Ant Algorithm Inspired Immune Intrusion Detector Generation Algorithm

Abstract

In this paper, we propose a novel generative method for immune intrusion detectors inspired by principle of ant algorithm. In detector generation stage, we use the characteristics of distributed processing of ant algorithm, which can quickly generate high quality detectors. In the stage of labeling the detectors, the label of the detectors is decided according to prior knowledge. The algorithm is simple, which do not need to undertake initial categories of division. Simulation experiments show that this algorithm performs as well as the state-of-the-art detector algorithms when recognizing known attacks, but is significantly more effective on identifying unknown attacks.