IPV6环境下的高维大规模包匹配算法

发布时间：2018-04-08 15:17

本文选题：差分演化　切入点：变异系数　出处：《电子学报》2013年11期

【摘要】：传统的包匹配算法不是无法运用于IPV6环境,就是性能太差.本文把基于实数编码的差分演化算法与传统的包匹配算法相融合.在适应值设计上引入变异系数的思想,从而使问题的处理更具有客观性.通过引入分布性特征,自适应调整变异的剧烈程度,从而动态权衡种群的多样性和收敛性之间的矛盾.数值实验表明此算法与传统算法相比,在速度、存储空间等综合性能上得到有效改善,另外本文提出的算法还有一个显著特点:包匹配的时间性能与规则数目之间具有很弱的相关性,从而本算法适合处理高维和大规模包匹配问题.本算法运用到IPV6网络,使数据包能快速转发.而且本文提出的方法具有普适性,适用于防火墙、路由器等网络设备.
[Abstract]:The traditional packet matching algorithm is either unable to be used in IPV6 environment, or its performance is too poor.In this paper, the differential evolution algorithm based on real number coding is combined with the traditional packet matching algorithm.The idea of coefficient of variation is introduced into the design of fitness value, so that the treatment of the problem is more objective.By introducing the distributive characteristics, the intensity of the variation can be adjusted adaptively, and the contradiction between the diversity and convergence of the population can be weighed dynamically.Numerical experiments show that compared with the traditional algorithm, the proposed algorithm can effectively improve the performance of the algorithm, such as speed, storage space, and so on.In addition, the algorithm proposed in this paper also has a remarkable feature: the performance of packet matching has a very weak correlation with the number of rules, so this algorithm is suitable for dealing with high dimensional and large scale packet matching problems.This algorithm is applied to IPV6 network, so that packets can be forwarded quickly.The method proposed in this paper is universal and suitable for network devices such as firewalls and routers.
【作者单位】：武汉大学计算机学院;南通大学计算机科学与技术学院;
【基金】：国家自然科学基金(No.61070008) 教育部人文社科项目(No.11YJC870012) 教育部人文社科基金(No.12YJCZH274) 江苏省自然科学基金(No.BK2010277)
【分类号】：TP393.01

【参考文献】