Method and system for reading data from a multilayer optical storage with dynamic interlayer crosstalk suppression

Abstract

The present application relates to the field of data storage, in particular to a kind of dynamic interlayer crosstalk suppression multilayer optical storage data reading method and system.The present application includes the following steps: through low-coherence interferometer, the layer analysis interference signal of target data layer in multilayer optical storage medium is collected, and the spatial frequency characteristic of adjacent data layer is synchronously obtained, the spatial frequency characteristic includes power spectral density, frequency bandwidth and phase distribution parameter.The present application is first realized to the dynamic analysis of transient coherent crosstalk intensity through the frequency domain convolution quantization of coherent coupling coefficient and the generation of depth network weight.The real-time aperture control (0.45-0.85 NA continuously changes) of variable aperture diaphragm is used to match the priority of frequency band suppression, break through the band-limited constraint of traditional fixed filter;combined with multichannel interference reference light path and second-order differential compensation algorithm, the interlayer spacing error is compressed to submicron level, and the phase drift caused by medium refractive index gradient is effectively suppressed.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data storage, in particular to a multi-layer optical storage data reading method and system for dynamic interlayer crosstalk suppression. BACKGROUND

[0002] With the development of multi-layer optical storage technology to the level of thousands of layers, the suppression of interlayer crosstalk faces a fundamental challenge: the optical crosstalk between adjacent data layers is no longer limited to static coherent coupling, but evolves into a composite interference source affected by the dynamic fluctuations of the refractive index of the medium, thermal mechanical deformation and transient errors of the interlayer spacing. The traditional static compensation method relies on a preset model and cannot adapt to the amplification effect of wavefront phase distortion caused by the non-uniformity of the medium in the actual reading and writing process. When the spatial frequency characteristics of the adjacent layers and the target layer form a nonlinear coherent superposition, the frequency band energy leakage will significantly worsen the signal fidelity, especially at the edge of the medium and during high-speed scanning.

[0003] In addition, the suppression of crosstalk by phase mask or fixed aperture filtering still lacks a quantitative mechanism for transient coherent coupling strength, resulting in band suppression failure, residual crosstalk forming asymmetric trailing noise in the time domain, and ultimately limiting the decoupling accuracy of the data layer. SUMMARY

[0004] The present disclosure proposes a multi-layer optical storage data reading method and system for dynamic interlayer crosstalk suppression, aiming to overcome at least one of the deficiencies in the prior art.

[0005] To achieve the above-mentioned purpose, the technical solutions disclosed by the present application are as follows:

[0006] According to one aspect of the present disclosure, a multi-layer optical storage data reading method for dynamic interlayer crosstalk suppression is provided, comprising the following steps:

[0007] Collecting the tomographic interference signal of the target data layer in the multi-layer optical storage medium by a low-coherence interferometer, and synchronously acquiring the spatial frequency characteristics of the adjacent data layer, the spatial frequency characteristics including power spectral density, frequency bandwidth and phase distribution parameters;

[0008] Calculating the coherent coupling coefficient of the adjacent layer and the target layer based on the phase distribution parameters, generating an initial crosstalk weight matrix through frequency domain convolution operation, and inputting the matrix into a deep confidence network to generate crosstalk suppression parameters including frequency band suppression priority and phase compensation amount;

[0009] Generating a control signal for a variable aperture diaphragm according to the frequency band suppression priority, adjusting the numerical aperture of the objective lens in real time through a piezoelectric ceramic driver, so that it continuously changes in the range of 0.45 to 0.85, and synchronously adjusting the incident angle of the scanning light beam to match the light beam propagation path after the change of the numerical aperture;

[0010] The interlayer spacing error caused by the non-uniformity of the medium refractive index is compensated at a sub-micron level through the cooperation of a multi-channel interference reference light path and a dynamic optical path compensator.

[0011] The aliasing signal is subjected to multi-scale decomposition in the signal processing module, an adaptive threshold filter is applied to the high-frequency component according to the frequency band suppression priority, the low-frequency component is subjected to phase reconstruction by using an asymmetric Gaussian function to eliminate residual crosstalk, and finally the corrected target layer data is output through inverse wavelet transform.

