论文部分内容阅读
随着Internet的发展,下一代互联网IPv6必然会最终代替目前的IPv4网络。相对IPv4而言,IPv6网络只是在网络层进行了比较大的改进,因此大多数网络安全问题对于IPV4和IPv6是相同的,例如DOS攻击、中间人攻击等。因此,IPv6网络安全形势同样不容乐观。作为教育网会员单位,作者所在单位于2011年接入CERNET2网络。本文基于Wireshark进行IPv6数据包的捕获解析并存储,然后使用Matlab聚类工具箱中的K均值算法和神经网络工具箱中的SOM算法,分别对包含多类攻击数据的IPv6流量进行处理从而实现了对于CERNET2网络的异常流量聚类识别。实验表明,本系统能够识别发生IPv6网络中的DOS攻击等几类针对ICMPv6的攻击,加强了校园网络的安全。
With the development of the Internet, the next generation of Internet IPv6 will eventually replace the current IPv4 network. Compared with IPv4, IPv6 network only makes a big improvement at the network layer. Therefore, most network security problems are the same for IPV4 and IPv6, such as DOS attacks, man-in-the-middle attacks, and the like. Therefore, the IPv6 network security situation is equally unsightly. As a member of Education Network, the author’s unit accessed CERNET2 network in 2011. In this paper, based on Wireshark, IPv6 packets are captured and parsed and stored. Then the K-means algorithm in Matlab clustering toolkit and the SOM algorithm in neural network toolbox are used respectively to process IPv6 traffic with multiple types of attack data, Abnormal traffic clustering identification for CERNET2 network. Experiments show that this system can identify several kinds of attacks against ICMPv6 such as DOS attack in IPv6 network and strengthen the security of campus network.