计算机应用 ›› 2010, Vol. 30 ›› Issue (05): 1198-1201.

• 信息安全 • 上一篇    下一篇

基于自适应直觉模糊推理的入侵检测方法

黄孝文1,张弛2   

  1. 1. 陕西三原空军工程大学导弹学院
    2. 空军工程大学导弹学院
  • 收稿日期:2009-11-30 修回日期:2009-12-28 发布日期:2010-05-04 出版日期:2010-05-01
  • 通讯作者: 黄孝文
  • 基金资助:
    国家自然科学基金资助项目;陕西省自然科学基金资助项目

Techniques for intrusion detection based on adaptive intuitionistic fuzzy reasoning

  • Received:2009-11-30 Revised:2009-12-28 Online:2010-05-04 Published:2010-05-01
  • Supported by:
    the National Natural Science Foundation of China under Grant

摘要: 将直觉模糊集理论引入信息安全领域,提出一种基于自适应直觉模糊推理的入侵检测方法。首先,分析现有入侵检测方法的特点与局限性,建立基于自适应神经—直觉模糊推理系统(ANIFIS)的Takagi-Sugeno型入侵检测模型。其次,设计系统的推理规则,确定各层输入输出的计算关系,以及系统输出结果的计算表达式。再次,设计网络学习算法,对网络结构进行调节以及对网络参数进行学习。最后,选择KDD CUP 99入侵检测数据集作为样本集,获得相应的检测结果,验证了方法的有效性和模型的正确性。

关键词: 信息安全, 入侵检测, 直觉模糊集, 自适应, 神经网络

Abstract: A technique for intrusion detection based on Adaptive Neuro-Intuitionistic Fuzzy Inference System (ANIFIS) was proposed with intuitionistic fuzzy set theory introduced into the area of information security. First, the properties and vulnerabilities of the existing intrusion detection methods were analyzed, and a model for intrusion detection on ANIFIS with Takagi-Sugeno type was established. Then, the inference rules of the system were devised, computational relations between layers of input and output and a synthesized computational expression of system output were ascertained. Subsequently, a learning algorithm of neural network was proposed, which could adjust network structure and study network parameters. Finally, the test results verify the validity of the technique and rationality of constructed model by providing intrusion detection instances with KDD CUP 99 dataset.

Key words: information security, intrusion detection, intuitionistic fuzzy set, self-adaptation, neural network