Research Of Artificial Immune-based Intrusion Detection Self Updating Method Based On Artificial Neural Network

With the development at full speed of global informationization, the computer online security question is outstanding day by day. The danger that the hacker invades, information reveals and the virus overflows has caused the great attention of the countries all over the world, so network information security has already become a fatal problem to settled. The traditional firewall, muter and the techniques of identity authentication and data encryption can not fit current network environment. Intrusion Detection System(IDS)is the research hotspot in the field of Network Security now, which plays an important role in safeguarding network security and people pay more attention to.The biological system is successfully at protecting the human body against a vast of foreign pathogens. Intrusion detection system deals with the similar problem. So more and more researchers working on network security start to apply the mechanisms derived from biological immune system into IDS due to their adaptability and dynamics, and some significant successes are gained. However, the once definition of normal and abnormal activities makes these Immune-based IDSs inadaptive in the real network environment. Moreover, the lack of quantitive descriptions in some Immune-based ID models makes them difficult for engineering application.This dissertation is dedicated to the research of the updating of the self-base. The main development work were as follows:(1)This dissertation introduces the concept and classification of intrusion detection, then, advantages and disadvantages of main methods, sums up the developing directions of intrusion detection.(2)After reviews of basic immunological material and artificial immune system(AIS) necessary for this dissertation, comprehensive formalization of the intrusion detection model based on AIS is presented. Analyses are derived from several typical algorithms of detector generation. The results of experiments are also presented.(3)This dissertation put forward a self-base updating method based on artificial neural network for artificial immune-based intrusion detection. This method, firstly, bring forward the rule on the updating of the mature detector; secondly, make use of the artificial neural network to update the self-base because of its excellent self-learning and self-adaptability, and then make the artificial immune-based dynamic intrusion detection fit the real network surroundings. Finally, This dissertation finished two emulation experiments, and the result shows that the new model performs much better and achieve the expected result.