[0012] Further, the dynamic optical path compensator captures the transient change of the interlayer spacing through a second-order differential term, and calculates the optical path compensation amount based on the thermal-optic coefficient of the medium and the mechanical deformation weight factor, and the steps include:

[0013] The temperature change and wavelength drift of each partition on the surface of the medium are monitored in real time by a distributed temperature sensor.

[0014] According to the thermal-optic coefficient of the medium and the mechanical deformation weight factor, the influence of the refractive index change caused by the temperature gradient on the optical path is calculated.

[0015] The transient change trend of the interlayer spacing error is captured through a second-order differential operation to generate a dynamic compensation signal.

[0016] The piezoelectric ceramic displacement stage is driven to adjust the optical path length of the reference light path with sub-micron accuracy, and the deviation of the focal position of the target layer from the theoretical value is controlled within ± 50 nm.

[0017] Further, the coherent coupling coefficient quantifies the strength and spatial correlation of the interlayer crosstalk by combining frequency domain integration and an exponential decay function, and the calculation method of the coherent coupling coefficient is:

[0018] where Γ i,j represents the coherent coupling coefficient between the i-th layer and the j-th layer, S i (λ) is the spectral signal of the i-th layer, S j (λ) is the spectral signal of the j-th layer, S j ∗ (λ) is the phase conjugate spectral signal of the j-th layer, Δz i,j is the interlayer spacing error, σ z is the standard deviation of the refractive index fluctuation of the medium, k is the spatial sampling position number, and N is the total number of sampling points.

[0019] Further, the phase reconstruction process of the asymmetric Gaussian function includes:

[0020] According to the time domain decay characteristic in the crosstalk suppression parameter, the widening coefficients of the positive and negative time axes are defined respectively.

[0021] The phase distribution of the low-frequency component is asymmetrically weighted to suppress the tailing effect caused by reflection of adjacent layers;

[0022] The spread factor is adjusted by an iterative optimization algorithm, so that the mean square error of the reconstructed waveform is less than a preset threshold.

[0023] Further, the data reading method further comprises a pre-calibration step:

[0024] A reference mark array is written on the surface of the medium, and layer spacing calibration data is generated by scanning with a low-power probe beam;

[0025] A three-dimensional distribution map of the refractive index of the medium is constructed according to the calibration data, and the three-dimensional sinc function interpolation algorithm is used to realize high-precision modeling of the spatial refractive index deviation, and the three-dimensional distribution map is pre-stored in the signal processing module;

[0026] The three-dimensional sinc function interpolation algorithm comprises:

[0027] The coordinates of the reference marks and the corresponding refractive index deviation coefficients are extracted;

[0028] A three-dimensional sinc function kernel is generated with each mark as the center, and the weight of the kernel is determined by the mark spacing and the thermal expansion coefficient of the medium;

[0029] A continuous refractive index distribution model is generated by linearly superimposing all kernel functions, and the model is discretized into grid data for real-time calling.

[0030] Further, the data reading method further comprises a post-validation step:

[0031] The corrected target layer data is calculated by a hash digest, and compared with a pre-stored hash value;

[0032] If the digest values do not match, a read operation of the redundant data layer is triggered, and the abnormal data is repaired by parity check bits and error correction codes;

[0033] The repaired data is re-written to the target layer and the hash value is updated.

[0034] Further, the training process of the deep belief network comprises:

[0035] Historical crosstalk signals under different medium types and layer spacing error conditions are collected as a training set;

[0036] The network weight is optimized by the contrast divergence algorithm, so that the mean square error of the network output crosstalk suppression parameter is reduced to less than 0.1%;

[0037] The noise diversity of the training data is enhanced by using the generative adversarial network, which is used to improve the generalization ability of the model to unknown media.

