【作者】 苏金泷； 欧阳钟辉；

【Author】 SU Jin-long~(1,2) OUYANG Zhong-hui~1 1 TSL School of Business and Information Technology,Quanzhou Normal University,Quanzhou 362000,China 2 School of Computer Engineering and Science,Shanghai University,Shanghai 200072,China

【机构】 泉州师范学院陈守仁工商信息学院；

【摘要】 不断增加的网络攻击将造成由泄密最终导致经济损失的不同程度的入侵伤害,因此信息安全建设成为大信息平台构建过程中不可缺少的部分。入侵检测设计一度成为研究小组的主要难题之一,这是由于不断增加的各类子网造成的复杂的网络异构性,无形中增加了计算机系统被从内部或外部的越权入侵的可能性。研究小组考察了现有入侵检测系统,引入了 FNN 算法,构造了一种基于 FNN 的入侵检测分类算法。该算法具有检测大部分入侵的潜力,以及较少的误报率,实际使用中已显示了该方法的有效性。更多还原

【Abstract】 For the increasing onslaught of computer attacks,which cause damage ranging from mere viola- tion of confidentiality and issues of privacy up to actual financial losses,the information security infrastructure has become an indispensable part on Comprehensive Information Platform design.In a time,the intrusion de- tection design was the most challenging task for us due to the proliferation of heterogeneous computer networks for the mounting up connectivity of computer systems with greater access to outsiders and makes it flimsiness for intruders to avoid identification.The research group investigated the existing intrusion detection models and fi- nally selected a kind of FNN-Base (fuzzy neural network) as a measure for building a goal oriented intrusion detection classifier.Practical performance shows that it is an effective approach with potential to detect all in- trusions and acceptable false alarm rate in our application environment.更多还原