基于对象特征的软件定义网络分布式拒绝服务攻击检测方法

发布时间：2018-04-10 07:24

本文选题：软件定义网络　切入点：元组　出处：《电子与信息学报》2017年02期

【摘要】：软件定义网络(SDN)受到分布式拒绝服务(DDoS)攻击时,攻击方会发送大量数据包,产生大量新的终端标识占用网络连接资源,影响网络正常运转。为准确发现受攻击对象,检测被占用资源,利用GHSOM技术,该文提出基于对象特征的DDoS攻击检测方法。首先,结合SDN网络及攻击特点,提出基于目的地址的检测7元组,并以此作为判断目标地址是否受到DDoS攻击的检测元素;然后,采用模块化设计,将GHSOM算法应用于SDN网络DDoS攻击的分析检测中,并在OpenDayLight的仿真平台上完成了仿真实验。实验结果显示,该文提出的检测7元组可有效检测目标对象是否受到DDoS攻击。
[Abstract]:When the software defined network (SDN) is attacked by the distributed denial of service (DDoS), the attacker will send a large number of packets, resulting in a large number of new terminal identifiers occupying the network connection resources, which will affect the normal operation of the network.In order to accurately find the attacked object and detect the occupied resources, this paper proposes a DDoS attack detection method based on object characteristics using GHSOM technology.First of all, combined with the characteristics of SDN network and attack, a 7 tuple detection based on destination address is put forward, which is used as the detection element to judge whether the target address is attacked by DDoS, and then the modularization design is used to determine whether the target address is attacked by DDoS.The GHSOM algorithm is applied to the analysis and detection of DDoS attack in SDN network, and the simulation experiment is completed on the OpenDayLight simulation platform.Experimental results show that the proposed 7 tuple detection can effectively detect whether the target object is attacked by DDoS.
【作者单位】：北京交通大学电子信息工程学院;
【基金】：国家973重点基础研究发展计划(2013CB329100) 国家863高技术研究发展计划(2015AA016103) 国家自然科学基金(61301081) 国家电网公司科技项目([2016]377)~~
【分类号】：TP393.08

【相似文献】