Spectral signal noise reduction method based on smoothing filtering with combination of VS-LMS

Abstract

The invention discloses a spectral signal noise reduction method based on smoothing filtering with combination of VS-LMS. The method mainly comprises the following steps: filtering signals by using a smooth filter and reducing the content of noise in the signals; and carrying out secondary filtering by using a VS-LMS algorithm, increasing resolution ratio of detail parts of a spectrum and clearly embodying the characteristics of the spectrum. Through the spectral signal noise reduction method based on smoothing filtering with combination of VS-LMS, the problems that the complexity of the algorithm is high, the network is unstable, the overall noise reduction performance is low, the influence of noise in the detail parts cannot be inhibited and the spectrum chart cannot reflect the real spectral characteristics in the prior art can be overcome.

Description

technical field [0001] The invention relates to the technical field of spectral signal noise reduction, in particular to a spectral signal noise reduction method based on smoothing filtering combined with VS-LMS. Background technique [0002] In recent years, spectral analysis has attracted more and more attention from analysis experts at home and abroad, and is widely used in agriculture, food industry, pharmaceutical industry and clinical medicine. However, in practical applications, the spectrum is inevitably disturbed by some noise, so that useful information cannot be obtained from the spectrum. Therefore, spectral denoising is particularly important in spectral analysis. [0003] For the spectral signal under low signal-to-noise ratio, EEMD combined with VS-LMS denoising algorithm, although the algorithm has a good overall denoising effect, it does not have a good suppression effect on the noise of some details, which makes the spectral image The same characteristics...

AI Technical Summary

Problems solved by technology

[0003] For the spectral signal under low signal-to-noise ratio, EEMD combined with VS-LMS denoising algorithm, although the algorithm is good in the overall denoising effect, but it does not have a good suppression effect on the noise of some details, so that the spectrogram The same characteristic cannot be well reflected

[0004] Smoothing filtering can effectively smooth the high-frequency components in the signal and improve the signal-to-noise ratio, but the processing of signal details is not ideal

After many experiments by the author, when dealing with signals with a signal-to-noise ratio lower than 10db, compared with wavelet and EEMD, LMS has better noise reduction performance, but the resolution of the peaks and valleys of the signal is not high, and there are steady-state errors and Paradox of Convergence Speed

Images