A network security situation perception model based on immune state transition

This paper proposes a new Model of Network Security Situation Perception based on artificial immunity which can realize immune state transition. This model is used to solve the self and non-self dynamic describing problem in computer immune system. It can realize the detection to known and unknown network intrusion. It has some characteristics like distributed behavior, self-learning, robustness etc; it also presented memory Immunological detection trees based on the antibody concentration. It adjusts each node´s order dynamically. Each time while testing, detector is extracted from high to low according to the antibody concentration to matching antigen. When antibody concentration is changed with the network attack, adjust the tree node sequence timely, make high concentrations detector match antigen prior to the low concentration detector for reducing invalid matching times and improving detection efficiency.