Thickness measurement signal processing method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention relates to the field of film measurement, and discloses a thickness measurement signal processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a sample thickness actual measurement signal generated by a to-be-measured sample through optical interference thickness measurement; the sample thickness actual measurement signal is modulated through a preset modulation function, a first processing signal is obtained, and the modulation function is generated based on overlapping parameters of an overlapping frequency band with the matching degree of an ideal thickness measurement signal of a preset thickness standard sample wafer and an actual thickness measurement signal of the thickness standard sample wafer reaching a preset standard in a frequency domain; the overlapping parameters comprise at least one of the central position, the width and the amplitude of the overlapping frequency band and the matching degree corresponding to the overlapping frequency band; and performing thickness analysis based on the first processing signal. According to the method, the frequency band carrying the correct thickness characteristic is screened out from the aliasing frequency and is purified, so that the thickness measurement and calculation precision of the signal containing noise is remarkably improved.

Description

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on June 4, 2025, with application number 2025107347336 and application name “Method, device, electronic device and storage medium for processing thickness measurement signals”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present invention relate to the field of thin film measurement technology, and in particular to a method, device, electronic device and storage medium for processing thickness measurement signals. Background Art

[0003] Currently, thin film thickness measurement methods are categorized into non-optical methods (including electrical and mechanical methods) and optical methods, with common methods including spectrophotometry, ellipsometry, confocal microscopy, and optical interferometry. Considering the application scenarios and measurement accuracy of in-line measurement for wafer thinning, optical interferometry based on reflectance spectroscopy is a thickness calculation algorithm that uses fiber optic technology to measure reflectivity curves. This method enables the combined assembly of signal transmitters and receivers, avoiding coaxial alignment. It is an effective means of achieving system miniaturization, with a simpler structure and easier integration, and is becoming a popular and mainstream research direction.

[0004] A review of current thin film thickness measurement methods reveals that optical interferometry is widely used in industrial environments due to its simplicity and efficiency. However, most interferometry-based film thickness calculation methods use reflectivity spectra or reflected light intensity spectra as intermediaries to determine film thickness, requiring highly stable light source intensity or the use of high-precision reference samples. Furthermore, due to their heavy reliance on reflection models, calculation distortion can occur when processing non-ideal signals containing noise. Summary of the Invention

[0005] The purpose of the present invention is to at least provide a method, device, electronic device and storage medium for processing thickness measurement signals, which can at least solve the problem that existing film thickness calculation methods based on optical interferometry are prone to calculation distortion when processing non-ideal signals containing noise, and at least improve the thickness measurement accuracy when processing non-ideal signals containing noise.

[0006] To solve the above technical problems, at least one embodiment of the present application provides a method for processing a thickness measurement signal, comprising: Obtaining a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; Modulating the sample thickness measurement signal using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on overlapping parameters of an overlapping frequency band in which a predetermined matching degree between an ideal thickness measurement signal of a thickness standard sample and an actual thickness measurement signal of the thickness standard sample in the frequency domain reaches a preset standard, the overlapping parameters including at least one of a center position, a width, and an amplitude of the overlapping frequency band, and a matching degree corresponding to the overlapping frequency band; Thickness analysis is performed based on the first processed signal.

[0007] In some optional embodiments, the modulation function includes an asymmetric window function or a dynamic filter function.

[0008] In some optional embodiments, the asymmetric window function includes an asymmetric Gaussian window function, the peak position of the asymmetric Gaussian window function is based on the center position of the overlapping frequency band, and the width of the asymmetric Gaussian window function is based on the width of the overlapping frequency band.

[0009] In some optional embodiments, performing thickness analysis based on the first processed signal includes: Converting the first processed signal to the frequency domain for peak search, using the found peak frequency as a calibration point, using frequency points with a preset frequency step before and after the calibration point on the frequency axis as boundary points, and determining the frequency range between the two boundary points as a selected frequency range, wherein the selected frequency range is used to determine the frequency band containing the thickness information; Converting the sample thickness measurement signal to the frequency domain, removing interference signals from the sample thickness measurement signal based on the frequency selection range, and then converting the sample thickness measurement signal after removing the interference signals to the time domain to obtain a correction signal; Thickness analysis is performed based on the corrected signal.

[0010] In some optional embodiments, removing interference signals from the sample thickness measurement signal based on the frequency selection range, and then converting the sample thickness measurement signal after the interference signals are removed to the time domain to obtain a correction signal, includes: Based on the frequency selection range, the measured signal of the sample thickness is divided into a signal within the frequency selection range and a signal outside the frequency selection range, the signal outside the frequency selection range is retained unchanged, and the signal within the frequency selection range is assigned zero to obtain a reconstructed signal; Converting the reconstructed signal to the time domain to obtain a second processed signal; The sample thickness measured signal is subtracted from the second processed signal to obtain the correction signal.

[0011] In some optional embodiments, removing interference signals from the sample thickness measurement signal based on the frequency selection range, and then converting the sample thickness measurement signal after the interference signals are removed to the time domain to obtain a correction signal, includes: retaining the signal of the sample thickness measurement signal within the selected frequency range, filtering out the signal outside the selected frequency range, and obtaining a third processed signal; The third processed signal is converted to the time domain to obtain the corrected signal.

[0012] In some optional embodiments, performing thickness analysis based on the corrected signal includes: Converting the corrected signal to the frequency domain for peak fitting to obtain fitting data; Target peak data is extracted from the fitting data, and thickness analysis is performed based on the target peak data.

[0013] In some optional embodiments, converting the corrected signal to the frequency domain for peak fitting to obtain fitting data includes: The corrected signal is zero-padded, and the corrected signal after zero-padded is transferred to the frequency domain for peak fitting to obtain the fitting data.

[0014] In some optional embodiments, the step of modulating the sample thickness measurement signal using a preset modulation function to obtain a first processed signal includes: Preprocessing the measured signal of the sample thickness, wherein the preprocessing includes at least one of noise reduction, detrending, and uniform sampling and reorganization; Extracting a data segment within a preset interval from the pre-processed sample thickness measurement signal to obtain a filtered signal; Performing empirical mode decomposition on the filtered signal, and modulating the decomposed intrinsic mode function using the modulation function to obtain the first processed signal, wherein the intrinsic mode function is the first decomposed intrinsic mode function component or a combination of multiple decomposed intrinsic mode function components.

