrusions into information systems have become a significant threat to our society with potentially severe consequences. DoS attack aims at degrading availability of the network system. Intrusion detection techniques can have a significant role in the detection of computer abuse such ss DoS attack.L This project describes a multivariate chi-square statistical approach to represent and detect intrusions. Multivariate technique is used because it uses more than one metrics to detect a single kind of attack, which is more accurate. Chi-square technique uses fewer computations than other multivariate techniques, so it was selected as the correlation algorithm. The DoS attacks are classified on the basis of the protocol they use. The prototype made on the basis of proposed architecture is a light weight detection and analysis engine to detect DoS attacks which cause no false alarms.