Blind source signal denoising method based on ensemble empirical mode decomposition

A technology that integrates empirical modes and blind source signals. It is applied in the field of signal processing and can solve the problems of Newton's method not converging and objective function value not converging.

Inactive Publication Date: 2015-02-25
SHENYANG JIANZHU UNIVERSITY
View PDF0 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the algorithm is sensitive to the selection of the initial value. When the initial point is far away from the minimum point, Newton’s method may not converge. The reason is that the direction of Newton’s iteration is not necessarily the direction of descent. After iteration, the value of the objective function may not converge

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Blind source signal denoising method based on ensemble empirical mode decomposition
  • Blind source signal denoising method based on ensemble empirical mode decomposition
  • Blind source signal denoising method based on ensemble empirical mode decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] In order to prove the effectiveness and superiority of the IEEMD method in processing speech signals, a test speech is used for specific analysis below. The selected voice signal is a section of female audio in the TIMIT voice library, and the content is: "She had your dark suit in greasy wash water all year".

[0062] The present invention is based on the blind source signal denoising method of ensemble empirical mode decomposition: firstly, the original EEMD algorithm is revised; secondly, the N-order IMF obtained by decomposing the revised EEMD is analyzed through stepwise regression, and false components are eliminated; then all IMFs are passed through ICA performs signal separation and reconstruction to maximize signal enhancement and eliminate noise; specifically includes the following steps: specifically includes the following steps:

[0063] First, through the enumeration experiment, that is, the signal is decomposed in the case of multiple groups of different w...

Embodiment 2

[0112] The difference from Example 1 is that this example selects a section of female audio in the TIMIT voice database as the experimental corpus for the analysis of the actual voice signal. The original sampling frequency of the voice signal is 16kHz, monophonic recording, 16bit quantization, the content is: "She had your dark suit in greasy wash water all year". figure 2 It is the 15th-order IMF component obtained after IEEMD decomposition. Unlike the simulation signal, it is impossible to intuitively judge which IMF component is a false component after the speech signal is decomposed by IEEMD, and the contribution of each IMF component to the original voice cannot be determined. Intuitive judgment.

[0113] After stepwise regression analysis of the 15th-order IMF components obtained by IEEMD decomposition, the results are shown in Table 1:

[0114] Table 1 IMF regression coefficients

[0115]

[0116] It can be clearly seen from Table 1 that the components after IMF8...

Embodiment 3

[0119] Different from Embodiment 1, this example defines the noise-added speech signal as y(t)=s(t)+n(t), where s(t) is a pure speech signal and n(t) is a noise signal. In order to visually verify the effectiveness of the algorithm proposed in this paper, a piece of female audio and a male audio were randomly selected in the TIMIT speech library for analysis, and the added noise signal n(t) is the Vehicle interior noise, F in the standard noise library noisex-92 -16 noise and factory noise. These three noises are added to the pure speech according to the signal-to-noise ratios of -5dB, -10dB, and -15dB. The input signal-to-noise ratio and output signal-to-noise ratio are defined as formulas (17) and (18).

[0120] SNR IN = 10 log 10 Σ i = 1 n ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a blind source signal denoising method based on ensemble empirical mode decomposition, and belongs to the technical field of signal processing. By means of the method, definitions on the white noise amplitude and the number of iterations in an original algorism are rectified. False component discrimination is conducted on the IMP component obtained after IEEMD is decomposed through a classic stepwise regression analysis method, features of original signals are effectively reserved, the false component generated by the IEEMD algorism is eliminated, and interference to the subsequent denoising algorism by the false component is eliminated. Finally, for the non-convergence phenomenon generated occasionally when the ICA algorism processes the high-frequency signals, a high-order TFastICA method is provided, features of the IEEMD and the TFastICA are combined, and rear-end processing is conducted on the IEEMD through the TFastICA method. The blind source signal denoising method based on the ensemble empirical mode decomposition has wide application prospects in the fields of removing mechanical vibration noise, voice signal noise, instantaneous underwater noise and other signal processing fields.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and in particular relates to a blind source signal denoising method based on ensemble empirical mode decomposition, which is applied to signal processing fields such as detecting and removing mechanical vibration noise, voice signal noise, and underwater instantaneous noise. Background technique [0002] The current denoising methods for non-stationary and nonlinear signals are mainly based on short-time Fourier transform and wavelet decomposition. The EMD decomposition method is a new nonlinear and non-stationary signal proposed by Dr. Norden E. Huang (Huang E) of the NASA Goddard Space Flight Center and his colleagues in the late 1990s. processing method. Different from wavelet decomposition, EMD decomposition does not require prior knowledge. It can adaptively decompose the fluctuations of different scales in the complex signal into a limited number of intrinsic mode functions (IMF)...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/00
Inventor 安冬须颖邵萌戴敬梁文峰杨谢柳
Owner SHENYANG JIANZHU UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap