Abstract: The paper presents a new intrusion detection system based on modified ART-2 neural network, adopts new non-linear transfer function and competitive learning algorithm to solve the problem that the same phasic data with different amplitude can’t be distinguished by the standard ART-2 neural network and make it more suitable for intrusion detection system. It can raise the accuracy of detection and the ability of adaptively detecting unknown intrusion.

摘要： 提出了一种基于改进的ART-2神经网络的入侵检测系统,采用新的非线性变换函数和竞争层学习算法克服了ART-2神经网络的“预处理信号畸变”和“同相位不可分”问题,使之更适用于入侵检测,提高了检测的准确率和对未知入侵的自适应识别能力。

关键词: 入侵检测系统, 自适应共振, 神经网络

CLC Number: 

• TP393.08