MZI signal denoising method based on neighborhood wavelet coefficients

Abstract

The invention discloses an MZI signal denoising method based on neighborhood wavelet coefficients, and relates to a signal processing method. The MZI signal denoising method comprises the steps that first, the method for introducing the neighborhood wavelet coefficients completely considers the relation between the wavelet coefficients in scales, and the denoising effect is improved; second, an improved threshold value can be changed along with the changes of a decomposition scale, and the phenomena of 'over-throttling' and 'over-reserving' in the denoising process of a traditional threshold value are well removed; at last, a novel threshold function is provided, the improved threshold value is combined for processing noisy signals, the denoising effect of a traditional wavelet is improved, a gauss formula is used for fitting the denoising signals, the wavelength corresponding to a peak value is obtained, and therefore the demodulation of spectral signals is achieved.

Description

technical field [0001] The invention relates to the field of signal processing of optical fiber sensing, in particular to a method for processing MZI sensing signals. technical background [0002] Mach-Zehnder interferometer (MZI) has good optical filtering characteristics, and has been widely used in optical sensing, fiber accelerators and fiber optic hydrophones. Demodulation is the core technology of the optical fiber sensing system, and the change of the measured signal parameters is deduced by detecting the offset of the wavelength. However, the signal is easily polluted by noise in the process of acquisition and transmission, which affects the demodulation accuracy. [0003] The demodulation technology includes denoising and peak finding. Peak finding is based on denoising, so denoising is the basis of demodulation. The existence of noise will not only affect the resolution of the system, but even interfere with the real signal, affecting the next step of signal proc...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims at the problem of relatively large reconstruction signal error when processing the current wavelet coefficient based on the neighborhood coefficient method in the prior art. Based on the traditional wavelet threshold denoising method and the NeighCoeff method, an improved threshold and Threshold function, the threshold can change with the number of decomposition layers, avoiding the phenomenon of "over strangling" or "over retaining" wavelet coefficients, and at the same time combined with the threshold function to process the MZI sensing signal, effectively solving the above existing problems insufficient

Images