[0015] In some optional embodiments, obtaining a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured includes: Obtain the sample thickness measurement signal generated by infrared light interferometry thickness measurement of the sample to be measured.

[0016] At least one embodiment of the present application further provides a device for processing a thickness measurement signal, comprising: A measured signal acquisition module is used to obtain a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; a first processed signal generating module, configured to modulate the sample thickness measurement signal using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on overlapping parameters of an overlapping frequency band in which a predetermined matching degree between an ideal thickness measurement signal of a thickness standard sample and an actual thickness measurement signal of the thickness standard sample in the frequency domain reaches a preset standard, the overlapping parameters including at least one of a center position, a width, an amplitude of the overlapping frequency band, and a matching degree corresponding to the overlapping frequency band; A thickness analysis module is configured to perform thickness analysis based on the first processed signal.

[0017] At least one embodiment of the present application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the above-mentioned method for processing thickness measurement signals.

[0018] At least one embodiment of the present application further provides a computer-readable storage medium storing a computer program, wherein the computer program implements the above-mentioned method for processing thickness measurement signals when executed by a processor.

[0019] The embodiments of the present application provide a method, device, electronic device and storage medium for processing thickness measurement signals, wherein the modulation function is generated based on the overlapping parameters of the overlapping frequency bands whose matching degree in the frequency domain between the ideal thickness measurement signal of the preset thickness standard sample and the actual thickness measurement signal of the thickness standard sample reaches the preset standard. The overlapping frequency band is the frequency band that carries the correct thickness characteristic. Therefore, the modulation function can be used to highlight the frequency band that carries the correct thickness characteristic in the actual thickness measurement signal of the sample, suppress other interference signals, and make the first processed signal obtained after modulation close to the ideal thickness measurement signal of the actual thickness of the sample to be measured. The accuracy of the thickness measurement result obtained by thickness analysis based on the first processed signal is greatly improved. This improves the situation in which the existing film thickness solution method based on optical interferometry is prone to solution distortion when processing non-ideal signals containing noise. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] One or more embodiments are exemplarily described by the figures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments.

[0021] Figure 1 is a flow chart of a method for processing a thickness measurement signal provided by one embodiment of the present application; Figure 2 is a schematic diagram of frequency broadening provided by an embodiment of the present application; Figure 3This is a schematic diagram showing a time-frequency spectrum comparison between an ideal thickness measurement signal and an actual thickness measurement signal of a thickness standard sample provided by an embodiment of the present application after Hilbert-Huang transform; Figure 4 This is a schematic diagram of the effect of adding an asymmetric window to the sample thickness measurement signal provided by an embodiment of the present application; Figure 5 An embodiment of the present application provides a signal Schematic diagram of the comparison with the decomposed eigenmode functions; Figure 6 Schematic diagram of an asymmetric Gaussian window function provided by an embodiment of the present application; Figure 7 is a schematic diagram of a frequency spectrum of a first processed signal provided by an embodiment of the present application; Figure 8 is a schematic diagram of a power density spectrum of a first processed signal provided by an embodiment of the present application; Figure 9 is a schematic diagram of a second processing signal provided by an embodiment of the present application; Figure 10 1 is a schematic diagram comparing an original sample thickness measurement signal and a corrected signal corrected_signal provided in an embodiment of the present application; Figure 11 This is a schematic diagram of a zero-padded signal obtained by performing zero-padded correction according to an embodiment of the present application; Figure 12 This is a schematic diagram of frequency stretching after performing fast Fourier transform on a zero-padded signal provided by an embodiment of the present application; Figure 13 1 is a schematic diagram of performing single-peak Gaussian fitting on a frequency domain signal after fast Fourier transform of a zero-padded signal, provided by an embodiment of the present application; Figure 14 This is a schematic diagram of comparing the power spectrum results of the original signal, the ideal thickness signal, and the frequency-selected signal after simultaneously performing Hilbert-Huang transform, provided by an embodiment of the present application; Figure 15 This is a statistical graph of thickness results of repeated tests on a sample with a thickness of 430 μm provided by an embodiment of the present application; Figure 16 This is a statistical graph of thickness results of repeated tests on a sample with a thickness of 200 μm provided by an embodiment of the present application; Figure 17 1 is a schematic structural diagram of a thickness measurement signal processing device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, each embodiment of the present application will be described in detail below with reference to the accompanying drawings. However, it will be understood by those skilled in the art that in each embodiment of the present application, many technical details are proposed to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in the present application can be implemented. The division of the following embodiments is for convenience of description and should not constitute any limitation on the specific implementation of the present application. The various embodiments can be combined with each other and referenced to each other under the premise of no contradiction.

[0023] Most current film thickness calculation methods based on optical interferometry use reflectivity spectra as an intermediary to determine film thickness. This requires highly stable light source intensity or the use of high-precision reference samples. Furthermore, due to their heavy reliance on the reflectivity model, calculation distortion is prone to occur when processing non-ideal signals containing noise.

[0024] The technical problem addressed by this application is how to use noisy reflectance spectra to establish a highly accurate reflectance model to obtain accurate film thickness information. Conventional methods often use signal processing and spectrum analysis based on fast Fourier transforms to address this problem. While these methods establish relatively accurate reflectance models (reflectivity models or reflected light intensity models), they can lead to multi-peak broadening.

[0025] In view of the fact that non-ideal signals in the actual acquisition process contain noise and frequency aliasing that causes solution distortion, this application proposes a thickness measurement signal processing method to solve the problem of multi-peak broadening caused by frequency aliasing when performing thickness analysis on the thickness measurement signal.

[0026] The implementation details of the thickness measurement signal processing method of this embodiment are described in detail below. The following content is only provided for easy understanding and is not necessary for implementing this solution.

[0027] In order to better understand this solution, the principle of optical interferometry is first explained here: The relationship between interference light intensity (reflectivity), wavelength and film is as follows

[0028] Where d is the thickness of the film or plate being measured, n is the refractive index of the sample being measured, I(λ) is the intensity of the light source, and λ is the wavelength.

[0029] That is, under ideal conditions, the interference period of the interference light intensity is related to the thickness and refractive index of the film to be measured. ; in, is the interference period, is the sampling frequency, is wavelength 1, The wavelength is 2.

