粗集神经网络在网络入侵中的应用研究

摘要

研究网络安全问题.网络入侵具有多样性、不确定性,收集数据包含大量冗余信息,传统网络入侵检测算法无法消除冗余消息,导致网络入侵检测的准确率低.为了提高网络安全性,提出了一种粗集神经网络的网络入侵检测算法.算法利用粗集理论对入侵样本数据属性约简,将不完整数据剔除,消除冗余信息,然后将约简后的数据输入BP神经网络,通过BP神经网络非线性学习能力,在输入与输出之间建立一种非线性映射关系,识别出网络入侵类型.仿真结果表明,相对于传统网络人侵检测算法,粗集BP神经网络不仅提高了网络检测的正确率,降低了误报率、漏报率,同时加快了网络入侵的检测速度,是一种有效、实时的网络入侵检测算法.%Research on network intrusion detection problem.Because network intrusion data samples have incomplete, uncertainty and redundant information, traditional methods cannot deal with these data effectively, resulting in low intrusion detection accuracy.In order to improve the accuracy of network intrusion detection, this paper presents network intrusion detection method based on rough set and neural network, using rough sets theory to reduce the sample data attribute, eliminate redundant information, then the data is used as the BP neural network input.Through the BP neural network training, non-linear relation between input and output is established.Simulation results show that, compared with traditional BP neural network, the proposed method accelerates the network intrusion detection speed, improves the detection accuracy, reduces the rate of false positives.It is an effective network intrusion detection model.