基于免疫原理的网络入侵检测技术研究-计算机系统结构专业论文.docxVIP

重庆大学硕士学位论文 重庆大学硕士学位论文 英文摘要 ABSTRACT With the development of network technology,the security problem of computer networks is becoming more important day by day．The problem mainly includes the information security in transmission and storage in network．Because the traditional network security techniques，such FireWall，IA(Identiflcation and Authentication)and on,are passive defense，the intrusion detection technology is necessary．Nowadays，the intrusion detection is one important task which fulfills active defense to networks and information．IDS(Intrusion Detection System)is analogous to the IS(Immune System)． The problem of pmtecting computer systems fiom malicious intrusions can similarly be viewed as the problem of distinguishing self from dangerous nonself(In term of IDS，self patterns represent acceptable and legal events；nonself patterns represent unacceptable and illegal events)．Immune systems provide naive template for researching and developing network IDS．So，the research ofIDS inspired by biological ilTtrnune system is one hot point after neural network and evolving calculation．The goal ofthe researeh is to solve network intrusion detection problem using irfa21une principles，architecture and algorithms extracted from im／1211rle system． This dissertation spreads its discussions around the issue of network intrusion detection based on immunulogical principles．An immunnological model of distributed detection，called negative detection，is studied．The emphasis is improving the method of intrusion detection and the negative selection algorithm．The primary works of the dissertation are as foUows： (1)The statistical data indicates that there are high-frequency connections in LAN， and these connections is sparse．Based the statistical data and its characters，the method ofthe negative detection is improved．The method for improvement is that filter is constructed with the