[0030] Using FFT or other methods to convert time domain signals into frequency domain signal analysis and extract frequency can improve the resolution and reliability of the calculation results. The relationship between peak frequency and refractive index and thickness can be expressed as: ; ; is the peak frequency. According to the above formula, the thickness information of the sample to be measured under the ideal interference signal can be quickly analyzed.

[0031] Example 1: The thickness measurement signal processing method of this embodiment can be applied to electronic devices with communication, computing and data storage capabilities. The specific process can be as follows: Figure 1 Shown, including: Step 110, obtaining a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; In this embodiment, the sample to be measured can be an object such as a thin film or a flat plate. Optical interferometry thickness measurement is a non-contact measurement technology based on the principle of optical interference. The sample thickness measured signal in this embodiment can be an interference signal formed by the light intensity spectrum of the upper and lower surfaces of the sample to be measured, or it can be an interference signal formed by the reflectivity spectrum of the upper and lower surfaces. By analyzing the frequency and phase of the interference signal, the thickness of the sample can be calculated. In some optional embodiments, the sample thickness measured signal generated by infrared light interferometry thickness measurement of the sample to be measured is obtained and processed.

[0032] Step 120: Modulating the sample thickness measurement signal using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on overlapping parameters of an overlapping frequency band in which a predetermined matching degree between an ideal thickness measurement signal of a thickness standard sample and an actual thickness measurement signal of the thickness standard sample in the frequency domain reaches a preset standard, wherein the overlapping parameters include at least one of a center position, a width, an amplitude of the overlapping frequency band, and a matching degree corresponding to the overlapping frequency band; It's understandable that when performing thickness analysis on a thickness measurement signal, taking the Fourier transform as an example, theoretically, a thickness measurement signal will have only one peak frequency after a Fast Fourier Transform (FFT), and the corresponding thickness result can be obtained based on this peak frequency. However, in practice, directly performing a FFT on the acquired signal often results in multiple peaks of a certain width, indicating frequency broadening. Therefore, it's impossible to determine the peak frequency corresponding to the thickness.

[0033] like Figure 2Figure 2 shows a schematic diagram of frequency broadening after direct Fast Fourier Transform (FFT) processing of the measured sample thickness signal. The horizontal axis represents frequency (Hz), and the vertical axis represents amplitude. Fast Fourier Transform processing of the full spectrum (0-512) and various sub-segments (0-256, 257-512) reveals varying degrees of frequency broadening, with peak frequencies located at different locations. This indicates that the acquired signal is a mixture of different frequency components, making it difficult to effectively extract the peak frequency. The accuracy of the peak frequency directly impacts the thickness measurement. Therefore, signal processing such as filtering and noise reduction is necessary. From the aliased frequency signal, the peak frequency can be extracted to further analyze the thickness. The core of interferometric thickness measurement based on Fast Fourier Transform (FFT) lies in feature extraction from the real-time acquired signal. The inability to accurately extract the peak frequency due to frequency aliasing directly impacts the accuracy of thickness measurement.

[0034] When frequency broadening exists, the sample thickness measurement signal is modulated using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on an overlapping parameter of an overlapping frequency band in which the ideal thickness measurement signal of the preset thickness standard sample and the actual thickness measurement signal of the thickness standard sample have a matching degree in the frequency domain that meets a preset standard.

[0035] In this embodiment, the modulation function is a pre-built function that can be used for samples of any thickness to be tested, and is used to improve the accuracy of extracting the main lobe peak of the sample thickness measurement signal. The specific construction process is as follows: 1) Calculate the ideal thickness measurement signal of the thickness standard sample. It can be calculated according to the following formula: Theoretical peak frequency as follows:

[0036] For example, for a 400um silicon wafer with an average refractive index of 3.45, the characteristic peak frequency is:

[0037] According to the theoretical periodic function, it is as follows:

[0038] in, Based on the above theoretical periodic function, by inputting the average refractive index, corresponding wavelength, thickness and other parameters, the theoretical curve of the ideal thickness measurement signal of the thickness standard sample can be obtained.

[0039] 2) Convert the ideal thickness measurement signal of the preset thickness standard sample and the actual thickness measurement signal of the thickness standard sample into the frequency domain for matching, and determine the overlapping frequency band whose matching degree meets the preset standard; Specifically, the ideal thickness measurement signal and the actual thickness measurement signal of the thickness standard sample are processed by Hilbert-Huang transform respectively, and the theoretical signal and the measured signal are converted into power density spectrum for analysis.

[0040] like Figure 3 The figure shows the comparison between the ideal thickness measurement signal and the actual thickness measurement signal of the thickness standard sample. The horizontal axis in the figure is the wave number ( ), with frequency (Hz) on the vertical axis. The figure shows a comparison of the ideal and measured thickness signals of two different thickness standard samples. Select the upper set of comparison graphs. The dark green signal, with its smooth and regular variation, represents the ideal thickness measurement signal of the thickness standard sample, while the light green signal, with its uneven peaks and skewed distribution, represents the actual thickness measurement signal of the thickness standard sample. It can be seen that there is overlap between the ideal and actual thickness measurement signals (indicated by the box). When the matching degree of these overlapping portions meets the preset standard, it can be assumed that the actual and ideal thickness measurement signals have a high degree of similarity in the frequency domain corresponding to the overlapping portions. This similarity matching determines the band in the actual thickness measurement signal that carries the correct thickness characteristic frequency.

[0041] For example, a spectrometer with 512 points between 1510 and 1590 nm may produce a thickness measurement signal (or a sample thickness measurement signal) with frequency aliasing. The correct thickness frequency may lie in the 1530-1540 nm band (i.e., the band corresponding to the overlapping portion). In this embodiment, the Hilbert-Huang transform is pre-processed on the power spectrum of the ideal and actual thickness measurement signals of the thickness standard sample to determine the band with the correct thickness characteristic frequency. Practical verification has shown that the error between the thickness value obtained for the thickness standard sample and the actual value is within the allowable range based on the overlapping frequency bands.

[0042] When measuring samples of other thicknesses, after the measured thickness signal of the sample is converted to the frequency domain, the frequency band carrying the correct thickness information will basically fall within the above-mentioned matched overlapping frequency band.

