基于防空作战体系的DDoS攻击防御技术研究  

Research on DDoS attack defense technology based on air defense combat system

作　　者：刘顺余 唐强 沙智伦 Liu Shunyu;Tang Qiang;Sha Zhilun(The Chinese People's Liberation Army 31635 Troops,Guangxi Guilin 541100)

机构地区：[1]中国人民解放军31635部队,广西桂林541100

出　　处：《网络空间安全》2020年第6期38-42,共5页Cyberspace Security

摘　　要：防空作战体系依托军事综合信息网、指挥专网、内部局域网等网络,连接各级作战值班室和作战阵地并实现空情共享.这类网络任务单一,防御能力较弱,易遭受敌军攻击,如分布式拒绝服务(DDoS).基于此,文章中提出了一种新颖的入侵检测方法—KNS.这种方法基于集成学习的思想,首先分别采用K-最近邻(K-Nearest Neighbor,KNN)、朴素贝叶斯分类器(Naive Bayes Classifier,NBC)和支持向量机(Support Vector Machine,SVM)对流量进行检测,其次对检测结果进行投票策略(Voting)整合,最后获得KNS的最终检测结果.这种方法在DDoS数据集进行了测试,结果表明,KNS具有较好的异常检测检测准确性、检测率、误报率.The air defense combat system relies on military integrated information networks,command special networks,internal LANs,and other networks to connect combat duty rooms and combat positions at all levels and achieve the sharing of air intelligence.This type of network has a single task,weak defense capabilities,and is vulnerable to enemy attacks-distributed denial of service(DDoS).In this paper,a novel intrusion detection method,which named KNS,is proposed.Based on the idea of ensemble learning,this method first uses K-Nearest Neighbor(KNN),Naive Bayes Classifier(NBC)and Support Vector Machine(SVM)to perform traffic Anomaly detection.Secondly,a voting strategy(Voting)is integrated on the test results,and finally KNS is obtained,and the final test results are obtained.This method has been tested on the DDoS data set,and the results show that KNS has good anomaly detection accuracy,detection rate,and false alarm rate.

关 键 词：防空作战体系 机器学习 入侵检测 分布式拒绝服务 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]