[0038] Further, the control signal generation logic of the variable aperture stop comprises:

[0039] According to the frequency band suppression priority, a target numerical aperture theoretical value is determined;

[0040] The actual numerical aperture of the objective lens is monitored in real time through a closed-loop feedback circuit, and compared with the theoretical value;

[0041] If the deviation exceeds 5%, the piezoelectric ceramic driving voltage is adjusted until the actual aperture converges within the theoretical range;

[0042] The adjustment step of the incident angle of the scanning light beam comprises:

[0043] According to the change amount of the numerical aperture of the objective lens, the refraction angle of the light beam inside the medium is calculated;

[0044] A wavefront phase distribution matched with the refraction angle is generated through a liquid crystal phase modulator;

[0045] A diffractive optical element is used to pre-distort the light beam to compensate for the focal point position to ensure axial stability.

[0046] According to another aspect of the present disclosure, a multi-layer optical storage data reading system for dynamic interlayer crosstalk suppression is provided for implementing the dynamic interlayer crosstalk suppression multi-layer optical storage data reading method as described above, and the data reading system comprises:

[0047] A tomographic signal acquisition module comprising a low-coherence interferometer and a high-speed photodetector is used to acquire interference signals of the target layer and adjacent layers;

[0048] A dynamic crosstalk prediction module integrates a deep confidence network and a frequency domain convolution unit to generate crosstalk suppression parameters according to phase distribution parameters;

[0049] An optical parameter control module composed of a variable aperture objective lens, a liquid crystal phase modulator, and a piezoelectric ceramic driver is used to dynamically adjust the numerical aperture and the wavefront phase;

[0050] An optical path compensation module including a multi-channel interference reference light path, a piezoelectric ceramic displacement stage, and a temperature control subsystem is used to compensate for sub-micron interlayer spacing errors;

[0051] A signal processing module embedded with an asymmetric wavelet transform algorithm and a hash verification unit is used for signal separation, correction, and data integrity verification.

[0052] Further, the optical path compensation module further comprises:

[0053] The dynamic optical path compensator, and the calculation method of the compensation amount is:

[0054] wherein, AL is the optical path compensation amount, a m and β are the medium thermal-optical coefficient and the mechanical deformation weight factor, ΔT m is the temperature change of the mth temperature control sub-zone, M is the total number of temperature control sub-zones, Δλ m is the wavelength drift amount, Δz is the axial deformation amount, n is the medium refractive index, and T is the absolute temperature of the medium.

[0055] a wavefront phase modulator configured to generate a conjugate phase field according to the crosstalk suppression parameter, for canceling the wavefront distortion of the interlayer crosstalk.

[0056] a temperature control subsystem for maintaining the surface temperature fluctuation of the medium within ±0.1°C.

[0057] The beneficial effects of the present application are:

[0058] The present application first realizes dynamic analysis of transient coherent crosstalk intensity through frequency domain convolution quantization of coherent coupling coefficient and deep network weight generation. The real-time aperture control (0.45-0.85 NA continuous change) of the variable aperture diaphragm matches the priority of the frequency band suppression, breaking through the band limitation of the traditional fixed filter; combined with the multi-channel interference reference light path and the second-order differential compensation algorithm, the interlayer spacing error is compressed to sub-micron level, effectively suppressing the phase drift caused by the refractive index gradient of the medium.

[0059] At the signal reconstruction level, the multi-scale decomposition technology based on asymmetric Gaussian function compensates the phase for the tailing characteristics of residual crosstalk, so that the mean square error of the reconstructed waveform is reduced to below the threshold. Finally, the four-order optimization chain of "coherent coupling dynamic modeling-numerical aperture frequency domain matching-optical path closed loop compensation-asymmetric noise reconstruction" is formed, which reduces the original bit error rate by an order of magnitude in a 50 nm thick thousand-layer medium, and the adaptability to medium thermal noise is improved.

