Modulation spectrum fast zoom method based on FFT

Abstract

The invention discloses a modulation spectrum fast zoom method based on FFT. The method comprises the following steps that (1) sampling frequency is preliminarily determined by the analysis frequency range; (2) observation window width is set so that the expected number of sampling data are obtained; (3) the data are grouped so that the actual number of the sampling data are obtained; (4) the sampling frequency is adjusted so that the observation window is enabled to be corresponding to the analysis frequency range; (5) the data are sampled and grouped FFT is performed; and (6) an initial spectral line is computed and zoom spectrum is obtained. The method has the following beneficial effects and advantages that 1. resolution for the fine structure of the modulation spectrum is enabled to be consistent with that of direct FFT and higher than that of ZFFT by adopting the algorithm based on combination of direct FFT and any base FFT; and 2. computing speed of the method is higher than that of direct FFT and ZFFT for more than 2 times and can be enhanced to be more than 4 times of computing speed of direct FFT by reasonably selecting the observation window width 2M, the half bandwidth expansion coefficient alpha or the sampling coefficient beta according to different application requirements.

Description

technical field [0001] The invention belongs to thinning spectrum analysis technology, in particular to a fast thinning method of modulation spectrum based on FFT. Background technique [0002] The modulation spectrum is the spectrum of the continuous modulated wave produced by the modulation of an analog signal. Signal modulation is to use a signal (modulation signal) to control another signal (carrier signal) as a carrier, so that a certain parameter (amplitude, frequency, phase, pulse width, etc.) of the latter changes according to the value of the former; usually in the form of A high-frequency sinusoidal signal is used as a carrier signal, and commonly used modulation methods include amplitude modulation, frequency modulation, phase modulation, and pulse width modulation. In actual engineering signal monitoring, due to interference, natural frequency and other modulation factors, the signal form is often as attached figure 2 shown. For attached figure 2 The acquisit...

AI Technical Summary

Problems solved by technology

[0004] In actual engineering signal monitoring, in order to quickly identify the fine structure in the modulation spectrum, the signal analysis must not only have high frequency resolution, but also have a wide frequency range, and meet the real-time performance of the system; however, the conventional FFT Both spectrum and frequency refinement methods are difficult to take into account at the same time

Whether it is possible to invent a method for quickly identifying the fine structure in the modulation spectrum has not yet been solved by the scientific and technological circles at home and abroad, and there is no relevant research report.

Images