| RED算法的随机模型和仿真 |
| |
| 引用本文： | 罗超,陈宗海,薛福珍.RED算法的随机模型和仿真[J].中国科学技术大学学报,2005,35(6):947-953. |
| |
| 作者姓名： | 罗超 陈宗海 薛福珍 |
| |
| 作者单位： | 中国科学技术大学自动化系,安徽,合肥,230027 |
| |
| 摘 要： | 运用随机分析的方法建立了一个关于RED(random early detection)队列长度及其指数加权滑动平均值(exponentially weighted moving average,EWMA)的随机模型.对该模型的动态仿真和数值分析表明,它能够很好地预测RED在忙期的动态行为.利用这个模型,RED的参数可以通过简单的数值分析方法进行设置,避免了经验方法在链路特性参数和流量参数变化时所出现的性能问题.
|
| 关 键 词： | 拥塞控制 队列长度 |
| 文章编号： | 0253-2778（2005）06-0947-07 |
| 收稿时间： | 2003-07-25 |
| 修稿时间： | 2005-02-02 |
Stochastic Modeling and Simulations of the RED Algorithm |
| |
| LUO Chao,CHEN Zong-hai,XUE Fu-zhen.Stochastic Modeling and Simulations of the RED Algorithm[J].Journal of University of Science and Technology of China,2005,35(6):947-953. |
| |
| Authors: | LUO Chao CHEN Zong-hai XUE Fu-zhen |
| |
| Institution: | Department of Automation, University of Science and Technology of China, Hefei 230027, China |
| |
| Abstract: | A stochastic model of the random early detection(RED) queue length and its exponentially weighted moving average (EWMA) value is developed based on stochastic analysis. By comparing research between dynamic simulation and numerical analysis, it has been verified that this model can predict the busy-period dynamics of RED very well, which makes it possible to set up RED parameters by simple numerical analysis, thus avoiding the performance problems that occur when links and traffic parameters oscillate and empirical methods are used. |
| |
| Keywords: | RED EWMA random early detection (RED) congestion control queue length exponentially weighted moving average (EWMA) |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
