| 基于SDN的DDoS攻击检测技术研究 |
| |
| 引用本文： | 张强强,李永忠.基于SDN的DDoS攻击检测技术研究[J].教育技术导刊,2019,18(7):205-208. |
| |
| 作者姓名： | 张强强 李永忠 |
| |
| 作者单位： | 江苏科技大学 计算机学院， 江苏 镇江 212003 |
| |
| 摘 要： | 为了在保证检测准确率的前提下提高检测效率，并优化SDN网络中基于流表特征的DDoS攻击检测算法，主要分析基于流表特征的DDoS攻击检测技术及其存在的不足，提出首先利用主成分分析优化流表特征，从中选出合适的特征子集，并采用支持向量机算法实现分类检测；然后搭建仿真网络环境，利用正常数据集与攻击数据集训练分类器进行测试实验；最后从检测准确率与检测时间两个维度对特征降维前后的检测方法进行对比。实验结果表明，经过特征降维的检测方法在不影响准确率的同时，有效提高了检测速率。
|
| 关 键 词： | 软件定义网络 DDoS攻击 主成分分析 支持向量机 |
| 收稿时间： | 2018-11-07 |
Research on DDoS Attack Detection Technology Based on SDN |
| |
| ZHANG Qiang-qiang,LI Yong-zhong.Research on DDoS Attack Detection Technology Based on SDN[J].Introduction of Educational Technology,2019,18(7):205-208. |
| |
| Authors: | ZHANG Qiang-qiang LI Yong-zhong |
| |
| Institution: | College of Computer， Jiangsu University of Science and Technology，Zhenjiang 211003，China |
| |
| Abstract: | In order to improve detection efficiency and optimize DDoS attack detection algorithm based on stream table characteristics in SDN network on the premise of guaranteeing detection accuracy， this paper mainly analyses DDoS attack detection technology based on stream table characteristics and its shortcomings， and proposes to optimize flow table features by using principal component analysis（PCA）， and select appropriate features from them. and support vector machine（SVM） algorithm is applied to realize classification detection. By building a simulated network environment， the classifier is trained with normal data sets and attack data sets and tested. Finally， the detection methods before and after feature dimensionality reduction are compared from two dimensions of detection accuracy and detection time. The results show that the detection method after feature dimensionality reduction improves the detection speed without affecting the accuracy. |
| |
| Keywords: | software defined network DDoS attack principal component analysis support vector machine |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