[0043] 3) Constructing a modulation function based on the overlapping parameters of the overlapping part. The overlapping parameters include at least one of the center position, width, amplitude of the overlapping frequency band and the matching degree corresponding to the overlapping frequency band. Correspondingly, there can be multiple types of modulation functions, and different modulation functions can be constructed based on different overlapping parameters. The purpose of the modulation function is to highlight the frequency band where the matching degree between the ideal thickness measurement signal of the preset thickness standard sample and the actual thickness measurement signal of the thickness standard sample meets the standard, and suppress the signals of other frequency bands, thereby achieving the purpose of purifying the signal. Similarly, by using the constructed modulation function to modulate the measured thickness information of the sample, the frequency band carrying the correct thickness signal in the measured thickness signal of the sample can also be highlighted. Therefore, after convolving the measured thickness signal of the sample with the modulation function, a first processed signal is obtained. The thickness information frequency band contained in the first processed signal accounts for a higher proportion, which helps to improve the accuracy of subsequent thickness analysis.

[0044] In some possible embodiments, the portion of the ideal thickness measurement signal of the thickness standard sample that matches the actual thickness measurement signal of the sample is input as a parameter into a preset matching algorithm, and the matching degree or similarity that reaches a preset standard, such as 90% or above, is considered to be a matching interval, that is, an overlapping portion.

[0045] Step 130: Perform thickness analysis based on the first processed signal.

[0046] Specifically, converting the first processed signal to the frequency domain significantly reduces the broadening of the signal, resulting in a prominent single peak. Extracting this peak for thickness analysis yields a more accurate thickness measurement.

[0047] In summary, in this embodiment, the modulation function is generated based on the overlapping parameters of the overlapping frequency bands whose frequency domain matching degree between the ideal thickness measurement signal of the preset thickness standard sample and the actual thickness measurement signal of the thickness standard sample reaches the preset standard. The overlapping frequency band is the frequency band that carries the correct thickness characteristic. Therefore, the modulation function can be used to highlight the frequency band that carries the correct thickness characteristic in the sample thickness measurement signal, suppress other interference signals, and make the first processed signal obtained after modulation close to the ideal thickness measurement signal of the actual thickness of the sample to be measured. The accuracy of the thickness measurement result obtained by thickness analysis based on the first processed signal is greatly improved. This improves the existing film thickness solution method based on optical interferometry, which is prone to solution distortion when processing non-ideal signals containing noise.

[0048] In some optional embodiments, the modulation function includes an asymmetric window function or a dynamic filter function. When constructing an asymmetric window modulation function, it can be constructed based on the center position and overlap width of the overlapping frequency bands. When constructing a dynamic filter modulation function, it can be constructed based on the matching degree corresponding to the overlapping frequency bands. The specific modulation function selected depends on actual needs and is not limited here.

[0049] It's worth noting that in this embodiment, an asymmetric window function was chosen over a symmetric one to more accurately highlight the frequency information in the overlapping regions. In actual measurements, for thicker samples, the aliasing frequency increases, and after the fast Fourier transform, a high number of aliasing frequencies remain. A symmetric window function can erroneously increase the sidelobe energy, affecting the accuracy of mainlobe peak extraction. Therefore, choosing an asymmetric window function can better adapt to the measurement requirements of different thicknesses.

[0050] In some optional embodiments, a dynamic filter function is used to modulate the first processed signal. For example, based on the above-mentioned Hilbert-Yang and STFT short-time Fourier transform results, it is determined that the portion carrying the thickness feature is mainly located at 1530-1545nm (0.647-0.654). In the sample thickness measured signal spectrum range of 1510-1590nm, the number of different smoothing points is adjusted for each 5nm segment to change the weight of the sample thickness measured signal. The higher the matching degree of the frequency band, the lower the number of smoothing points. According to this state filter function, the thickness of different samples is measured, and the thickness results obtained have a small error with the actual value.

[0051] In one embodiment, the asymmetric window function includes an asymmetric Gaussian window function, the peak position of the asymmetric Gaussian window function is based on the center position of the overlapping frequency band, and the width of the asymmetric Gaussian window function is based on the width of the overlapping frequency band.

[0052] like Figure 4 The figure shows the effect of adding an asymmetric window to the actual signal of sample thickness. The horizontal axis in the figure is the wave number ( ), the vertical axis is the amplitude. This part of the signal is the first processed signal. Figure 6 As shown in the figure, it is a schematic diagram of the asymmetric Gaussian window function. The horizontal axis in the figure is the wave number ( ), with the vertical axis representing the weight / percentage. The center position determines the center of symmetry of the Gaussian function. The Gaussian function reaches its maximum at the center position and is symmetric about the center position. The overlap width determines the spread (i.e., standard deviation) of the Gaussian function. A larger overlap width indicates a wider Gaussian function; a smaller overlap width indicates a narrower Gaussian function.

[0053] In this embodiment, the asymmetric Gaussian window function has a flexible shape adjustment capability, can better adapt to the characteristics of different non-stationary signals, reduce boundary effects, and improve the local feature extraction capability.

[0054] In some optional embodiments, the asymmetric window function may also use an asymmetric probability distribution function, such as a gamma distribution, a Weibull distribution, a lognormal distribution, etc. The left and right coefficients of the symmetric window function may also be adjusted, such as using different coefficients on the left and right sides of the Hanning window to generate an asymmetric window function.

[0055] In an optional embodiment, the thickness analysis based on the first processed signal includes: converting the first processed signal to the frequency domain for peak search, using the peak frequency found as the calibration point, and using the frequency points on the frequency axis with a preset frequency step before and after the calibration point as boundary points, and determining the frequency band range between the two boundary points as the selected frequency range, and the selected frequency range is used to determine the frequency band containing thickness information; converting the sample thickness measured signal to the frequency domain, eliminating the interference signal in the sample thickness measured signal based on the selected frequency range, and then converting the sample thickness measured signal after the interference signal is eliminated to the time domain to obtain a corrected signal; and performing thickness analysis based on the corrected signal.

[0056] Specifically, the first processed signal is transferred to the frequency domain for peak search, which can be done by performing FFT on the first processed signal to obtain the amplitude spectrum Y_magnitude of the first processed signal. The amplitude spectrum is as follows: Figure 7 shown. Figure 7 The horizontal axis is frequency (Hz) and the vertical axis is amplitude. You can also use the spectrum analysis tool pwelch to perform power density spectrum analysis. The power density spectrum is as follows: Figure 8 shown. Figure 8 The horizontal axis represents frequency (Hz), and the vertical axis represents energy density. In this embodiment, FFT is performed on the first processed signal to perform peak search.

[0057] After that, the threshold is determined by frequency peak extraction. If there are multiple peaks, the peaks are found according to certain parameters, such as using functions such as findpeak.

