机构地区: 中南大学信息科学与工程学院
出 处: 《通信学报》 2005年第4期24-29,共6页
摘 要: 提出了一种用于网络测量的动态时间驱动分组采样技术,根据网络流量的变化动态调整采样次数,从而能将最后的测量误差率控制在一个给定的范围内,在保证精度的同时极大地降低了测量的代价。并且该方法的实现复杂度与“1out of N”采样技术近似,文章中给出了有关的理论分析和实验研究。实验结果表明该方法达到了设计的目标。根据网络运行状态动态调整网络测量方案的思想可以广泛用于网络测量当中。 Proposed a dynamic time-driven packet sampling technique. This dynamic packet sampling could vary sampling degree with the change of network traffic, thus guarantee the error rate under a given level. In this way, high accuracy could be got at a low cost of network measurement. Moreover, this packet sampling technique was nearly as simple as “1 out of N”, theoretic analysis and experimental study were given. Experiment results show that this technique achieves design object. The idea of dynamic changing measurement scheme according to network state can be widely used in network measurement.
领 域: [自动化与计算机技术] [自动化与计算机技术]