基于卷积神经网络的拒绝服务攻击数据流检测  

Denial-of-Service Attack Data Flow Detection Based on Convolutional Neural Network

作　　者：谢洁 韩德志[1] XIE Jie;HAN De-zhi(College of Information Engineering,Shanghai Maritime University,Shanghai 201306)

机构地区：[1]上海海事大学信息工程学院,上海201306

出　　处：《现代计算机（中旬刊）》2018年第9期43-47,共5页Modern Computer

摘　　要：卷积神经网络在相邻层中加入局部连接的卷积操作,用来提取数据的特征。利用卷积神经网络的这种特性,丰富云环境中拒绝服务攻击数据流的特征信息,将数据流通过训练后的卷积神经网络模型,可以对云环境中的拒绝服务攻击数据流进行初步分类检测。提出一种基于卷积神经网络的拒绝服务攻击数据流检测模型,将粒子群算法应用于卷积神经网络的学习,实验结果表明,检测模型的拒绝服务攻击数据流的平均检测精确度达到98%。Convolutional neural networks add locally concatenated convolutions in adjacent layers to extract data features. Uses the characteristics of convolutional neural networks to enrich the characteristic information of denial of service attack data flow in cloud environment. Through the trained convolutional neural network model, we can initially perform data denial of service attack in cloud environment. Classificationtest. Proposes a data flow detection model for denial of service attacks based on convolutional neural networks. The particle swarm algorithm is applied to the learning of convolutional neural networks. The experimental results show that the average detection accuracy of the denial of service attack data flow detection model is 98%.

关 键 词：云环境 拒绝服务攻击 卷积神经网络 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]