| 人工神经网络在ATM链路容量控制中的应用 |
| |
| 引用本文： | 黄英,郑君里.人工神经网络在ATM链路容量控制中的应用[J].清华大学学报(自然科学版),1997(1). |
| |
| 作者姓名： | 黄英 郑君里 |
| |
| 作者单位： | 清华大学电子工程系 |
| |
| 摘 要： | 研究人工神经网络（ＡＮＮ）在异步转移模式（ＡＴＭ）链路容量分配中的应用，设计了神经网络预测器以解决ＡＴＭ链路容量分配中的困难。与其它方法相比较，该方法不需对网络进行复杂的数学分析，借助足够的学习样本，即可自动学习输入和输出之间的关系，预测网络性能。文中分别采用呼损率和链路利用率两种优化目标进行计算机模拟，结果表明该方法能正确完成容量分配，并且易于实现和推广。
|
| 关 键 词： | 人工神经网络（ANN） 异步转移模式（ATM） 虚通道VP 链路容量分配 |
Application of artificial neural network in the link capacity control of ATM |
| |
| Huang Ying,Zheng Junli.Application of artificial neural network in the link capacity control of ATM[J].Journal of Tsinghua University(Science and Technology),1997(1). |
| |
| Authors: | Huang Ying Zheng Junli |
| |
| Institution: | Huang Ying,Zheng Junli Department of Electronic Engineering,Tsinghua University,Beijing 100084 |
| |
| Abstract: | This paper explores ANN's (artificial neural network) application in the link capacity control of ATM (asynchronous transfer mode) and proposes an ANN predictor to solve the problem encountered in ATM network. Compared with other methods, the new way does not need any complicated mathematical calculations, it can learn the relationship between input and output patterns automatically and predict the performance of the network. Computer simulation results using two optimization object methods show that the ANN way can assign the link capacity correctly and is easy to implement and generalize. |
| |
| Keywords: | artificial neural network (ANN) asynchronous transfer mode (ATM) virtual path link capacity control |
| 本文献已被 CNKI 等数据库收录！ |
