基于规则的多模型集成化异常入侵检测系统  

Multiple-Model Integrated Anomaly Intrusion Detection System Based on Rules

作　　者：赵丽[1] 秦振吉[1] 常文萃[1] 

机构地区：[1]晋中学院计算机科学与技术系,晋中030600

出　　处：《现代计算机》2008年第11期83-85,92,共4页Modern Computer

摘　　要：在讨论相关技术的基础上,提出一种多模型集成的入侵检测系统,然后运用不同的模型和方法,构建系统的正常行为,并将其使用基于规则的方法集成,实验证明系统的检测率得到提高的同时,大大降低了误警率。Proposes a multiple-model intrusion detection system on the basis of related technology. It constructs normal behavior of the system by using different models and methods. The models are integrated based on rule methods subsequently. In the end, the experiment proves that the detection rare of the system is improved and at the same time the false positive rate is reduced greatly.

关 键 词：入侵检测系统(IDS) 隐马尔可夫模型(HMM) 系统调用 神经网络 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术] TQ02[自动化与计算机技术—计算机科学与技术]