Unlock instant, AI-driven research and patent intelligence for your innovation.

Mask signal frequency selection method based on energy index

A mask signal and frequency selection technology, applied in the field of signal processing, can solve problems such as energy leakage, mode aliasing, and the inability to completely retain the highest-order frequency signal, so as to ensure integrity and reduce errors

Pending Publication Date: 2020-12-22
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The key to constructing the mask signal is to determine the frequency of the mask signal. The traditional mask signal uses the average instantaneous frequency of the first-order component of the EMD decomposition result as the mask signal frequency, but the first-order component cannot be completely preserved. The information of the highest-order frequency signal will leak some energy into other components; on the other hand, this frequency selection method is not applicable to all signals, and the problem of modal aliasing still occurs in actual use

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
  • Mask signal frequency selection method based on energy index
  • Mask signal frequency selection method based on energy index
  • Mask signal frequency selection method based on energy index

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] Such as figure 1 As shown, a mask signal frequency selection method based on an energy index is characterized in that it comprises the following steps:

[0047] Step 1, perform EMD decomposition for a given signal x(t), including the following steps:

[0048] 101. Find all local extremum points of the signal x(t), wherein: t is a time variable;

[0049] 102. Fit the local maxima and local minima sequences with cubic splines respectively to obtain the upper and lower envelopes e u1 、e d1 ;

[0050] 103. Calculate the mean of the upper and lower envelopes

[0051] 104. Subtract the mean m from the original signal 11 (t), get the residual signal h 1 (t)=x(t)-m 11 (t);

[0052] 105. Judgment h 1 (t) Whether the given sieving stop criterion is satisfied, if so, consider h 1 (t) is an intrinsic mode function (intrinsic mode function, IMF), if not satisfied, let h 1 (t) instead of x(t), repeat the step 101 to the step 104 until the remaining signal h that satisfie...

Embodiment 2

[0070] A mask signal frequency selection method based on an energy index, comprising the following steps:

[0071] Perform EMD decomposition on the signal x(t) to be processed to obtain all intrinsic mode functions (IMF) and trend items; perform Hilbert transformation on the first-order IMF, and obtain the instantaneous frequency sequence f of the first-order component by constructing an analytical signal (t) and the instantaneous amplitude sequence a(t); use the energy mean method to obtain the average instantaneous frequency And find the mask signal amplitude Among them: p represents the position of the point in the instantaneous amplitude sequence, q represents the length of the instantaneous amplitude sequence, a(p) represents the instantaneous amplitude corresponding to the point, and f(p) represents the instantaneous value corresponding to the point Frequency; determine the range of the parameter m [2,4], the step size is 0.01, that is, there are 201 possible values ​...

Embodiment 3

[0073] Next, the mask signal frequency selection method of the present invention is used to test the simulated signal to verify the suppression effect on EMD modal aliasing.

[0074] Suppose the signal x(t) to be processed is a combined signal of three sinusoidal signals

[0075] x(t)=sin(2πt·21.2)+sin(2πt·17.7)+sin(2πt·12.4), figure 2 is the time domain plot of x(t), image 3 is the spectrogram of x(t);

[0076] Perform traditional EMD decomposition on x(t), Figure 4 It is the spectrogram of the first-order IMF component. It can be seen from the figure that there are three modal frequencies in the first-order IMF component that should only contain a single frequency signal. The result of traditional EMD decomposition is not very ideal;

[0077] Analyze the first-order IMF component u(t) obtained by EMD decomposition, and perform Hilbert transformation on it Where: τ represents the time offset, d τ Integrate the variable τ to construct an analytical signal Z(t)=u(t)+jν...

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 discloses a mask signal frequency selection method based on an energy index. The method comprises the following steps: step 1, performing EMD decomposition on a given signal; and step 2,constructing a mask signal on the basis of an EMD decomposition result. According to the method, a mask frequency parameter most suitable for the signal is selected according to different signals through a method of formulating the energy index, and meanwhile, the filtering range does not need to be judged manually, so that the self-adaptive characteristic of EMD is not damaged, and errors causedby human intervention can be reduced. Besides, different from a traditional mask signal method in which the mask result is directly used as the first-order component of the EMD, only the frequency information of the mask result is extracted, the band-pass filtering range is determined based on the frequency information, the signal containing the modal information of the order is separated out, and the integrity of the signal information is guaranteed.

Description

technical field [0001] The invention belongs to the field of signal processing, in particular to a mask signal frequency selection method based on an energy index. Background technique [0002] In signal processing in the engineering field, a large number of non-stationary signals need to be analyzed. In 1998, Chinese-American scientist Norden E. Huang et al. proposed a time-frequency analysis method that can adaptively analyze non-stationary signals: Hilbert-Huang transform (HHT), which is composed of two parts: empirical mode decomposition (EMD) and Hilbert spectrum. The purpose of EMD decomposition is to decompose complex signals into several signal components according to different frequencies. However, if multiple frequencies are close to each other in the original signal, the EMD method will not be able to identify these frequencies, and modal aliasing will occur. In response to this problem, Deering et al. proposed a masking signal method (masking signals) to suppres...

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): G06K9/00G06F17/15G06F17/18
CPCG06F17/15G06F17/18G06F2218/04G06F2218/08
Inventor 周丽卞超范石磊
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS