A method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition

Abstract

This invention provides a method for improving the signal-to-noise ratio (SNR) during DPOAE testing using EMD decomposition, belonging to the field of otoacoustic emission detection technology. The method includes the following steps: S1: EMD decomposition of the acquired raw signal into different IMF signals; S2: Time-frequency analysis of the different IMF signals to obtain a first-type IMF signal containing transient noise and a second-type IMF signal without transient noise; S3: Weighted reconstruction of the first-type and second-type IMF signals to form a processed signal; S4: Calculation of the processed signal's SNR using a coherent averaging algorithm. This invention weights and reconstructs the first-type and second-type IMF signals to form a processed signal, and calculates the processed signal's SNR using a coherent averaging algorithm to reduce the impact of other transient noise signals on the coherent averaging algorithm.

Description

Technical Field

[0001] This invention relates to the field of otoacoustic emission (OAE) detection technology, and more specifically to a method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition. Background Technology

[0002] DPOAE stands for Distortion Product Otoacoustic Emissions. DPOAEs include multiple emission frequencies, such as 2f1-f2 and f2-f1, among which 2f1-f2 has the most stable occurrence rate and the highest intensity, and is currently the frequency generally used in clinical applications. The amplitude of DPOAEs is generally 60 dBSPL lower than the stimulus intensity, and the amplitude of DPOAEs is affected by many factors, such as the intensity difference between the two stimulus signals and the frequency ratio. Studies have shown that a value of f2 / f1 between 1.22 and 1.25 yields the maximum amplitude of DPOAEs in the mid-frequency range of 1–4 kHz, and the amplitude of DPOAEs is maximized when the stimulus intensity of f1 is 10 dB or more higher than that of f2.

[0003] Since DPOAEs signals are very weak and are basically submerged in noise, the common method to extract effective DPOAEs signals is coherent averaging. When the noise occurs randomly, coherent averaging can effectively improve the signal-to-noise ratio of the signal. However, during the measurement process, other transient noise signals will occur. These signals only occur in one or two occasional tests. If such data is also added to the coherent averaging algorithm, the signal-to-noise ratio of the DPOAEs signal will be greatly reduced. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the prior art by proposing a method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition.

[0005] This invention proposes a method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition, comprising the following steps:

[0006] S1: The raw signal acquired during the DPOAE test is decomposed into different IMF signals using EMD;

[0007] S2: Perform time-frequency analysis on different IMF signals to obtain the first type of IMF signal containing transient noise and the second type of IMF signal not containing transient noise;

[0008] S3: Weighted reconstruction of the first type IMF signal and the second type IMF signal are performed respectively to remove transient noise and form a processed signal;

[0009] S4: The signal-to-noise ratio of the processed signal is improved by calculating the coherent averaging algorithm.

[0010] Furthermore, step S1 specifically includes:

[0011] S11: Acquire raw signals of a preset length. ;

[0012] S12: Transfer the original signal Based on termination criteria EMD decomposition formation To obtain different IMF signals, where and These are two time series in the process of screening IMF components, where T represents the signal duration and Sd represents the standard deviation. express The intrinsic modulus function, Represents the residual.

[0013] Furthermore, the EMD decomposition in step S12 specifically includes:

[0014] S121: Draw the upper and lower envelope lines based on the upper and lower extreme points of the original signal;

[0015] S122: Obtain the mean of the upper and lower envelopes, and draw the mean envelope based on the mean;

[0016] S123: Subtract the mean envelope from the original signal to obtain the intermediate signal;

[0017] S124: Determine whether the intermediate signal satisfies the intrinsic modulus function (IMF) determination principle. If it does, treat the intermediate signal as an IMF component and return to step S121 after subtracting the IMF component from the original signal to obtain the new original signal. If it does not satisfy the principle, return to step S121 after treating the intermediate signal as the new original signal.

[0018] Furthermore, the principle for determining the intrinsic modulus function is as follows: the number of local extrema and zero-crossing points of the function is equal or differs by at most one throughout the entire time range; at any given time point, the average of the envelopes of the local maximum and the envelopes of the local minimum is zero.