[0058] In this embodiment, the frequency selection range is determined by taking the peak frequency as the calibration point. The average of the peak values ​​found can be used as the calibration point (or frequency selection threshold) threshold_freq to specify the frequency selection threshold. The frequency point of 200hz is used as threshold_freq1 and threshold_freq2 (i.e. boundary points) to determine the frequency selection range. It should be noted that the preset frequency step size can be 200hz, can also be 50hz, or 1 Hz. Select according to actual situation, no restriction here. The frequency selection range further refines the frequency band containing thickness information.

[0059] Next, perform a fast Fourier transform on the measured signal of the sample thickness, eliminate the interference signals in the measured signal of the sample thickness based on the selected frequency range, and then transfer the measured signal of the sample thickness after eliminating the interference signals to the time domain to obtain a corrected signal; perform thickness analysis based on the corrected signal. After the processing of the above steps, the proportion of the effective frequency band carrying the correct thickness information in the corrected signal increases significantly. After performing a fast Fourier transform on the corrected signal, the number of its peaks and broadening are further reduced. At this time, extract the first peak data after performing a fast Fourier transform on the corrected signal, and perform thickness analysis based on the first peak data, and the accuracy of the analyzed thickness data is significantly improved.

[0060] In an optional embodiment, the eliminating the interference signals in the measured signal of the sample thickness based on the selected frequency range, and then transferring the measured signal of the sample thickness after eliminating the interference signals to the time domain to obtain a corrected signal includes: based on the selected frequency range, dividing the measured signal of the sample thickness into signals within the selected frequency range and signals outside the selected frequency range, keeping the signals outside the selected frequency range unchanged, setting the signals within the selected frequency range to zero to obtain a reconstructed signal; transferring the reconstructed signal to the time domain to obtain a second processed signal; subtracting the second processed signal from the measured signal of the sample thickness to obtain the corrected signal.

[0061] Specifically, for example, the calibration point is threshold_freq, and the selected frequency threshold is specified. The frequency point of 200 hz is used as threshold_freq1 and threshold_freq2 (i.e., the boundary points). After performing a fast Fourier transform on the measured signal of the sample thickness, it is specified that in the frequency-domain signal after the transform, f > threshold_freq1 and f < threshold_freq2 are signals outside the frequency range, and the values of this part of the signals remain unchanged. The signals within the range of threshold_freq 200 hz are signals within the frequency range, and the values of this part of the signals are set to 0 to obtain a reconstructed signal. The reconstructed signal retains the low-frequency noise and high-frequency interference in the measured signal of the sample thickness.

[0062] Perform an inverse Fourier transform IFFT on the first processed signal (i.e., the reconstructed signal) after assignment to obtain a second processed signal recovered_signal. As Figure 9 shown, the time-domain signal below is a schematic diagram of the second processed signal. Figure 9 In it, the abscissa is the wave number ( ), and the ordinate is the amplitude.

[0063] Then, the second processed signal recovered_signal is subtracted from the sample thickness measured signal to obtain the corrected signal corrected_signal. Figure 10 The figure shows a comparison diagram of the original sample thickness measured signal (blue) and the corrected signal corrected_signal (yellow). Figure 10 The horizontal axis is the wave number ( ), the ordinate is the amplitude. At this point, the low-frequency noise and high-frequency interference in the original sample thickness measurement signal are eliminated, and the corrected signal is obtained.

[0064] In this embodiment, the signal is divided into signals within the selected frequency range and signals outside the selected frequency range based on the selected frequency range, achieving precise signal separation. The signals within the selected frequency range are assigned zero, while the signals outside the selected frequency range remain unchanged. An inverse Fourier transform is then performed on the signals outside the selected frequency range to generate a second processed signal, which restores low-frequency noise and high-frequency interference. The measured sample thickness signal is subtracted from the second processed signal to remove unwanted frequency components while retaining the characteristic thickness frequency components, resulting in a corrected signal.

[0065] In some optional embodiments, the interference signal in the sample thickness measurement signal is eliminated based on the frequency selection range, and then the sample thickness measurement signal after the interference signal is eliminated is converted to the time domain to obtain a correction signal, including: retaining the signal of the sample thickness measurement signal within the frequency selection range, filtering out the signal outside the frequency selection range, and obtaining a third processed signal; converting the third processed signal to the time domain to obtain the correction signal.

[0066] Specifically, filtering out signals outside the selected frequency range can be achieved by setting a bandpass filter corresponding to the selected frequency range, or by selecting signal components of a specific scale (the selected frequency range) through wavelet decomposition. The method of selecting the appropriate method depends on the actual situation and is not limited here.

[0067] In some optional embodiments, the method of modulating the measured thickness signal of the sample using a preset modulation function to obtain a first processed signal includes: preprocessing the measured thickness signal of the sample, wherein the preprocessing includes at least one of noise reduction, detrending and uniform sampling and recombination; extracting data segments within a preset interval in the preprocessed measured thickness signal of the sample to obtain a filtered signal; performing empirical mode decomposition on the filtered signal, and modulating the decomposed intrinsic mode function using the modulation function to obtain the first processed signal, wherein the intrinsic mode function is the first decomposed intrinsic mode function component or a combination of multiple decomposed intrinsic mode function components.

[0068] Specifically, the process of preprocessing and empirical mode decomposition of the sample thickness measured signal includes the following: a. Detrend the measured signal of sample thickness to obtain ; b.Yes Perform wavelet transform to reduce noise and obtain cleaned_signal. When using wavelet transform for noise reduction, the wavelet coefficients, number of denoising layers, threshold, etc. can be adjusted based on the specific characteristics of the sample thickness measured signal and the noise reduction goal.

[0069] c. Calculate the average sampling frequency (Non-uniform sampling): Calculate wave number ,in, is the wavelength; ,in, The function calculates the difference between adjacent wave numbers, and abs takes the absolute value; Calculating the sampling frequency ; Calculate the average sampling frequency

[0070] d. Use Butterworth bandpass filter for filtering, and the parameters can be adjusted including the low cutoff frequency , high cutoff frequency , the order of the filter and passband ripple , and get the filtered signal In some optional embodiments, it can be replaced by Cheby filtering or the like.

