A laser gyro signal denoising method and device based on empirical mode decomposition

Abstract

The present invention provides a laser gyro signal denoising method and device based on empirical mode decomposition, comprising the following steps: S1, using the EMD method to decompose the laser gyro output signal sequence, obtaining all IMF components and a residual sequence after the signal decomposition, and sorting all the decomposed signals from high to low according to frequency; S2, selecting the IMF components of noise aliasing; S3, filtering the selected IMF components of noise aliasing to obtain the filtered signal; S4, superimposing all the filtered signals, unselected IMF components and residual sequences, and reconstructing the denoised signal. The laser gyro signal denoising processing method proposed by the present invention can obtain an excellent denoising effect.

Description

Technical Field

[0001] The present invention belongs to the technical field of laser gyro signal processing, and in particular relates to a laser gyro signal denoising method and device based on empirical mode decomposition. Background Art

[0002] Among inertial devices, laser gyroscopes have the characteristics of wide dynamic range, fast startup speed, high reliability and digital output, and have always been highly valued by researchers. The signal output quality of laser gyroscopes is directly related to the performance of inertial devices. Due to the influence of design principles, manufacturing processes, signal measurement methods and other factors, the signals of laser gyroscopes are accompanied by a lot of noise. Domestic and foreign scholars have conducted extensive research on laser gyroscope signal noise reduction and proposed a series of signal noise reduction processing methods including the use of frequency domain filters, wavelet analysis, and forward prediction methods. However, the design bandwidth of frequency domain filters is single, and the effect of wavelet noise reduction technology on noise suppression depends on the wavelet basis, the number of decomposition layers and the threshold estimation method. The forward prediction noise filtering depends on the order of the filter. Based on the limitations of existing technologies, it is necessary to find a new method to suppress the interference of noise on the sampled signal. Summary of the invention

[0003] In order to overcome the deficiencies in the prior art, the inventors have conducted intensive research and provided a laser gyro signal denoising method based on empirical mode decomposition, in which weighted permutation entropy is introduced to extract noise aliasing components, the proposed components are filtered by the Savitzky-Golay algorithm, and the signal is reconstructed, thereby obtaining an excellent denoising effect, thereby completing the present invention.

[0004] The technical solution provided by the present invention is as follows:

[0005] In a first aspect, a laser gyro signal denoising method based on empirical mode decomposition comprises the following steps:

[0006] S1, use the EMD method to decompose the laser gyro output signal sequence, obtain all IMF components and a residual sequence after the signal decomposition, and sort all the decomposed signals from high to low according to frequency;

[0007] S2, select the IMF component of noise aliasing;

[0008] S3, filtering the selected noise aliased IMF component to obtain a filtered signal;

[0009] S4, superimpose all filtered signals, unselected IMF components and residual sequences to reconstruct the denoised signal.

[0010] In a second aspect, a laser gyro signal denoising device based on empirical mode decomposition comprises:

[0011] The signal decomposition module is used to decompose the laser gyro output signal sequence using the EMD method, obtain all IMF components and a residual sequence after the signal decomposition, and sort all the decomposed signals from high to low according to the frequency;

[0012] A component selection module, used for selecting IMF components of noise aliasing;

[0013] A filtering module is used to filter the selected IMF component of the noise aliasing to obtain a filtered signal;

[0014] The signal superposition module is used to superimpose all filtered signals, unselected IMF components and residual sequences to reconstruct the denoised signal.

[0015] In a third aspect, a laser gyro signal noise reduction device based on empirical mode decomposition includes:

[0016] one or more processors;

[0017] a storage device for storing one or more programs,

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the laser gyro signal denoising method described in the first aspect.

[0019] In a fourth aspect, a readable storage medium stores a computer program, which, when executed by a processor, implements the laser gyro signal noise reduction method described in the first aspect.

[0020] A laser gyro signal denoising method based on empirical mode decomposition provided by the present invention has the following beneficial effects:

[0021] The present invention provides a laser gyro signal denoising method based on empirical mode decomposition, which introduces weighted permutation entropy to extract noise aliasing components, filters the proposed components through a Savitzky-Golay algorithm, and reconstructs the signal, thereby achieving an excellent denoising effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The present invention provides a flow chart of a laser gyro signal denoising method based on empirical mode decomposition. DETAILED DESCRIPTION

[0023] The following detailed description of the present invention will make the features and advantages of the present invention more clear and explicit.

