摘要
根据免疫原理,对入侵检测中动态克隆选择算法的群体循环数据标准归一化处理、亲和力计算以及变异方案进行研究和改进.实验表明,改进后可使基于该算法的入侵检测引擎有较好的自适应性和学习能力,在提高检测率的基础上,降低了检测误报率.
Based on the principle of immunity and according to the dynamic clone selection algorithm in intrusion detection, this paper conducts a research on the unitary standardization process of the data in colony circulation from the algorithm, and on the calculation of affinity and the variation project in the hope of improvement. Experiments prove that after improvement, the intrusion detection engine based on the algorithm achieves a goal of strong adaptive and learning ability, thus reducing the false alarm rate besides improving the detection rate.
出处
《中南林业科技大学学报》
CAS
CSCD
北大核心
2008年第1期139-142,共4页
Journal of Central South University of Forestry & Technology
基金
湖南省教育厅科研项目(05D035)
关键词
软件工程
网络技术
网络安全
免疫原理
入侵检测
动态克隆选择算法
变异
software engineering
networks security
network technique
immune principle
intrusion detection
dynamic cloneselection algorithm
variation
