An Intelligent Detection Method for Network Security

To dynamically discover network attacks hidden in network data, an intelligent detection method for network security is proposed. Biological immune principles and mechanisms are adopted to judge whether network data contain illegal network packets. Signature library of network attacks and section library of attack signatures are constructed. They store attack signatures and signature sections, respectively. They are used to make the initial detection ability of proposed method. Detectors are defined to simulate immune cells. They evolve dynamically to adapt the network security. Signatures of network data are extracted from IP packets. Detectors match network data's signatures which mean some attacks. Warning information is formed and sent to network administrators according to recognized attacks.