[0024] The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise noted.

[0025] The present invention provides a laser gyro signal denoising method based on empirical mode decomposition, such as Figure 1 As shown, the following steps are included:

[0026] S1, take the laser gyro output signal as the original input signal. Use the EMD method to decompose the laser gyro output signal sequence, obtain all IMF components and a residual sequence after signal decomposition, and sort all the decomposed signals from high to low according to frequency. The specific steps are as follows:

[0027] S1.1: According to the EMD algorithm, the original signal x(t) is composed of different IMF components. The maximum and minimum values ​​of the original signal x(t) are found, and the upper envelope xu(t) and the lower envelope xd(t) are obtained according to the found maximum and minimum values.

[0028] S1.2: According to the upper envelope xu(t) and the lower envelope xd(t), determine the mean m(t) and the difference d(t):

[0029] m(t)=(xu(t)+xd(t)) / 2, d(t)=x(t)-m(t)

[0030] S1.3: If d(t) satisfies the two conditions required for the IMF component, it is taken as the first IMF component and recorded as c1. If it does not meet the conditions, steps S1.1 and S1.2 are repeated until the conditions of the IMF component are met. The first IMF component c1 is separated from the signal x(t) to obtain the residual sequence r1: r1 = x(t)-c1;

[0031] Among them, the two necessary conditions for the IMF component are: 1. The number of local extreme points and zero-crossing points of the IMF component must be equal or differ by at most one in the entire time range; 2. At any time point, the average of the envelope of the local maximum value (upper envelope) and the envelope of the local minimum value (lower envelope) must be zero.

[0032] S1.4: Repeat steps S1.1, S1.2 and S1.3 for the residual sequence r1. When the residual sequence is monotonic or less than a preset constant value, the decomposition is terminated. At this time, the original signal x(t) will be decomposed into n-1 IMF components and the final residual sequence r will be obtained.

[0033] S2, determine the weighted permutation entropy of all IMF components and residual sequences, obtain the mean of all weighted permutation entropies, and select all IMF components whose weighted permutation entropy is greater than the mean as the IMF components of noise aliasing.

[0034] S3, filtering the selected IMF component by using the Savitzky-Golay filtering algorithm, the processing method is as follows:

[0035] A sequence s(k) with a data length of m=2n+1 and a center point of s(0) can be expressed as s(-n), s(-n+1), s(-n+2)...s(0),...s(n-2), s(n-1)...s(n). A p-order polynomial is used for fitting, and the expression is:

[0036]

[0037] Among them, w is the fitting coefficient; k is the sequence number; i is the number of each term in the summation expression; and p is the highest order of the polynomial.

[0038] The fitting polynomial of 2n+1 points of the sequence s(k) can be expressed as the following matrix:

[0039]

[0040] It can be expressed as a matrix formula:

[0041] S (2n+1)×1 =N (2n+1)×(p+1) W (p+1)×1 +E (2n+1)×1 (3)

[0042] Among them, S is the observed output vector, N is the Vandermonde matrix, W is the coefficient vector, and E is the error vector. The total square error can be expressed as:

[0043]

[0044] The problem of minimizing e can be transformed into the problem of minimizing the coefficient vector W that satisfies the requirements:

[0045] argmin||N·WS|| 2 (5)

[0046] To make its gradient equal to 0:

[0047] ▽||N·WS|| 2 =2N T (N·WS)=0 (6)

[0048] The solution of the coefficient vector W can be found as:

[0049] W=(N T ·N) -1 ·N T ·S (7)

[0050] Substituting formula (7) into (1), the predicted value s at point k = 0 is * (0) and Consistent:

[0051] s * (0) = ((e0) T (N T N) -1 N T )S (8)

[0052] where e0=[1,0,0,...] T For R p+1 The first standard basis vector of . Formula (8) represents s * (k) The prediction output at k = 0 is equal to the linear combination of the observations in the window of 2n + 1 around k = 0. By moving the window, the prediction output at other k can be obtained. The output of the filter can be expressed as the convolution operation of the weight coefficient sequence C and the measurement sequence S with the center point of any k and the window of 2n + 1:

[0053]

[0054] The filtered signal is expressed as IMF d .

[0055] S4, superimpose all filtered signals IMF d , unselected IMF components and residual sequence, and reconstruct the denoised signal.

