A signal de-noising method for a MEMS gyroscope based on improved VMD-WTD. (8th June 2021)

• Record Type: Journal Article
• Authors: Ding, Mingkuan; Shi, Zhiyong; Du, Binhan; Wang, Huaiguang; Han, Lanyi
• Is Part Of: Measurement science & technology. Volume 32:Number 9(2021)
• Publication Date: 2021-06-08
• Subjects: MEMS gyroscope -- signal de-noising -- endpoint extension -- grid-search -- triangle waveform matching method -- variational mode decomposition (VMD) -- wavelet threshold de-noising (WTD); Physical measurements -- Periodicals; Scientific apparatus and instruments -- Periodicals; Equipment and Supplies -- Periodicals; Science -- instrumentation -- Periodicals; Technology -- instrumentation -- Periodicals; Mesures physiques -- Périodiques; Physical measurements; Scientific apparatus and instruments; Periodicals; 502.87
• DOI: 10.1088/1361-6501/abfe33
• Languages: English
• ISSNs: 0957-0233

Abstract

Abstract: To reduce the influence of MEMS gyroscope random errors on navigation systems, the improved variational mode decomposition-wavelet threshold de-noising (WTD) method is proposed in this paper. First, to suppress the endpoint effect caused by the signal truncation and Hilbert transform during decomposition, the triangular waveform matching method is used to search for the waveform which most matches the endpoints in the whole signal. Secondly, the grid search algorithm is used to select the optimal parameters for the extended signal to realize the optimal decomposition. Finally, all the components are analyzed to determine autocorrelation characteristics. According to the variance of the autocorrelation function, all the components can be divided into noise components, mixed components and signal components. The noise components are directly removed, and the mixed components are de-noised using the WTD method. Then, the signal components and the mixed components are reconstructed after de-noising. Analysis of simulation and measured data de-noising experiments proves the effectiveness of the proposed method. For the static signal, the mean square error (MSE) of the proposed method is reduced by 10.1% and the signal to noise ratio (SNR) is increased by 14.2%. For the dynamic signal, the MSE of the proposed method decreases by 16.9% and the SNR increases by 18.8%.