摘要
文中简要分析了当前入侵检测技术存在的问题和发展方向,结合人体免疫理论,提出一种基于检测代理的分布式网络入侵检测免疫模型。为了提高免疫系统识别异常的轻负荷和适应性,引入粗集理论的约简算法,用于待检数据的预处理和抗体基因库的生成,并结合抗体进化原理不断对基因实施进化。该模型可降低待检数据的冗余,保持抗体基因的进化和适应性,提高入侵检测的效率和准确性。
In this paper,the problems and developing tendency of intrusion detection system were analyzed. Imitating the human immune system,a distributed network intrusion detection model based on intrusion agents was raised. For enhancing adaptability and lightweight of the immune system,we introduced reduction algorithm based on rough sets theory to preprocess a plenty of security data and to generate the antibodys gene bank. According to antibody evolutionism,these genes can autoevolutes constantly. This model not only can reduce the extra data,but also can keep the evolution and adaptability of antibodys gene. So this model can raise the efficiency and accurate of intrusion detection system greatly.
出处
《计算机应用》
CSCD
北大核心
2003年第7期26-28,共3页
journal of Computer Applications
关键词
入侵检测
免疫系统
抗体基因库
粗集约简
基因进化
intrusion detection
immune system
antibody gene bank
rough sets reduction
gene evolution
