Restudying the Artificial Immune Model for Network Intrusion Detection

In order to quicken the affinity maturation process of detector population and improve the efficiency of network intrusion detection, this paper describes detailed vaccine operator, algorithm of adaptive extracting vaccine and immune evolutionary algorithm (IEA), and then design a novel artificial immune model and algorithm for network intrusion detection which integrates negative selection algorithm (NSA) with IEA. This model can also satisfy three requirements of distributed, self-organizing and lightweight. The network intrusion detection experiments based on the novel model and algorithm are designed to compare with Kim´s artificial immune model for network intrusion detection which is based on clonal selection algorithm (CSA) and NSA. Experimental results show that the novel model and its algorithm quickens the affinity maturation process of detector population and stably increases the detection rate along with increasing evolutionary generation; but in Kim´s conceptual mode, the affinity maturation process of detector population takes more time, the detection rate falls into a little degradation and maintains invariant for a long time.