摘要
针对目前入侵检测系统不能有效检测已知攻击的变种和未知攻击行为的缺陷,受免疫系统中动态克隆选择算法的启发,提出了一种基于改进的动态克隆选择算法的入侵检测模型.该模型可以适应连续改变的环境,动态地学习变化的“正常”模式以及预测新的“异常”模式.经实验证明,该模型在降低误报率的情况下,提高了检测率.
Because intrusion detection systems couldn＇t detect the mutant of existing intrusion behavior and undefined intrusion behavior effectively, according to the Dynamic Clonal Selection algorithms in the biological immune system, this paper presents an intrusion detection model based on extended Dynamic Clonal Selection algorithms. This model can adapt to continuously changing environments, dynamically learning the fluid normal patterns and predict new anomaly patterns. Experiment results show that this model improves the detection rate and maintains a low false alarm rate.
出处
《哈尔滨理工大学学报》
CAS
2006年第1期82-85,89,共5页
Journal of Harbin University of Science and Technology
基金
哈尔滨市学科后备带头人基金项目[2003AFXXJ013]
黑龙江省教育厅科学技术研究项目[10541044]
关键词
入侵检测
动态克隆选择
自体
非自体
intrusion detection
dynamic clonal selection
self
non-self
