| 一类基于非线性PCA准则的复数信号盲分离算法 |
| |
| 引用本文： | 倪晋平,马远良,鄢社锋.一类基于非线性PCA准则的复数信号盲分离算法[J].信号处理,2002,18(1):52-56. |
| |
| 作者姓名： | 倪晋平 马远良 鄢社锋 |
| |
| 作者单位： | 1. 西北工业大学航海工程学院,西安,710072;西安工业学院,西安,710032
2. 西北工业大学航海工程学院,西安,710072 |
| |
| 基金项目： | 国家自然科学基金资助项目（60072052） |
| |
| 摘 要： | 在阵列信号处理过程中，经常遇到复数信号盲分离问题。例如，卷积混合型的源信号的盲分离；声纳信号盲分离。本文提出了一类基于非线性准则的复数信号盲分离算法。将非线性函数引入学习过程，由算法自动调节学习速率。计算机仿真实验验证了算法的有效性，文中给出了验证结果。
|
| 关 键 词： | 盲源分离 独立分量分析 非线性主分量分析 神经网络 |
| 修稿时间： | 2001年7月5日 |
A Class of Blind Separation of Complex Valued Signals Based on Nonlinear PCA Criterion |
| |
| Ni Jinping Ma Yuanliang,Yan Shefeng.A Class of Blind Separation of Complex Valued Signals Based on Nonlinear PCA Criterion[J].Signal Processing,2002,18(1):52-56. |
| |
| Authors: | Ni Jinping Ma Yuanliang Yan Shefeng |
| |
| Abstract: | Separation of complex valued signals is a frequently arising problem in array signal processing. For example, separation of convolutively mixed signals such as underwater acoustic signals in receiving array involves computations on complex valued signals. A new set of complex blind separation algorithm based on nonlinear criterion is proposed. Complex valued recursive least square (RLS) algorithm is one of them. It needs not choosing the learning step. And the convergence rate corresponds to other s adaptive blind separation algorithms for complex valued signals. Used nonlinear function in learning, the algorithms can adjust automatically convergent rate. The performance of proposed complex algorithms is verified by computer simulation. The results of simulation are given. |
| |
| Keywords: | Blind source separation Independent component analysis(ICA) Nonlinear PCA Neural networks |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
