Intrusion Detection System Using BCO and Genetic Algorithm and a Comparison with ACO

Abstract

Intrusion Detection System (IDS) that is increasingly becomes a major part of defense system are used to recognize suspicious activities in a computer system and to protect its security. In the network, an intruder moves between various nodes to trace the path of attack. For the security of computer networks, the Intrusion detection and its prevention play an extremely important role. The traditional intrusion detection system has some limitations, to overcome from them, alerts are made which are exchanged and correlated in a co-operative fashion. So, the rapid development of attack detection is the basic necessity and there is requirement to develop intrusion detection algorithms based on fast machine learning having low false negative rates and high detection rates. In this paper, presents an intelligent learning approach using Bee Colony Optimization (BCO) and Genetic Algorithm (GA) to detect intrusions in the network and their comparison with the Ant Colony Optimization (ACO). The results and the experimental analysis of the hybrid system shows that system can achieve higher accuracy rate in identifying whether the activities are normal or abnormal ones and obtained reasonable detection rate with low false alarm rate.