[0071] e. Interpolation uniform recombination, Perform uniform sampling and obtain . Cubic spline interpolation or linear interpolation can be used, and the number of interpolation points is adjustable; f.Yes Select the points in the preset interval (such as the 10th to 400th points in 512 points) and get , that is, the above-mentioned filtered signal is obtained. The valid data interval is selected to eliminate the data with large front-end and back-end noise. In some optional embodiments, all 1:512 data are selected.

[0072] g.Yes Perform empirical mode decomposition to obtain multiple intrinsic mode function components. Figure 5 As shown, Schematic comparison of (a) and the first decomposed intrinsic mode function (b). Figure 5 The horizontal axis is the wave number ( ), where the ordinate is the amplitude. In actual processing, the first decomposed IMF component, IMF1, can be convolved with an asymmetric window function. Alternatively, the IMF obtained by combining multiple IMF components can be convolved with an asymmetric window function. For example, if IMF2 contains significant characteristic information, IMF1 and IMF2 can be combined to reconstruct a single IMF component.

[0073] The modulation function is used to modulate the intrinsic mode function to obtain the first processed signal, thereby completing the preliminary screening of the frequency bands carrying the correct thickness characteristic in the measured sample thickness signal. This process improves the quality of the measured sample thickness signal through signal preprocessing. On this basis, empirical mode decomposition is further performed, and the modulated signal is used to modulate the decomposed intrinsic mode function. This further enhances the local characteristics of the measured sample thickness signal, reduces boundary effects, and significantly improves the accuracy and effectiveness of signal analysis.

[0074] In some optional embodiments, the thickness analysis based on the corrected signal includes: converting the corrected signal to the frequency domain for peak fitting to obtain fitting data; extracting target peak data from the fitting data, and performing thickness analysis based on the target peak data.

[0075] In actual processing, after the correction signal is transformed by FFT, there may still be a certain degree of broadening. The reason is that in the step of determining the frequency selection range by using the frequency corresponding to the first peak data as the calibration point, the frequency selection threshold threshold_freq parameter selection range is relatively large, for example, threshold_freq 200 Hz. Considering that the ideal characteristic frequency conversion range of the refractive index converter is approximately +-1 Hz, a too small frequency selection threshold can easily lose the original signal and cause system errors. Therefore, by appropriately increasing the selected frequency range, when the frequency domain information has a certain degree of broadening but also has a clear peak trend, further peak fitting can be used to obtain the peak frequency, thereby improving the accuracy of peak frequency calculation and, in turn, the accuracy of thickness measurement.

[0076] In some optional embodiments, converting the corrected signal to the frequency domain for peak fitting to obtain fitting data includes: performing zero-padding on the corrected signal, converting the zero-padding corrected signal to the frequency domain for peak fitting to obtain the fitting data.

[0077] like Figure 11 As shown, the zero-padded signal padded_signal is obtained by padding the correction with zero. Figure 11 The horizontal axis is the data frame, and the vertical axis is the amplitude. The fast Fourier transform (FFT) of the zero-filled signal padded_signal is performed, and the frequency broadening is as follows Figure 12 shown. Figure 12 The horizontal axis is frequency (Hz), and the vertical axis is amplitude. The peak value after fast Fourier transformation of the zero-filled correction signal is fitted to obtain fitting data.

[0078] In this embodiment, zero padding is performed on the correction signal, so that the interpolation points of the spectrum after the correction signal is subjected to fast Fourier transform are increased, the position of the peak is more accurate, and the details of the thickness characteristic frequency can be more easily distinguished.

[0079] In some optional embodiments, the fitting of the peak value after fast Fourier transform of the zero-padded corrected signal includes: fitting the peak value after fast Fourier transform of the zero-padded corrected signal using a unimodal Gaussian function.

[0080] like Figure 13 Figure 1 shows a schematic diagram of a single-peak Gaussian fit performed on the frequency domain signal after Fast Fourier Transform (FFT) of the zero-padded signal padded_signal. The horizontal axis in the figure is frequency (Hz) and the vertical axis is amplitude.

[0081] In this embodiment, the single-peak Gaussian function has a simple mathematical form, a high computational efficiency in the fitting process, and can quickly and accurately describe the peak shape. In some optional embodiments, a Lorentz fitting, a Voigt fitting, or other methods can also be used to fit the peak. The fitting function is selected based on actual needs and is not specifically limited here.

[0082] In some optional embodiments, after constructing a modulation function based on the overlapped portion of the power spectrum between the ideal thickness measurement signal of the thickness standard sample and the actual thickness measurement signal of the thickness standard sample, which matches the preset standard, the modulation function is convolved with the actual thickness measurement signal of the thickness standard sample to obtain a third processed signal. A frequency threshold is determined as in the above embodiment, a frequency selection range is set based on the frequency threshold, and then the third processed signal is assigned a value to remove interference signals from the third processed signal to obtain a fourth processed signal. The fourth processed signal is subjected to FFT and peak fitting, and the thickness measurement value is calculated based on the characteristic peak frequency.

[0083] Next, the modulation function parameters are adjusted based on the thickness measurement compared to the reference thickness. For example, if the modulation function is an asymmetric Gaussian window, hyperparameters such as mu, sigma, and the findpeak function, threshold_freq1, and 2 are adjusted to ensure the algorithm matches the reference sample. The calculated results from each parameter selection are compared with the known thickness, and the various parameters are adjusted at each stage to achieve calibration. After calibration, unknown thicknesses are measured using the fixed parameters.

[0084] Example 2: Based on the above embodiment, this embodiment provides an application example. The thickness measurement signal processing method in this embodiment can be applied to electronic devices with communication, computing and data storage capabilities, as follows: Step 210, obtaining a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; Step 220 , preprocessing the measured signal of the sample thickness, including conventional detrending, wavelet transform noise reduction, calculation of average sampling frequency (non-uniform sampling), Butterworth bandpass filter and interpolation uniform reconstruction; Step 230, performing EMD modal decomposition on the pre-processed sample thickness measured signal to obtain the first intrinsic mode function IMF (:, 1); Step 240: Obtain a preset modulation function, which is an asymmetric Gaussian window function. Add the asymmetric Gaussian window function to the IMF(:, 1) to obtain a first processed signal. The construction process and usage of the asymmetric Gaussian window function in this embodiment are similar to those in the above embodiment and will not be repeated here to avoid repetition. Step 250, performing fast Fourier transform processing on the first processed signal, extracting the peak frequency as a frequency threshold, and determining a frequency selection range based on the frequency domain; Step 260: Perform FFT transformation on the measured signal of sample thickness, assign values ​​to the signals inside and outside the selected frequency range, and restore the spectrum distribution and noise characteristics. The data outside the range of 50 are subjected to inverse Fourier transform IFFT to reconstruct the spectral distribution of high-frequency interference and low-frequency light source to obtain the reconstructed signal.

