基于HHT的非平稳信号分析仪的研究

本文介绍了希尔伯特-黄变换（HHT）的原理，首先通过经验模态分解（EMD），信号被分解成一系列固有模态函数（IMF），再通过Hilbert变换得到每个IMF的瞬时频率（IF）和瞬时幅值函数，最终得到原始信号的IF分布和Hilbert谱。Hilbert谱是信号的时间-频率-能量分布。为使HHT能有效分析非平稳信号，引入了改进HHT的方法，即在HHT过程中，将小波包变换（WPT）作为预处理器，外加IMF的筛选。采用虚拟仪器开发技术研制了一台基于HHT的非平稳信号分析仪。最后以HHT去噪为例，介绍了基于HHT的非平稳信号分析仪的应用。

Research on non-stationary signal analyzer based on HHT

The principle of Hilbert-Huang transform (HHT) is introduced. Firstly, the analyzed signal is decomposed into a collection of intrinsic mode functions (IMF) through empirical mode decomposition. Then, the instantaneous frequencies (IF) and instantaneous amplitude of each IMF are derived through the Hilbert transform, hence, the IF distribution and Hilbert spectrum are obtained. Hilbert spectrum has an energy-frequency-time distribution of the signal. In order to extract the characteristics of non-stationary signal using HHT, an improved HHT method is introduced, which uses wavelet packet transform (WPT) as preprocessor and has the IMF selection function added. Based on HHT, a non-stationary signal analyzer was developed, which adopts virtual instrument technology. Finally, an example of noise removal is given to show the application of the signal analyzer.