摘要
入侵检测是网络安全的一个重要组成部分。针对常用基于长短期记忆网络的入侵检测算法的收敛速度慢 的问题,考虑到网络流量数据同时存在时间和空间的特征,提出了一种基于卷积神经网络和门控循环单元的入侵 检测方法。使用卷积神经网络将流量数据的空间特征提取出来,以多个相同参数的小卷积核来替代大卷积核,使 得网络结构加深。将提取出的特征输入到门控循环单元中,学习流量数据的时序特征后进行分类。数据集采用 KDD99,实验表明,该模型能够对网络流量进行有效的识别。
Intrusion detection is an important part of network security.Aiming at the problem of slow convergence of com⁃monly used intrusion detection algorithms based on long and short-term memory networks,considering that the network traf⁃fic data have both time and space characteristics,an intrusion detection method based on convolutional neural networks and gated recurrent units is proposed.Use convolutional neural network to extract the spatial characteristics of the traffic data,and replace the large convolution kernel with multiple small convolution kernels with the same parameters to deepen the net⁃work structure.Input the extracted features into the gated recurrent unit,and classify after learning the time series features of the traffic data.The data set uses KDD99,and experiments show that this model can effectively identify network traffic.
作者
张行健
王怀彬
ZHANG Xingjian;WANG Huaibin(Laboratory of Network Information Security,School of Computer Science and Engineering,Tianjin University of Technology,Tianjin 300384,China)
出处
《天津理工大学学报》
2022年第1期37-42,共6页
Journal of Tianjin University of Technology
基金
国家自然科学基金(61773286)。
关键词
入侵检测
卷积神经网络
门控循环单元
intrusion detection
convolutional neural network
gated recurrent unit