[0085] Step 270 , subtracting the reconstructed signal from the sample thickness measured signal to obtain a corrected signal; In step 280, the corrected signal is subjected to FFT processing to obtain a second frequency domain signal with obvious peak characteristics; a single-peak Gaussian function is fitted to the second frequency domain signal to extract peak data. Optionally, the fitting model can be replaced with a Lorentz model or a high-order polynomial fitting model.

[0086] Step 290: Output the thickness result based on the peak value data and the thickness calculation formula.

[0087] Effect verification: like Figure 14 As shown in the figure, the power spectrum results of the ideal thickness measurement signal (yellow) of the thickness standard sample are compared with the actual thickness measurement signal (blue) and the correction signal (gray) after Hilbert Huang transform. The horizontal axis in the figure is the wave number ( ), with amplitude as the ordinate. The actual thickness measurement signal contains many aliasing frequencies, and the corrected signal is essentially close to the ideal signal periodic frequency. This technical approach effectively decomposes the frequency characteristics related to thickness from the aliasing frequencies. This allows for accurate extraction of effective thickness information using fast Fourier transforms and other techniques to calculate the thickness of the measured sample.

[0088] Under this technical route, the repeatability of the thickness calculation results after processing the actual collected data is less than 0.1um. Figure 15 and Figure 16 As shown, the test data are for two pieces with different standard thicknesses (430um, 200um). Figure 15 The test data of three different test points of 430um sample are shown below. Figure 15 Figures a1, b1, and c1 show the thickness results of 100 measurements at Point 1, Point 2, and Point 3 on a standard thickness sample. The horizontal axis in Figures a1, b1, and c1 represents the number of measurements, and the vertical axis represents the thickness (µm). Figure a2 shows the frame data from a single measurement in Figure a1, i.e., the actual thickness signal. Similarly, b2 and c2 show the frame data from a single measurement in b1 and c1, respectively. The horizontal axis in Figures a2, b2, and c2 represents the wavelength (nm), and the vertical axis represents the reflected light intensity.

[0089] Similarly, Figure 16 The test data of three different test points for 200um sample, Figure 16 The meaning of each figure is the same as Figure 15 The same, no further description here.

[0090] Signals were collected at three different measurement points on samples of different materials, and the stability was verified. The thickness average and repeatability results were as follows:

[0091] Example 3: Another embodiment of the present application relates to a thickness measurement signal processing device. The implementation details of the thickness measurement signal processing device of this embodiment are specifically described below. The following content is only for the convenience of understanding the implementation details and is not necessary for the implementation of this solution. The schematic diagram of the thickness measurement signal processing device of this embodiment can be as follows: Figure 17 As shown, it includes a thickness measurement signal acquisition module 2010 , a first processed signal generation module 2020 and a thickness analysis module 2030 .

[0092] The thickness measurement signal acquisition module 2010 is used to obtain a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; a first processed signal generating module 2020 configured to modulate the sample thickness measurement signal using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on overlap parameters of an overlapping frequency band in which a predetermined matching degree between an ideal thickness measurement signal of a thickness standard sample and an actual thickness measurement signal of the thickness standard sample in the frequency domain reaches a preset standard, wherein the overlap parameters include at least one of a center position, an overlap width, an amplitude, and a matching degree corresponding to the overlapping frequency band; The thickness analysis module 2030 is configured to perform thickness analysis based on the first processed signal.

[0093] It is worth mentioning that all modules involved in this embodiment are logical modules. In actual applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, to highlight the innovation of this application, this embodiment does not include units that are not closely related to solving the technical problem proposed by this application. However, this does not mean that other units do not exist in this embodiment.

[0094] In an optional embodiment, the thickness analysis module includes: a frequency selection range determining unit, configured to convert the first processed signal into the frequency domain for peak search, use the found peak frequency as a calibration point, use frequency points on the frequency axis that are a preset frequency step away from the calibration point as boundary points, and determine the frequency range between the two boundary points as a frequency selection range, wherein the frequency selection range is used to determine the frequency band containing thickness information; a correction signal generating unit, configured to convert the sample thickness measurement signal into the frequency domain, remove interference signals from the sample thickness measurement signal based on the frequency selection range, and then convert the sample thickness measurement signal after the interference signals are removed into the time domain to obtain a correction signal; The thickness analysis unit is used to perform thickness analysis based on the correction signal.

[0095] In an optional embodiment, the correction signal generating unit includes: a reconstruction subunit, configured to divide the measured signal of the sample thickness into a signal within the frequency selection range and a signal outside the frequency selection range based on the frequency selection range, retain the signal outside the frequency selection range unchanged, assign zero to the signal within the frequency selection range, and obtain a reconstructed signal; a second processed signal generating subunit, configured to convert the reconstructed signal into a time domain to obtain a second processed signal; The second correction subunit is used to subtract the sample thickness measured signal from the second processed signal to obtain the correction signal.

[0096] In an optional embodiment, the correction signal generating unit includes: The first correction subunit is used to retain the signal of the sample thickness measurement signal within the selected frequency range, filter out the signal outside the selected frequency range, and obtain the correction signal.

[0097] a third processed signal generating subunit, configured to retain the signal of the sample thickness measured signal within the selected frequency range and filter out the signal outside the selected frequency range to obtain a third processed signal; The third correction subunit is used to convert the third processed signal into the time domain to obtain the corrected signal.

[0098] In an optional embodiment, the thickness analysis unit includes: A fitting subunit, configured to convert the corrected signal into the frequency domain for peak fitting to obtain fitting data; The analysis subunit is used to extract target peak data from the fitting data and perform thickness analysis based on the target peak data.

[0099] In an optional embodiment, the fitting subunit is used to perform zero-padding processing on the corrected signal, and transfer the zero-padding corrected signal to the frequency domain for peak fitting to obtain the fitting data.

