Intrusion Detection Technique based on Dendritic Cell Algorithm and Dempster Belief Theory

Abstract

Today traditional intrusion detection systems are unable to detect intrusion attacks. Huge number of false alarm generated by the system results in financial loss of an organization. The unique features of artificial immune system encourage and motivate the researchers to employ this technique in variety of applications and especially in intrusion detection systems. Recently Artificial immune system (AIS) has been applied for anomaly based intrusion detection in computer networks. Artificial immune system is a new technique which is applied for solving various problems in the field of information security. In this paper we presents a intrusion detection system based on one of the algorithm of artificial immune system called the Dendritic Cell Algorithm (DCA) and Dempster–Belief Theory (DBT) in order to minimise the rate of the generation of intrusion detection system , false positive rate and improve correlation factor in the designed intrusion detection system. With the help of Dempster–Belief theory we calculate the degree of uncertainty and with the help of event gathering calculate the entropy, which help us to determine the intrusion in the given system. Data having higher entropy is regarded as the “intruder” and generate the alarm. Thus with the help of this dual detection technique we can not only minimize the false positive and false negative rate but also improved the correlation technique and decrease the intrusion rate in the system. KeywordsArtificial Immune System, intrusion detection system, human immune system, danger theory, negative selection algorithm, DCA, Dempster–Belief theory.