面向拒绝服务攻击的多标签IP返回追踪新方法

发布时间：2018-06-19 05:47

本文选题：网络安全 + IP返回追踪　；参考：《西安交通大学学报》2013年10期

【摘要】：针对网络安全中拒绝服务攻击难以防御的特点,提出面向拒绝服务攻击的多标签IP返回追踪方法(iTrace-DPPM),用以识别基于互联网控制报文协议(ICMP)的直接和反射式拒绝服务攻击的真实源地址。该方法首先结合ICMP数据段大小及最大传输单元阀值,计算单一ICMP数据报文可携带的路由标记数,再根据数据包的幸存时间推断路由器与攻击源头的距离,将路由器的标记概率设定为距离的倒数,并针对每个标记域独立地执行概率标记算法,最后受害目标根据接收的标记信息,实现转发路径的重构及源头识别。与已有的动态概率包标记方法相比,iTrace-DPPM方法具有路径重构所需数据包少、支持部分部署及无额外负载的优点。NS2环境下的模拟实验结果证实,路径重构所需的攻击包数降为DPPM方法的路由标记数的倒数。
[Abstract]:In view of the characteristics that denial of service attack in network security is difficult to defend, A multi-label IP return tracking method for denial of service attack (DoS) is proposed to identify the real source addresses of direct and reflective denial-of-service attacks based on ICMP-based Internet Control packet Protocol (ICMPP). The method first calculates the number of routing markers that can be carried by a single ICMP data packet, combining the size of the ICMP data segment and the maximum transmission unit threshold, and then extrapolates the distance between the router and the source of the attack based on the survival time of the packet. The marking probability of the router is set to the reciprocal of the distance, and the probability marking algorithm is implemented independently for each tag domain. Finally, the victim target reconstructs the forwarding path and identifies the source according to the received tag information. Compared with the existing dynamic probabilistic packet marking method, the proposed method has fewer packets needed for path reconfiguration, and supports partial deployment and no additional load. The number of attack packets needed for path reconstruction is reduced to the reciprocal of the number of routing markers of the DPPM method.
【作者单位】：西安交通大学制造系统工程国家重点实验室;上海交通大学信息安全工程学院;上海交通大学电子信息与电气工程学院;
【基金】：国家科技支撑计划资助项目(2012BAH38B04) 国家自然科学基金资助项目(61271316) 机械制造系统工程国家重点实验室开放基金资助项目(sklms 2012005)
【分类号】：TP393.08

【参考文献】