Method for extracting the instantaneous velocity of a Doppler signal on the basis of continuous wavelet transformation

Abstract

The invention discloses a method for extracting the instantaneous velocity of a Doppler signal on the basis of continuous wavelet transformation and belongs to the technical field of signal processing. The method is characterized by obtaining the instantaneous frequency of an output signal by means of the continuous wavelet transform, reducing errors by performing polynomial fitting on the instantaneous frequency at each moment, and solving the instantaneous velocity at each moment depending on the fitted instantaneous frequency at each moment. The method is applicable to an interferometer without a delay optical path and a reference optical path, may simultaneously keep time-domain and frequency-domain information, acquires signals with better smoothness and without the restriction of a boundary effect, is wide in application range, high in precision, accurately extracts the velocity speed at each moment.

Description

technical field [0001] The invention relates to a method for extracting the instantaneous velocity of a Doppler signal based on continuous wavelet transform, in particular to extracting the instantaneous velocity of a Doppler signal output by a Mach-Zehnder interferometer, and belongs to the technical field of signal processing. Background technique [0002] Speed ​​measurement technology is widely used in daily life, such as the measurement of vehicle speed and the measurement of motor speed. Laser Doppler velocimetry is a common method of velocimetry, among which dual-beam Doppler velocimetry can achieve high precision and is widely used. However, this method is difficult to measure out of focus, and it is more difficult to measure the speed of solid motion. Usually, it can only be used to measure the motion of fluid and the small vibration of solid. The emergence of laser interference velocimetry has solved the problem of measuring the velocity of solids well. The princi...

AI Technical Summary

Problems solved by technology

[0003] At present, although there are many types of laser interferometers used to measure speed, there are mainly three signal processing methods. The first method is to divide the output signal into two channels, and then use an oscilloscope or a data acquisition card to measure the phase difference of the two signals. To achieve speed measurement, the biggest disadvantage of this method is the addition of one output signal, which requires an additional channel to collect data, and the output signal of the laser interferometric velocimetry method often has a higher frequency, so the cost of adding one channel is relatively large; The second method is to measure the speed by measuring the change of the interference fringe. Although this method does not need to divide the output signal into two paths, it is not easy to directly observe the change of the fringe, so it is difficult to achieve high accuracy. CCD cameras can solve this problem to a certain extent, but the cost will increase a lot; the last method is to directly extract the instantaneous frequency of the signal through the signal processing algorithm, and then extract the instantaneous speed according to the instantaneous frequency. Although this method will not bring To increase the cost, but currently there is no very mature algorithm that can accurately extract the velocity signal from the Doppler signal, although the FFT (Fast Fourier Transform) algorithm can be used to process the Doppler signal, but due to the Doppler The frequency of the Le signal changes rapidly with time, and it is difficult to preserve the frequency and time information of the signal at the same time by Fourier transform, so the measurement accuracy will be limited, and the calculation speed still needs to be improved

Images