[0019] Further, step S2 includes: transforming different IMF signals to obtain corresponding Hilberc spectra, performing time-frequency analysis on the IMF signals based on the Hilberc spectra, and if the IMF signal contains transient noise signals, then the IMF signal is marked as a first type of IMF signal; if the IMF signal does not contain transient noise signals, then the IMF signal is marked as a second type of IMF signal.

[0020] Furthermore, step S2 specifically includes:

[0021] S21: Obtain the amplitude spectrum , ;

[0022] S22: Based on amplitude spectrum For each The signal is transformed by HHT transformation to obtain the transformed signal;

[0023] S23: Perform time-domain plot analysis on each transformed signal to obtain transient noise signals. Transformed signals containing transient noise signals are labeled as Type I IMF signals, and transformed signals without transient noise signals are labeled as Type II IMF signals.

[0024] Furthermore, step S21 specifically includes:

[0025] S211: Perform Hilbert transform on each IMF signal to obtain... , Where s(n) represents the actual signal included in the response x(n) after each stimulus, u(n) represents the conventional noise signal included in the response x(n) after each stimulus, and t represents time. Indicates frequency;

[0026] S212: Obtain the induced response signal x(n) through the formula x(n) = s(n) + u(n), where s(n) represents the actual signal and u(n) represents the conventional noise signal;

[0027] S213: Based on The phase function is obtained from x(n) respectively. and instantaneous amplitude , , ,based on Get instantaneous frequency , ;

[0028] S214: Based on instantaneous amplitude and instantaneous frequency Original signal Perform HHT transformation to obtain ;

[0029] S215: Based on Obtaining the amplitude spectrum , .

[0030] Furthermore, step S3 specifically includes:

[0031] The first type of IMF signal is weighted and attenuated, the second type of IMF signal is normally weighted, and all IMF signals are reconstructed to form the processed signal.

[0032] Furthermore, based on formulas Acquire and process signals, where The attenuation coefficient is... , Indicates the processing signal. express The intrinsic modulus function, Represents the residual.

[0033] Furthermore, the value of Sd ranges from 0.2 to 0.3.

[0034] The method of improving the signal-to-noise ratio during DPOAE testing using EMD decomposition according to the present invention has the following beneficial effects:

[0035] The acquired raw signal is decomposed into different IMF signals using EMD. Time-frequency analysis is then performed on these IMF signals to obtain the first type of IMF signal containing transient noise and the second type of IMF signal without transient noise. This identifies whether transient noise exists in the different IMF signals. If transient noise is present, it is removed through weighted reconstruction, thus eliminating the transient noise component from the original signal and reducing its impact on the coherent algorithm. Weighted reconstruction is performed on the first and second type of IMF signals to remove transient noise, forming a processed signal. Finally, a coherent averaging algorithm is used to calculate the signal-to-noise ratio (SNR) of the processed signal, reducing the impact of other transient noise signals on the coherent averaging algorithm and improving the SNR of the signal with transient noise, thereby improving the testing efficiency of otoacoustic emissions. Attached Figure Description

[0036] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention. In these drawings, similar reference numerals are used to denote similar elements. The drawings described below are some embodiments of the invention, but not all embodiments. Other drawings will be readily available to those skilled in the art based on these drawings without any inventive effort.

[0037] Figure 1 This is a flowchart illustrating a method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition, according to an embodiment of the present invention.

[0038] Figure 2 This is a flowchart illustrating the process of eliminating transient noise in a method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition, according to an embodiment of the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0040] Please see Figure 1-2 An embodiment of the present invention provides a method for improving the signal-to-noise ratio during DPOAE testing using EMD decomposition, comprising the following steps:

[0041] S1: The raw signal acquired during the DPOAE test is decomposed into different IMF signals using EMD;

[0042] S2: Perform time-frequency analysis on different IMF signals to obtain the first type of IMF signal containing transient noise and the second type of IMF signal not containing transient noise;

[0043] S3: Weighted reconstruction of the first type IMF signal and the second type IMF signal are performed respectively to remove transient noise and form a processed signal;

[0044] S4: The signal-to-noise ratio of the processed signal is improved by calculating the coherent averaging algorithm.

