CAJ | 학술논문

虽然目前在危害Internet的安全中，分布式拒绝服务即DDOS已经成为重要的隐患之一，但是由于对DDOS攻击进行防范、检测和反击的研究工作还没有取得实质性的重大突破，因此，目前还没有一种有效的方法来准确及时的预测DDOS攻击的发生。本文提出一种根据网络业务自相似性，通过实时建模和动态分析检测拒绝服务攻击发生前后网络流量自相似性参数Hurst的变化来检测DDoS攻击。通过仿真试验，该方法能够快速准确的检测出DDoS攻击的发生。
수연목전재위해Internet적안전중，분포식거절복무즉DDOS이경성위중요적은환지일，단시유우대DDOS공격진행방범、검측화반격적연구공작환몰유취득실질성적중대돌파，인차，목전환몰유일충유효적방법래준학급시적예측DDOS공격적발생。본문제출일충근거망락업무자상사성，통과실시건모화동태분석검측거절복무공격발생전후망락류량자상사성삼수Hurst적변화래검측DDoS공격。통과방진시험，해방법능구쾌속준학적검측출DDoS공격적발생。
Although the harmfulness of Internet security, distributed denial of service DDOS has become one of the important potential danger, but because of the work of prevention, detection and response to DDOS attacks have not achieved a major breakthrough, there is not an efficient way to predict DDOS attacks accurately and timely. This paper presents a self similarity based on network traffic, by real-time modeling and dynamic analysis to detect denial of service attacks occurred before and after the self similar network traffic changes of parameters of Hurst to detect DDoS attacks. Through the simulation test, the method can quickly and accurately detect the occurrence of DDoS attack.