基于支持向量机与移动Agent的入侵检测系统模型.pdfVIP

第l7卷 第1期 云南民族大学学报(自然科学版) V01．17 N0．1 2008年1月 Journal of Yunnan Nationalities University(Natural Sciences Edition) Jan．2008 基于支持向量机与移动Agent的入侵检测系统模型 谷 雨 周 波 戴燕翎 (1．云南民族大学教育学院，云南 昆明650031； 2．云南科技信息职业学院，云南昆明650106；3．云南省文山州电信分公司，云南文山663000) 摘 要 目前要发展新的入侵检测系统必须解决检测准确性、高效性的问题，同时要考虑分布式智能化的检测方法．提 出一种采用基于支持向量机与移动Agent技术的入侵检测系统模型，利用支持向量机对小样本、高维非线性数据良好的分类 性能，将其作为检测工具；利用移动Agent的智能性、移动性，在网络节点间进行迁移检测入侵．给出了相应的模型结构． 关键词 人侵检测系统；支持向量机；移动Agent；A~et 【中图分类号]TP393．08 【文献标识码】A 【文章编号】1672--8513(2008)01—0068—04 Intrusion Detection System Model Based on Support Vector Machine and Mobile Agent Gu Yu ，Zhou Bo ，Dai Yanling3 (1．School of Education，Yunnan Nationalities University，Kunming 65003 1，China； 2．Yunnan Institute of Technology and Information，Kunming 650106，China； 3．Wenshan Branch，Yunnan Telecom Company Limited，Wenshan 663000，China) Abstract：To develop a new IDS，the problem of detecting accuracy and efficienly need to be solved，and in- tellectualized distributed detecting method should be put into consideration．In this paper，we bring forward an IDS based on Suppo~Vector Machine and mobile Agent technology．It makes use of SVM as a detecting tool because of its good classification performance when dealing with problems of hi曲dimension nonlinear data with small sample set and hish dimension non—linear data．Intellectuality and transferability of mobile Agent is utilized to move and detect intrusion among network nodes．A configuration of system model based on SVM and mobile Agent was pres- ented，and some functional model design and system environment were introduced． Key words：intrusion detection system；suppo~vector machine；mobile Agent；Aglet 既有对新的智能