An Intrusion Detection Technology Based on Immune Theory

Using the specific immunity and clonal expansion principles of B , T cells of human immune system in intrusion detection is a significative research. We proposed a new kind of structure of detection, which is using vulnerability and its corresponding data, and setting up tree code mode. Employing the improved CLOPE cluster´s algorithm at antibody muster to distill the fragment of variable region, and then rearrange the gene. It can integrate the antibody muster, detect unknown attack. Experiments show that the method can not only detect defined intrusion fast and accurate, but also undefined intrusion, especially anamorphosis of defined attack.