[0100] In an optional embodiment, the first processing signal generating module includes: a preprocessing unit, configured to preprocess the sample thickness measurement signal, wherein the preprocessing includes at least one of noise reduction, detrending, and uniform sampling and reorganization; a screening unit, configured to extract data segments within a preset interval from the pre-processed signal of the sample thickness measurement to obtain a screened signal; An empirical mode decomposition unit is used to perform empirical mode decomposition on the filtered signal, and modulate the decomposed intrinsic mode function using the modulation function to obtain the first processed signal, wherein the intrinsic mode function is the first decomposed intrinsic mode function component or a combination of multiple decomposed intrinsic mode function components.

[0101] Example 4: Another embodiment of the present application relates to an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the thickness measurement signal processing method of the above-mentioned embodiments.

[0102] The memory and processor are connected using a bus, which can include any number of interconnected buses and bridges. The bus connects various circuits of one or more processors and memories. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits. These are all well known in the art and are therefore not described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single component or multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over a wireless medium via an antenna. Furthermore, the antenna receives data and transmits it to the processor.

[0103] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0104] Embodiment 5: Another embodiment of the present application relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0105] That is, those skilled in the art will understand that all or part of the steps in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a program. The program is stored in a storage medium and includes a number of instructions for causing a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps in the methods described in the various embodiments of this application. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0106] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application, and that in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present application.

Claims

1. A method for processing a thickness measurement signal, characterized in that: include: Obtaining a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; Modulating the sample thickness measurement signal using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on overlapping parameters of an overlapping frequency band in which a predetermined matching degree between an ideal thickness measurement signal of a thickness standard sample and an actual thickness measurement signal of the thickness standard sample in the frequency domain reaches a preset standard, the overlapping parameters including at least one of a center position, a width, and an amplitude of the overlapping frequency band, and a matching degree corresponding to the overlapping frequency band; Thickness analysis is performed based on the first processed signal.

2. The method for processing thickness measurement signals according to claim 1, wherein: The modulation function includes an asymmetric window function or a dynamic filter function.

3. The method for processing thickness measurement signals according to claim 2, wherein: The asymmetric window function includes an asymmetric Gaussian window function, a peak position of the asymmetric Gaussian window function is determined based on a center position of the overlapping frequency band, and a width of the asymmetric Gaussian window function is determined based on a width of the overlapping frequency band.

4. The method for processing thickness measurement signals according to claim 1, wherein: The performing thickness analysis based on the first processed signal includes: Converting the first processed signal to the frequency domain for peak search, using the found peak frequency as a calibration point, using frequency points with a preset frequency step before and after the calibration point on the frequency axis as boundary points, and determining the frequency range between the two boundary points as a selected frequency range, wherein the selected frequency range is used to determine the frequency band containing the thickness information; Converting the sample thickness measurement signal to the frequency domain, removing interference signals from the sample thickness measurement signal based on the frequency selection range, and then converting the sample thickness measurement signal after removing the interference signals to the time domain to obtain a correction signal; Thickness analysis is performed based on the corrected signal.

5. The method for processing thickness measurement signals according to claim 4, characterized in that: Eliminating the interference signal in the sample thickness measurement signal based on the frequency selection range, and then converting the sample thickness measurement signal after the interference signal is eliminated to the time domain to obtain a correction signal, including: Based on the frequency selection range, the measured signal of the sample thickness is divided into a signal within the frequency selection range and a signal outside the frequency selection range, the signal outside the frequency selection range is retained unchanged, and the signal within the frequency selection range is assigned zero to obtain a reconstructed signal; Converting the reconstructed signal to the time domain to obtain a second processed signal; The sample thickness measured signal is subtracted from the second processed signal to obtain the corrected signal.

6. The method for processing thickness measurement signals according to claim 4, characterized in that: Eliminating the interference signal in the sample thickness measurement signal based on the frequency selection range, and then converting the sample thickness measurement signal after the interference signal is eliminated to the time domain to obtain a correction signal, including: retaining the signal of the sample thickness measurement signal within the selected frequency range, filtering out the signal outside the selected frequency range, and obtaining a third processed signal; The third processed signal is converted to the time domain to obtain the corrected signal.

7. The method for processing thickness measurement signals according to claim 4, wherein: The performing thickness analysis based on the correction signal includes: Converting the corrected signal to the frequency domain for peak fitting to obtain fitting data; Target peak data is extracted from the fitting data, and thickness analysis is performed based on the target peak data.

8. The method for processing thickness measurement signals according to claim 7, wherein: The step of converting the corrected signal to the frequency domain for peak fitting to obtain fitting data includes: The corrected signal is zero-padded, and the corrected signal after zero-padded is transferred to the frequency domain for peak fitting to obtain the fitting data.

9. The method for processing thickness measurement signals according to any one of claims 1 to 8, characterized in that: The method of modulating the sample thickness measurement signal using a preset modulation function to obtain a first processed signal includes: Preprocessing the measured signal of the sample thickness, wherein the preprocessing includes at least one of noise reduction, detrending, and uniform sampling and reorganization; Extracting a data segment within a preset interval from the pre-processed sample thickness measurement signal to obtain a filtered signal; Performing empirical mode decomposition on the filtered signal, and modulating the decomposed intrinsic mode function using the modulation function to obtain the first processed signal, wherein the intrinsic mode function is the first decomposed intrinsic mode function component or a combination of multiple decomposed intrinsic mode function components.

10. The method for processing thickness measurement signals according to any one of claims 1 to 8, characterized in that: The method of obtaining a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured comprises: Obtain the sample thickness measurement signal generated by infrared light interferometry thickness measurement of the sample to be measured.

11. A device for processing thickness measurement signals, characterized in that: include: A thickness measurement signal acquisition module is used to obtain a sample thickness measurement signal generated by optical interferometry thickness measurement of the sample to be measured; a first processed signal generating module, configured to modulate the sample thickness measurement signal using a preset modulation function to obtain a first processed signal, wherein the modulation function is generated based on overlap parameters of an overlapping frequency band in which a predetermined matching degree between an ideal thickness measurement signal of a thickness standard sample and an actual thickness measurement signal of the thickness standard sample in the frequency domain reaches a preset standard, the overlap parameters including at least one of a center position, an overlap width, an amplitude, and a matching degree corresponding to the overlapping frequency band; A thickness analysis module is configured to perform thickness analysis based on the first processed signal.

12. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method for processing thickness measurement signals according to any one of claims 1 to 10.

13. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for processing a thickness measurement signal according to any one of claims 1 to 10 is implemented.

Images