[0045] Specifically, a common method to improve the signal-to-noise ratio is the coherent averaging method, the algorithm of which is described below:

[0046] Since DPOAEs signals are very weak and are basically submerged in noise, a common processing method to extract effective DPOAEs signals is to superimpose a large amount of test data so that the noise cancels each other out and the useful DPOAEs signals are highlighted.

[0047] Specifically, this application pertains to a method for otoacoustic emission (OAE) detection, applied to instruments with OAE detection functions. This application is primarily used in distortion product otoacoustic emission (DPOAE) detection, mainly to remove transient noise during the DPOAE test. The recorded evoked response signal x(n,i) is composed of the actual signal s(n,i) and the noise signal u(n,i), i.e.: x(n,i) = s(n,i) + u(n,i); i = 1, 2, ..., M, where M represents the total number of stimuli and i is the stimulus number. For each stimulus, a response x(n,i) will be generated. The noise is often stronger than the response, and the noise is essentially random, so each x(n,i) is different, making it difficult to determine the shape of the signal s(n) from a single record. However, because OAE is repeatable, s(n,i) remains essentially unchanged each time under the same stimulus conditions. Therefore, under the same stimulus conditions, s(n) can be considered approximately a deterministic signal. Let u(n) be a stationary random signal with zero mean and variance w^2, and that each stimulus is independent of the others, i.e.: E{u(n, i)u(n, j)} = 0. When n != j, sum the corresponding M records of x(n) and take the average: The operation of this formula is called "coherent averaging," and the smoothness of the signal depends on the magnitude of the stimulus M. If the power of s(n) is denoted as P, then for each stimulus, the signal-to-noise ratio (SNR) of x(n,i) is P / w^2. After averaging M samples, the signal power remains P, and the mean noise remains zero, but the variance becomes w^2 / M. Thus, the signal-to-noise ratio becomes: SNR = P / (w^2 / M) = M*P / w^2, which is M times higher than the signal-to-noise ratio without coherent averaging. In the formula... The requirement is that the noise signal u(n, i) is random, and only when they are coherently averaged can the noise signal be canceled out.

[0048] Specifically, however, during actual measurements, transient noise signals may occur, such as friction between the test cable and clothing, swallowing saliva, and environmental noise. These signals are not constant but only occur occasionally in one or two tests. If such data is also included... In the process of coherent averaging, transient noise significantly reduces the signal-to-noise ratio (SNR) of DPOAEs signals, necessitating an increase in the number of coherent averaging iterations to improve the SNR. Measurement time also increases. Therefore, it is necessary to first decompose the original signal into multiple IMF signals. Then, time-frequency analysis is used to obtain the first type of IMF signal containing transient noise and the second type of IMF signal without transient noise. All IMF signals are then weighted and reconstructed. The first type of IMF signal containing transient noise is weighted and reduced, while the second type of IMF signal without transient noise is not reduced. Finally, all IMF signals are reconstructed to remove transient noise, eliminate its impact on coherent averaging, and improve the SNR.

[0049] Step S1 specifically includes:

[0050] S11: Acquire raw signals of a preset length. ;

[0051] S12: Transfer the original signal Based on termination criteria EMD decomposition formation To obtain different IMF signals, where and These are two time series in the process of screening IMF components, where T represents the signal duration and Sd represents the standard deviation. express The intrinsic modulus function, Represents the residual.

[0052] Specifically, assuming that during the screening of IMF components, and These are two of the time series, according to the termination criteria. Based on this, EMD decomposes the input signal into several intrinsic mode functions and a residual, i.e., by the formula... Composition, in which Indicates the input signal. express The intrinsic modulus function, Sd represents the residual, T represents the signal duration, and Sd represents the standard deviation.