[0056] Table 1 is a comparison of the ALLAN variances before and after signal denoising. It can be seen that the ALLAN variances of the signals after denoising are smaller than those before denoising. Compared with the wavelet denoising method using the db1 basis function, it can be seen that the denoising performance of the method of the present invention is better than that of the wavelet denoising method.

[0057] Table 1 shows the ALLAN variance comparison before and after signal noise reduction

[0058]

[0059]

[0060] According to a second aspect of the present invention, there is provided a laser gyro signal denoising device based on empirical mode decomposition, comprising:

[0061] The signal decomposition module is used to decompose the laser gyro output signal sequence using the EMD method, obtain all IMF components and a residual sequence after the signal decomposition, and sort all the decomposed signals from high to low according to the frequency;

[0062] A component selection module, used for selecting IMF components of noise aliasing;

[0063] A filtering module is used to filter the selected IMF component of the noise aliasing to obtain a filtered signal;

[0064] The signal superposition module is used to superimpose all filtered signals, unselected IMF components and residual sequences to reconstruct the denoised signal.

[0065] In a preferred embodiment, the signal decomposition module performs signal decomposition on the laser gyro output signal sequence to obtain all IMF components and a residual sequence after the signal decomposition, which is implemented in a manner including the following steps:

[0066] S1.1: According to the EMD algorithm, the original signal x(t) is composed of different IMF components. The maximum and minimum values ​​of the original signal x(t) are found, and the upper envelope xu(t) and the lower envelope xd(t) are obtained according to the found maximum and minimum values.

[0067] S1.2: Determine the mean m(t) and the difference d(t) based on the upper envelope xu(t) and the lower envelope xd(t);

[0068] S1.3: If d(t) satisfies the two conditions required for the IMF component, it is taken as the first IMF component and recorded as c1. If it does not meet the conditions, steps S1.1 and S1.2 are repeated until the conditions of the IMF component are met. The first IMF component c1 is separated from the signal x(t) to obtain the residual sequence r1: r1 = x(t)-c1;

[0069] S1.4: Repeat steps S1.1, S1.2 and S1.3 for the residual sequence r1. When the residual sequence is monotonic or less than a preset constant value, the decomposition is terminated. At this time, the original signal x(t) will be decomposed into n-1 IMF components and the final residual sequence r will be obtained.

[0070] In a preferred embodiment, the component selection module selects the IMF component of noise aliasing, which is implemented by the following steps: determining the weighted permutation entropy of all IMF components and residual sequences, obtaining the mean of all weighted permutation entropies, and selecting all IMF components whose weighted permutation entropies are greater than the mean as the IMF component of noise aliasing.

[0071] In a preferred implementation, the filtering module performs filtering processing on the selected noise aliased IMF component by using a Savitzky-Golay filtering algorithm.

[0072] According to a third aspect of the present invention, there is provided a laser gyro signal denoising device based on empirical mode decomposition, comprising:

[0073] one or more processors;

[0074] a storage device for storing one or more programs,

[0075] When the one or more programs are executed by the one or more processors, the one or more processors implement the laser gyro signal denoising method described in the first aspect.

[0076] According to a fourth aspect of the present invention, a readable storage medium stores a computer program, which, when executed by a processor, implements the laser gyro signal noise reduction method described in the first aspect.

[0077] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices, equipment and readable storage media can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0078] The device, equipment and readable storage medium technical solution of the present application can be essentially or partially embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0079] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the present invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented using software, the functions may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any media that facilitates the transmission of a computer program from one place to another. The storage medium may be any available medium that a general or special-purpose computer can access.

[0080] The present invention has been described in detail above in conjunction with specific implementations and exemplary examples, but these descriptions cannot be understood as limiting the present invention. Those skilled in the art understand that, without departing from the spirit and scope of the present invention, a variety of equivalent substitutions, modifications or improvements may be made to the technical solution of the present invention and its implementation methods, all of which fall within the scope of the present invention. The scope of protection of the present invention shall be subject to the attached claims.

