基于生物不确定性记忆的入侵检测模型研究-计算机应用技术专业论文.docxVIP

Abstract The popularization and application of Internet has made the present computer system become a complicated and open system, which has brought great facility to the development of society and peoples life. But while the network size is expanding constantly, the network is easier to be invaded. The intrusion detection system which is one of the key technologies for network safe has got the extensive concern. This text is devoted to incorporating the biological memory mechanism and uncertain principle in the detection system to make the invasion detection system more intelligent.Based on the analyses of the current situation of present network security and intrusion detection system, the main work are listed below: through the combining with memory principle and fuzzy principle, and expanding the traditional memory cell, A new intrusion detection model based on biological fuzzy memory principles are proposed, which can remember normal degree and unusual degree of a behavior of the system at the same time, while providing the mian frame and detailed definition for the model, the memory cell updating 、 decay course , relevant algorithms and instance analysis results are put forward; A new intrusion detection model based on Both-Branch fuzzy memory and markov processes is proposed by the combining with the Both-Branch fuzzy memory and biological memory principle, which uses the markov processes to analyse the order activity on the network at the same time, this model remembers the system activity influence degree to network normal or unusual when memorying normal degree and unusual degree, uses the the cerebellum and brain to distinguish memory the rules invade and not rules invade, while carrying on memory upgrading , decay , use the latent Markovs course to analyse , the instance proves there is very good measuring results in it; based on the combining with the Biological Immune And Catastrophe Theory,A new instrusion tolerance trigger model based on biological