[0053] Step S12, EMD decomposition, specifically includes:

[0054] S121: Draw the upper and lower envelope lines based on the upper and lower extreme points of the original signal;

[0055] S122: Obtain the mean of the upper and lower envelopes, and draw the mean envelope based on the mean;

[0056] S123: Subtract the mean envelope from the original signal to obtain the intermediate signal;

[0057] S124: Determine whether the intermediate signal satisfies the intrinsic modulus function (IMF) determination principle. If it does, treat the intermediate signal as an IMF component and return to step S121 after subtracting the IMF component from the original signal to obtain the new original signal. If it does not satisfy the principle, return to step S121 after treating the intermediate signal as the new original signal.

[0058] Specifically, it is determined whether the intermediate signal satisfies the intrinsic mode function (IMF) determination principle. If it does, the signal is an IMF component. If not, the analysis is repeated based on the signal. The acquisition of IMF components usually requires several iterations. After obtaining the first IMF component IMF1, the original signal is subtracted from IMF1 to obtain the new original signal. The analysis is repeated again, and IMF2 is obtained. This process is repeated to complete the EMD decomposition.

[0059] The criteria for determining intrinsic modulus functions are: the number of local extrema and zero-crossing points of the function is equal or differs by at most one throughout the entire time range; at any given time point, the average of the envelopes (upper envelope) of the local maximum and the envelopes (lower envelope) of the local minimum is zero.

[0060] Specifically, in physics, for instantaneous frequency to be meaningful, the function must be symmetric, have a local mean of zero, and possess the same number of zero-crossing points and extrema. Any signal consists of several intrinsic modulo functions, and an intrinsic modulo function must satisfy the eigenfunction criterion. Regarding the second condition above, the classical global requirement is modified to a local requirement, so that the instantaneous frequency is no longer affected by unnecessary fluctuations caused by asymmetrical waveforms. In reality, this condition should be "the local mean of the data is zero." However, for non-stationary data, calculating the local mean involves the concept of a "local time scale," which is difficult to define. Therefore, the second condition uses the average of the local maximum envelope and the local minimum envelope being zero to replace it, making the signal waveform locally symmetric.

[0061] Step S2 includes: transforming different IMF signals to obtain corresponding Hilberc spectra, performing time-frequency analysis on the IMF signals based on the Hilberc spectra, and if the IMF signal contains transient noise signals, then the IMF signal is marked as a first-type IMF signal; if the IMF signal does not contain transient noise signals, then the IMF signal is marked as a second-type IMF signal.

[0062] Step S2 specifically includes:

[0063] S21: Obtain the amplitude spectrum , ;

[0064] S22: Based on amplitude spectrum For each The signal is transformed by HHT transformation to obtain the transformed signal;

[0065] S23: Perform time-domain plot analysis on each transformed signal to obtain transient noise signals. Transformed signals containing transient noise signals are labeled as Type I IMF signals, and transformed signals without transient noise signals are labeled as Type II IMF signals.

[0066] Specifically, analyze m items. Signal, according to the formula Calculate each The HHT transform is used. Because transient noise is discontinuous in the time domain, the HHT transform of different IMF components is used to analyze in which IMF signal the transient noise belongs. If there is no noise, the corresponding IMF signal cannot be identified.

[0067] Specifically, Empirical Mode Decomposition (EMD) is the technical analysis premise of this invention. HHT mainly consists of two parts: Empirical Mode Decomposition (EMD) and Hilbert Spectral Analysis (HSA). HHT decomposes the signal to be decomposed into several IMFs (Intrinsic Mode Functions) using EMD, and then performs Hilbert transform on these IMF components to obtain the Hilbert spectrum, thereby performing time-frequency analysis on the signal. These IMFs are components that satisfy certain conditions; then, Hilbert transform is performed on each IMF to obtain the corresponding Hilbert spectrum, with each IMF represented in the joint time-frequency domain; finally, summing the Hilbert spectra of all IMFs yields the Hilbert spectrum of the original signal.

