Wavelet-based Signal Analysis And Processing

Short-time Fourier transform (STFT) is a common time-frequency analysis tools, it is a kind of linear transformation and avoid the interference of the cross in other type high times the non-stationary signal analysis method, Wavelet Analysis is localized time-frequency analysis tool. It uses time and frequency jointly to denote the characteristics of signals. Both of them have been widely applied in signal detection, analysis and interference suppression, etc.This paper firstly carries the comprehensive review on the way of signal analysis development and introduced the basic principle and properties of short-time Fourier transform and wavelet analysis respectively from two aspects of continuous and discrete which can be the foundation to the next chapters. Then emphatically introduces the time-frequency analysis theory, the application based on short-time Fourier transform and Wavelet Analysis in MATLAB is simulated, the related applications including non-stationary signal time-frequency analysis and spectrum analysis of speech. Through the experiment, we can get the picture of time-frequency distribution by making short-time Fourier transform on the non-stationary signal, the frequency component can be pinpointed at a time and frequency plane anytime. Use the short-time Fourier transform processing speech signal, then the picture of sounds and voiced time-frequency signal two-value function could be got, which gives characteristic of the frequency spectrum of the voiced and unvoiced speech and energy distribution. And the seismic signal denoising based on wavelet analysis. Simulation results show that the short-time Fourier transform and wavelet analysis are powerful tools in analyzing the non-stationary signal, and receive good effect.