摘要
在研究通信网络的流量控制及网络性能的时候,业务源模型是一个十分重要的组成部分。传统的业务流模型大多是基于泊松或贝努力过程的,而这些模型表现出的行为与实际网络的测量结果不符。大量业务的测量表明在这些业务中普遍存在着自相似性(即长相关性)。采用自相似建模表征业务到达过程的长时相关特性具有较高的精度,其中Hurst系数是表征业务突发特性的重要参数。经研究发现,无线网络中仍然存在自相似性。考察CDPD(蜂窝数字分组数据网)的特性,发现自相似性对网络的排队性能有着显著的影响,可以通过增加队列长度,加大缓冲存储空间,实现降低延时。
Traffic models play a very important part when studing the flow control problems and performance of communication networks. The conventional models are mostly based on Poisson model or Bernoulli process, but the activity of these models is not consistent with the measured results of the real networks. Many measurement on traffic have shown that there exist intrincately self-similar or long-range dependence(LRD) characteristics in the traffic. The long-range correlation as self-similar process can be efficiently modeled, and Hurst index is the key value of this model representing the burstiness of traffic source. After some research ,we can conclude that the self-similar performance also exist in wireless networks, and the self-similarity is important to the queuing analysis for communication networks. Enlarging the length of the queue and the buffer can reduce the queuing delay.
出处
《微计算机信息》
北大核心
2005年第1期193-194,共2页
Control & Automation