[0068] Step S21 specifically includes:

[0069] S211: Perform Hilbert transform on each IMF signal to obtain... , Where t represents time, Indicates frequency;

[0070] S212: Obtain the evoked response signal x(n) by formula x(n) = s(n) + u(n), where s(n) represents the actual signal included in the response x(n) after each stimulus, and u(n) represents the conventional noise signal included in the response x(n) after each stimulus;

[0071] S213: Based on The phase function is obtained from x(n) respectively. and instantaneous amplitude , , ,based on Get instantaneous frequency , ;

[0072] S214: Based on instantaneous amplitude and instantaneous frequency Original signal Perform HHT transformation to obtain ;

[0073] S215: Based on Obtaining the amplitude spectrum , .

[0074] Specifically, compared to wavelet analysis, the most significant feature of EMD is that it overcomes the problem of the lack of adaptability of basis functions. Wavelet analysis requires the selection of a specific wavelet basis, and the choice of the wavelet basis has a significant impact on the overall results. Once the wavelet basis is determined, it cannot be changed during the entire analysis process. Even if the wavelet basis may be optimal globally, it may not be optimal locally. Therefore, the basis functions of wavelet analysis lack adaptability. Performing a Hilbert transform on each IMF signal yields... This allows us to obtain the phase function. Instantaneous frequency Instantaneous amplitude The last sampled signal After HHT transformation, the following is obtained: ,Will Represented as amplitude spectrum .

[0075] Step S3 specifically includes:

[0076] The first type of IMF signal is weighted and attenuated, the second type of IMF signal is normally weighted, and all IMF signals are reconstructed to form the processed signal.

[0077] Based on formulas Acquire and process signals, where The attenuation coefficient is... , Indicates the processing signal. express The intrinsic modulus function, Represents the residual.

[0078] Specifically, according to The transient noise signal is analyzed by calculating the time-domain plot of the signal, and the noise signal is weighted and attenuated based on the formula. For m The original signal is reconstructed and combined into a processed signal. IMF components with transient noise are attenuated by a factor. weaken, coefficient This can be derived from experience gained during testing.

[0079] The value of Sd ranges from 0.2 to 0.3.

[0080] Specifically, DPOAE stands for Distortion Product Otoacoustic Emission, HHT stands for Hilbert-Huang Transform, and EMD stands for Empirical Mode Decomposition.

[0081] Specifically, the following implementation steps can be adopted in this application:

[0082] ① Collect a signal of a certain length ;

[0083] ② The signal is based on the formula EMD decomposition is performed using the criteria described above. Using EMD decomposition can overcome the lack of adaptability in the time-frequency domain basis functions. The final signal can be expressed by the formula... Express;

[0084] ③Analyze m items Signal, according to the formula Calculate each The HHT transform is used to analyze which IMF signal the transient noise belongs to. If there is no noise, the corresponding IMF signal cannot be identified.

[0085] ④According to The transient noise signal is analyzed by calculating the time-domain plot of the signal, and the noise signal is weighted and attenuated according to the formula. For m The original signal is reconstructed and combined to form a new original signal;

[0086] ⑤ Perform coherent averaging to calculate the DPOAE response.

[0087] The original signal is decomposed into different IMF signals using the EMD decomposition method. The presence of transient noise in each IMF signal is then identified. If transient noise is present, it is removed through weighted reconstruction, thereby eliminating the transient noise component from the original signal, reducing its impact on the coherent algorithm, and improving the signal-to-noise ratio.

[0088] The above-described contents can be implemented individually or in various combinations, and these variations are all within the protection scope of this invention.

