| 一种新的基于极大似然估计的系统辨识方法 |
| |
| 引用本文： | 朱幼莲,刘琚.一种新的基于极大似然估计的系统辨识方法[J].现代雷达,1998,20(1):33-38. |
| |
| 作者姓名： | 朱幼莲 刘琚 |
| |
| 作者单位： | 常州技术师范学院，东南大学 |
| |
| 摘 要： | 进化计算是一种基于自然选择和自然变异机制的并行随机优化算法。由于递推极大似然估计法难以实现性能指标函数的全局寻优，所以本文提出了将改进的进化计算用于系统参数辨识的方法。计算机仿真结果表明，该算法能够实现似然函数的全局寻优。
|
| 关 键 词： | 极大似然估计，信号处理，进化计算 |
A New Method of System Identification Based on Maximum Likelihood Estimate |
| |
| Zhu Youlian.A New Method of System Identification Based on Maximum Likelihood Estimate[J].Modern Radar,1998,20(1):33-38. |
| |
| Authors: | Zhu Youlian |
| |
| Affiliation: | Zhu Youlian (Changzhou Normal Institute of Technology Changzhou 213001) Liu Ju He Zhenya(Southeast University Nanjing 210096) |
| |
| Abstract: | A new method of system identification based on maximum likelihood estimate is presented in this paper. Evolutionary computation is a stochastic optimal algorithm based on mechanics of natural selection and mutation. It has more advantages over gradient maximum likelihood algorithm, which is difficult to global optimization of object function. Experimental results show that evolutionary algorithm can reach the maximization of the global likelihood function. |
| |
| Keywords: | maximum likelihood estimate signal processing evolutionary computation |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《现代雷达》浏览原始摘要信息 |
|
点击此处可从《现代雷达》下载全文 |