[0081] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

Claims

1. A laser gyro signal denoising method based on empirical mode decomposition, characterized in that: include: Use the EMD method to decompose the laser gyro output signal sequence, obtain all IMF components and a residual sequence after the signal decomposition, and sort all the decomposed signals from high to low according to frequency; Select the IMF components of noise aliasing, specifically: determine the weighted permutation entropy of all IMF components and residual sequences, obtain the mean of all weighted permutation entropies, and select all IMF components whose weighted permutation entropies are greater than the mean as the IMF components of noise aliasing; Filtering the selected noise aliased IMF component to obtain a filtered signal; All filtered signals, unselected IMF components and residual sequences are superimposed to reconstruct the denoised signal.

2. The laser gyro signal denoising method based on empirical mode decomposition according to claim 1, characterized in that: The step of using the EMD method to decompose the laser gyro output signal sequence to obtain all IMF components and a residual sequence after the signal decomposition includes the following sub-steps: S1.1: According to the EMD algorithm, the original signal x(t) is composed of different IMF components. The maximum and minimum values ​​of the original signal x(t) are found, and the upper envelope xu(t) and the lower envelope xd(t) are obtained according to the found maximum and minimum values. S1.2: Determine the mean m(t) and the difference d(t) based on the upper envelope xu(t) and the lower envelope xd(t); S1.3: If d(t) satisfies the two conditions required for the IMF component, it is taken as the first IMF component and recorded as c1. If it does not meet the conditions, steps S1.1 and S1.2 are repeated until the conditions of the IMF component are met. The first IMF component c1 is separated from the signal x(t) to obtain the residual sequence r1: r1 = x(t)-c1; S1.4: Repeat steps S1.1, S1.2 and S1.3 for the residual sequence r1. When the residual sequence is monotonic or less than a preset constant value, the decomposition is terminated. At this time, the original signal x(t) will be decomposed into n-1 IMF components and the final residual sequence r will be obtained.

3. The laser gyro signal denoising method based on empirical mode decomposition according to claim 1, characterized in that: The step of filtering the selected noise aliased IMF component is specifically filtering the selected noise aliased IMF component by using a Savitzky-Golay filtering algorithm.

4. A laser gyro signal denoising device based on empirical mode decomposition, characterized in that: include: The signal decomposition module is used to decompose the laser gyro output signal sequence using the EMD method, obtain all IMF components and a residual sequence after the signal decomposition, and sort all the decomposed signals from high to low according to the frequency; A component selection module is used to select the IMF components of noise aliasing, including: determining the weighted permutation entropy of all IMF components and residual sequences, obtaining the mean of all weighted permutation entropies, and selecting all IMF components whose weighted permutation entropies are greater than the mean as the IMF components of noise aliasing; A filtering module is used to filter the selected IMF component of the noise aliasing to obtain a filtered signal; The signal superposition module is used to superimpose all filtered signals, unselected IMF components and residual sequences to reconstruct the denoised signal.

5. The laser gyro signal denoising device based on empirical mode decomposition according to claim 4, characterized in that: The signal decomposition module performs signal decomposition on the laser gyro output signal sequence to obtain all IMF components and a residual sequence after the signal decomposition, which is implemented by a method including the following steps: S1.1: According to the EMD algorithm, the original signal x(t) is composed of different IMF components. The maximum and minimum values ​​of the original signal x(t) are found, and the upper envelope xu(t) and the lower envelope xd(t) are obtained according to the found maximum and minimum values. S1.2: Determine the mean m(t) and the difference d(t) based on the upper envelope xu(t) and the lower envelope xd(t); S1.3: If d(t) satisfies the two conditions required for the IMF component, it is taken as the first IMF component and recorded as c1. If it does not meet the conditions, steps S1.1 and S1.2 are repeated until the conditions of the IMF component are met. The first IMF component c1 is separated from the signal x(t) to obtain the residual sequence r1: r1 = x(t)-c1; S1.4: Repeat steps S1.1, S1.2 and S1.3 for the residual sequence r1. When the residual sequence is monotonic or less than a preset constant value, the decomposition is terminated. At this time, the original signal x(t) will be decomposed into n-1 IMF components and the final residual sequence r will be obtained.

6. The laser gyro signal denoising device based on empirical mode decomposition according to claim 4, characterized in that: The filtering module performs filtering processing on the selected noise aliased IMF component through Savitzky-Golay filtering algorithm.

7. A laser gyro signal noise reduction device based on empirical mode decomposition, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the laser gyro signal noise reduction method according to any one of claims 1 to 3.

8. A readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the laser gyro signal noise reduction method according to any one of claims 1 to 3 is implemented.

Images