[0089] It should be noted that in the description of this application, the terms "upper end," "lower end," and "bottom end," indicating orientation or positional relationship, are based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this application is in use. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. Moreover, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements, but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise limited, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition, characterized in that, Includes the following steps: S1: The raw signal acquired during the DPOAE test is decomposed into different IMF signals using EMD; S2: Perform time-frequency analysis on different IMF signals to obtain the first type of IMF signal containing transient noise and the second type of IMF signal not containing transient noise; S3: Weighted reconstruction of the first type IMF signal and the second type IMF signal are performed respectively to remove transient noise and form a processed signal; S4: The signal-to-noise ratio of the processed signal is improved by calculating the coherent averaging algorithm. Step S2 includes: transforming different IMF signals to obtain corresponding Hilberc spectra, performing time-frequency analysis on the IMF signals based on the Hilberc spectra, and if the IMF signal contains transient noise signals, then the IMF signal is marked as a first type of IMF signal; if the IMF signal does not contain transient noise signals, then the IMF signal is marked as a second type of IMF signal. Step S2 specifically includes: S21: Obtain the amplitude spectrum H(w, n), S22: Based on the amplitude spectrum H(w,n), perform analysis on each IMF. m (n) The signal is transformed by HHT transformation to obtain the transformed signal; S23: Perform time-domain plot analysis on each transformed signal to obtain transient noise signals. Transformed signals containing transient noise signals are labeled as Type I IMF signals, and transformed signals without transient noise signals are labeled as Type II IMF signals.

2. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 1, characterized in that, Step S1 specifically includes: S11: Acquire the raw signal I(n) of a preset length; S12: Apply the original signal I(n) based on the termination criterion EMD decomposition formation To obtain different IMF signals, where IMF m-1 (n) and IMF m (n) represents two time series in the process of screening IMF components, where T represents the signal duration and Sd represents the standard deviation. m (n) represents M th The intrinsic modulus function, Res M (n) represents the residual.

3. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 2, characterized in that... Step S12, EMD decomposition, specifically includes: S121: Draw the upper and lower envelope lines based on the upper and lower extreme points of the original signal; S122: Obtain the mean of the upper and lower envelopes, and draw the mean envelope based on the mean; S123: Subtract the mean envelope from the original signal to obtain the intermediate signal; S124: Determine whether the intermediate signal satisfies the intrinsic modulus function (IMF) determination principle. If it does, treat the intermediate signal as an IMF component and return to step S121 after subtracting the IMF component from the original signal to obtain the new original signal. If it does not satisfy the principle, return to step S121 after treating the intermediate signal as the new original signal.

4. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 3, characterized in that, The principle for determining the intrinsic modulus function is as follows: the number of local extrema and zero-crossing points of the function is equal or differs by at most one throughout the entire time range; at any given time point, the average of the envelopes of the local maximum and the envelopes of the local minimum is zero.

5. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 1, characterized in that, Step S21 specifically includes: S211: Perform a Hilbert transform on each IMF signal to obtain y(n). Where t represents time and τ represents frequency; S212: Obtain the evoked response signal x(n) by formula x(n)=s(n)+u(n), where s(n) represents the actual signal included in the response x(n) after each stimulus, and u(n) represents the conventional noise signal included in the response x(n) after each stimulus; S213: Obtain the phase function Φ(n) and instantaneous amplitude a(n) based on y(n) and x(n) respectively. The instantaneous frequency w(n) is obtained based on Φ(n). S214: Obtain the original signal I(n) by performing HHT transformation based on the instantaneous amplitude a(n) and instantaneous frequency w(n). S215: Based on Obtain the amplitude spectrum H(w, n).

6. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 1, characterized in that, Step S3 specifically includes: The first type of IMF signal is weighted and attenuated, the second type of IMF signal is normally weighted, and all IMF signals are reconstructed to form the processed signal.

7. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 6, characterized in that: Based on formulas Acquire and process the signal, where α m I'(n) represents the attenuation coefficient corresponding to the first type of IMF signal, and I′(n) represents the processed signal. m (n) represents M th The intrinsic modulus function, Res M (n) represents the residual.

8. The method for improving the signal-to-noise ratio of a signal during DPOAE testing using EMD decomposition as described in claim 2, characterized in that: The value of Sd ranges from 0.2 to 0.3.

Images