模糊网络入侵中多层序列特征自动提取方法研究

发布时间：2018-06-25 17:30

  本文选题：模糊网络 + 入侵检测　； 参考：《现代电子技术》2017年10期

【摘要】：模糊网络中入侵特征较为多样化,无法通过固定的阈值进行合理判断。为了解决模糊网络入侵检测方法存在检测率低、误报率高和检测速度慢等问题,提出一种基于量子神经网络的层序列特征自动提取方法。在该算法中,通过对模糊网络进行层次划分,运用量子BP神经网络模型以量子形式形态的空间思维结构来提取信息,通过量子空间结构中量子门的移位与旋转变化对神经网络量子形态相位进行操作,完成多层序列特征自动提取。仿真实验表明,该算法具有较好高的检测率和检测效率,并且误报率较低。
[Abstract]:The intrusion features in fuzzy networks are diverse and can not be reasonably judged by fixed thresholds. In order to solve the problems of low detection rate, high false alarm rate and slow detection speed in fuzzy network intrusion detection method, a layer sequence feature automatic extraction method based on quantum neural network is proposed. In this algorithm, the fuzzy network is divided into layers, and the quantum BP neural network model is used to extract the information from the spatial thinking structure of the quantum form. The phase of quantum morphology in neural network is operated by the shift and rotation of quantum gate in quantum space structure, and the multi-layer sequence feature is automatically extracted. Simulation results show that the algorithm has high detection rate and detection efficiency and low false alarm rate.
【作者单位】： 武汉大学经济与管理学院;义乌工商职业技术学院机电信息学院;
【基金】：浙江省2015年度高等教育教学改革项目(JG2015343)
【分类号】：TP393.08
，

本文编号：2066922