| 基于熵调整模糊c-均值聚类的时频能量混合模型 |
| |
| 引用本文： | 田光明,陈光.基于熵调整模糊c-均值聚类的时频能量混合模型[J].信号处理,2005,21(1):1-6. |
| |
| 作者姓名： | 田光明 陈光 |
| |
| 作者单位： | 1. 电子科技大学自动化工程学院,成都,610054;中国工程物理研究院结构力学研究所,绵阳,621900
2. 电子科技大学自动化工程学院,成都,610054 |
| |
| 摘 要： | 本文提出了一种改进由时频不相交分量组成信号的双线性时频分布的分辨率和可读性的方法。用修正的Xie-Beni聚类有效性指标对熵调整模糊c-均值聚类算法进行拓展将模糊聚类与密度估计相结合,实现了信号时频分量的识别和建模;信号的时频能量混合模型给出了信号分量的数目及其在时频面上所占据的区域。这些信息可以用于分离信号分量,设计适合于每个分离分量的光滑核。仿真结果表明,对于由时频不相交分量组成的信号,本方法可以识别出其中的信号分量,并得到较优的时频分布。
|
| 关 键 词： | 时频分布 混合模型 密度估计 模糊c-均值聚类 熵调整 |
| 修稿时间： | 2003年11月6日 |
Time-Frequency Energy Mixture Model Using Fuzzy c-Means Cluster with Entropy Regularization |
| |
| Tian Guangming Chen Guangju.Time-Frequency Energy Mixture Model Using Fuzzy c-Means Cluster with Entropy Regularization[J].Signal Processing,2005,21(1):1-6. |
| |
| Authors: | Tian Guangming Chen Guangju |
| |
| Abstract: | This paper presents a method for improvement of the resolution and clarity of bilinear time-frequency distributions (TFDs) generated from signals consisting of a number of approximately time-frequency (TF) disjoint components. Using a modified Xie-Beni cluster validity index, the fuzzy c-means cluster algorithm with entropy regularization is extended to integrate fuzzy cluster and density estimation to identify TF components and derive TF energy mixture model that indicates the number of components in the signal, and the regions they occupy in the TF plane. This information is used to isolate the components, and smoothing kernels are designed using the properties of each isolated component. Simulation results indicate that, to the signals consisted of numerous TF disjoint components, our method can extract components from it, and get better TFDs. |
| |
| Keywords: | time-frequency distribution mixture model density estimation fuzzy c-means cluster entropy regularization |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
