摘要
防范DDo S攻击的关键在于攻击源的定位,包标记是攻击源定位技术研究的热点。针对传统概率包标记存在的问题,提出选择IPv4数据报首部的选项字段作为标记区域,采用Huffman编码压缩标记信息,减少路径重构时所需标记包的数量;利用IPv6的隧道模式,在IPv4到IPv6网络时增加一个复制操作,将标记信息转存到IPv6的hop-by-hop字段,增加改进算法的适用范围。实验结果表明,改进算法快速、准确和高效,只需一个数据报即可完成路径重构,适用于IPv4和IPv6网络。
The key to prevent DDoS attacks is locating attack source, and packet marking is the hot spot of attack source locating technology. Aiming at the problems of packet marking, an improved algorithm is proposed. The improved algorithm chooses option field of IPv4 datagram header as the marking area and uses Huffman code to reduce the number of marked packets during path reconstruction. Packets pass from IPv4 network to IPv6 network, adding a copy operation to copy marking information to IPv6 extension header of hop - by - hop. Thus, it increases the application scope. The experimental results show that the improved algorithm is rapid, accurate and efficient. It can complete path reconstruction only needing a datagram, which can be applied to IPv4 and IPv6 network.
出处
《河池学院学报》
2015年第5期74-78,共5页
Journal of Hechi University
基金
广西教育厅高校科研项目(YB2014325)
河池学院青年科研课题(2013A-N002)
河池学院"计算机网络与软件新技术"重点实验室资助项目(院科研〔2013〕3号)