| 引用本文: | 丁 锋, 萧德云, 丁 韬.衰减激励条件下最小均方算法的收敛性[J].控制理论与应用,2003,20(1):109~112.[点击复制] |
| DING Feng, XIAO De-yun, DING Tao.Convergence of least mean squares algorithm under attenuating excitation conditions[J].Control Theory and Technology,2003,20(1):109~112.[点击复制] |
|
| 衰减激励条件下最小均方算法的收敛性 |
| Convergence of least mean squares algorithm under attenuating excitation conditions |
| 摘要点击 1382 全文点击 1221 投稿时间：2000-08-28 修订日期：2001-05-21 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 |
| 2003,20(1):109-112 |
| 中文关键词 参数估计 辨识 衰减激励 最小均方算法 |
| 英文关键词 parameter estimation identification attenuating excitation least mean square algorithm |
| 基金项目 国家自然科学基金(60074029); 国家自然科学基金重点项目(69934010); 清华大学信息学院创新基金资助项目 |
|
| 中文摘要 |
| 给出了衰减激励信号的定义,并在衰减激励条件下,利用随机过程理论,研究了随机系统最小均方算法的收敛速率,阐述了参数估计误差收敛时,衰减指数和算法中设计参变量 (收敛因子或步长 )的选择方法.分析表明:在衰减激励条件下,最小均方算法也具有良好的性能:当衰减指数和设计参变量满足一定条件时,则参数估计误差一致收敛于零. |
| 英文摘要 |
| The definition of attenuating excitation signal is given, and the convergence rate of the least mean square algorithm, using stochastic process theory, is studied for stochastic systems under attenuating excitation. The way to choose the attenuating index and design variable (convergent factor or stepsize) is stated for guaranteeing the convergence of the parameter estimates, and the analysis indicates that under attenting excitation the least mean square algorithm also has a good performance:i.e., the parameter estimation error given by the least mean square algorithm uniformly converges to zero when the attenuating index and design variable satisfy some proper conditions. |
|
|
|
|
|