| 广义预测误差模型及其GLMS自适应算法 |
| |
| 引用本文： | 刘世忠 刘曼铃. 广义预测误差模型及其GLMS自适应算法[J]. 四川大学学报(自然科学版), 1990, 27(3): 313-323 |
| |
| 作者姓名： | 刘世忠 刘曼铃 |
| |
| 作者单位： | 四川大学无线电电子学系(刘世忠)，四川大学无线电电子学系(刘曼铃) |
| |
| 摘 要： | 提出了广义预测误差模型,其二次型性能面的形状受控于模型参数α,在最小均方误差准则下,推导出GLMS自适应算法.通过简化的GLMS算法和LMS算法的类比分析,证明了模型参数α对算法的收敛特性具有明显的影响.在α<0时,可获得比LMS算法更好的收敛特性,而计算量增加不多.数字仿真结果与理论分析相符.
|
| 关 键 词： | 自适应算法 模型参数估计 预测滤波 |
GENERALIZED PREDICTION ERROR MODEL AND ITS GLMS ADAPTIVE ALGORITHM |
| |
| Liu Shizhong Liu Manlin. GENERALIZED PREDICTION ERROR MODEL AND ITS GLMS ADAPTIVE ALGORITHM[J]. Journal of Sichuan University (Natural Science Edition), 1990, 27(3): 313-323 |
| |
| Authors: | Liu Shizhong Liu Manlin |
| |
| Affiliation: | Department of Radio Electronics |
| |
| Abstract: | A generalized prediction error model in presented in this paper. Its performance surface is contrqled by model parameter a. Based on the minimun mean-square error criterion, a GLMS(Generalized LMS) adaptive algorithm is derived.By comparing a simplified GLMS algorithm with LMS algorithm, it is shown that the model parameter a has evident affects on the algorithm convergence. And when a <0, the convergence property of the simolified GLMS is better than of LMS algorithm, with a slight increase in the computational complexity.Extensive computer simulations support our theoretical analysis. |
| |
| Keywords: | adaptive algorithm model parameter estimate pridiction filter. |
| 本文献已被 CNKI 维普 等数据库收录！ |
