引用本文
  • 龚宇翔,曹进,付玉龙,郭敏.针对LTE-A网络中的DDoS攻击流量检测模型[J].信息安全学报,2019,4(1):27-38    [点击复制]
  • GONG Yuxiang,CAO Jin,FU Yulong,GUO Min.A DDoS attack detection model for LTE-A network[J].Journal of Cyber Security,2019,4(1):27-38   [点击复制]
【打印本页】 【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

←前一篇|后一篇→

过刊浏览    高级检索

本文已被:浏览 6885次   下载 5664 本文二维码信息
码上扫一扫!
针对LTE-A网络中的DDoS攻击流量检测模型
龚宇翔1, 曹进1, 付玉龙1, 郭敏2
0
(1.西安电子科技大学 网络与信息安全学院 西安 中国 710126;2.北京计算机技术及应用研究所 北京 中国 100854)
摘要:
近年来,4G LTE-A技术发展迅猛,移动设备的普及以及各种承载于4G网络的业务和应用已经成为我们日常不可或缺的部分。但网络攻击技术也不断的在发展,特别是近年来针对4G LTE-A网络的攻击技术的不断演进,已成为危害人们切身利益的关键问题。DDoS作为DoS攻击的一种,对网络带来了更大的危害,因此需要研究一种攻击检测模型。文章提出了一个针对LTE-A网络中的DDoS攻击流量检测模型,模型利用熵作为特征之一,并使用随机森林算法训练模型分类器,可将其部署在eNB上对流经该eNB的DDoS流量进行识别。通过验证,所提出的模型的检测准确率可达99.956%。
关键词:  机器学习  随机森林  DDoS  LTE网络  
DOI:10.19363/J.cnki.cn10-1380/tn.2019.01.03
投稿时间:2018-09-29修订日期:2018-11-07
基金项目:本课题得到国家重点研发计划项目(No.2016YFB0800700)与国家自然科学基金项目(No.61772404,No.61602359)资助
A DDoS attack detection model for LTE-A network
GONG Yuxiang1, CAO Jin1, FU Yulong1, GUO Min2
(1.School of Cyber Engineering, XiDian University, Xi'an 710126, China;2.Beijing Computer Technology and Application Institute, Beijing 100854, China)
Abstract:
In recent years, 4G LTE-A technology has developed rapidly,and the popularity of mobile devices and various services based on 4G networks have become an indispensables part of our daily life. However, attack means is also constantly developing. The continuous evolution of attack means for 4G LTE-A networks in recent years has become a key issue that threatens our legal right. DDoS is a kind of denial of service attack, which brings more harm. Therefore, it is necessary to study an attack detection model. In this paper,a DDoS attack detection model for LTE-A network has been proposed. The model uses entropy as one of the features and uses random forest algorithm to train a classifier which can be equipped on an eNB to recognize the DDoS flow through the eNB. The experiment result shows that the detection accuracy of the proposed model can reach to 99.956%.
Key words:  machine learning  random forest  DDoS  LTE network  entropy