[0060] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, and to implement the content of the specification, the following describes the preferred embodiments of the present application in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0061] Figure 1 is the flow chart of the multi-layer optical storage data reading method of the dynamic interlayer crosstalk suppression of the present application;

[0062] Figure 2 is the schematic diagram of the interlayer crosstalk coherent coupling principle of the present application;

[0063] Figure 3 is the schematic diagram of the dynamic optical path compensation mechanism of the present application;

[0064] Figure 4A schematic diagram of the numerical aperture dynamic regulation effect of the present application is shown in the figure;

[0065] Figure 5 A schematic diagram of the asymmetric Gaussian phase reconstruction process of the present application is shown in the figure;

[0066] Figure 6 A schematic diagram of the three-dimensional refractive index spatial distribution of the present application is shown in the figure;

[0067] Figure 7 A schematic diagram of the deep belief network training process of the present application is shown in the figure. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0069] In the embodiments of the present application, the words such as "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design presented as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of the words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0070] The present application provides the following preferred embodiments:

[0071] Embodiment I

[0072] In order to solve the problem of dynamic interlayer crosstalk in multi-layer optical storage media, the present embodiment provides a multi-layer optical storage data reading method for suppressing dynamic interlayer crosstalk. The method effectively suppresses interlayer crosstalk and improves the accuracy and reliability of data reading by precisely controlling and adjusting optical parameters and using advanced signal processing technology. As shown in the figure, Figure 1 The flow of the multi-layer optical storage data reading method is shown in the figure.

[0073] S100: Collect the tomographic interference signals of the target data layer in the multi-layer optical storage medium by a low-coherence interferometer, and synchronously acquire the spatial frequency characteristics of the adjacent data layer, including power spectral density, frequency bandwidth, and phase distribution parameters.

[0074] S200: Calculate the coherent coupling coefficient of the adjacent layer and the target layer based on the phase distribution parameters, generate an initial crosstalk weight matrix through frequency domain convolution operation, and input the matrix into a deep belief network to generate crosstalk suppression parameters containing frequency band suppression priority and phase compensation amount.

[0075] S300: generating a control signal of the variable aperture diaphragm according to the band suppression priority, and adjusting the numerical aperture of the objective lens in real time through a piezoelectric ceramic driver to continuously change in the range of 0.45 to 0.85, and synchronously adjusting the incident angle of the scanning light beam to match the light beam propagation path after the numerical aperture is changed.

[0076] S400: compensating for the interlayer spacing error caused by the non-uniformity of the medium refractive index at a sub-micron level through the cooperation of the multi-channel interference reference light path and the dynamic optical path compensator.

[0077] S500: performing multi-scale decomposition on the aliasing signal in the signal processing module, applying adaptive threshold filtering to the high-frequency component according to the band suppression priority, and using an asymmetric Gaussian function to reconstruct the phase of the low-frequency component to eliminate residual crosstalk, and finally outputting the corrected target layer data through inverse wavelet transform.

[0078] First, the tomographic interference signal of the target data layer in the multi-layer optical storage medium is collected by a low coherence interferometer, and the spatial frequency characteristics of the adjacent data layers are synchronously obtained. As shown in Figure 2 As shown in FIG. 1, the low coherence interferometer synchronously detects the reflection spectrum of the adjacent data layer when scanning the target data layer. Specifically, the low coherence interferometer (for example, Thorlabs OS-5305) works at a center wavelength of 635 nm, and the power spectral density, frequency bandwidth and phase distribution parameters of the adjacent layer are analyzed through a double-channel spectral decomposition module. It should be understood that the extraction of these parameters depends on the Fourier domain integration operation after grating splitting, so that the coherent coupling relationship between the adjacent layer and the target layer can be accurately described.

[0079] Further, based on the obtained phase distribution parameters, the coherent coupling coefficient of the adjacent layer and the target layer is calculated. An initial crosstalk weight matrix is generated through a frequency domain convolution operation, and the matrix is input into a deep belief network (DBN) to generate crosstalk suppression parameters including band suppression priority and phase compensation amount. It should be understood that this method based on frequency domain convolution can more accurately quantify the strength and spatial correlation of interlayer crosstalk.

[0080] Further, according to the generated band suppression priority, a control signal of the variable aperture diaphragm is generated. The numerical aperture (NA) of the objective lens is adjusted in real time through a piezoelectric ceramic driver to continuously change in the range of 0.45 to 0.85. As shown in Figure 4As shown, the point spread function (PSF) under different numerical apertures will change, thereby affecting the accuracy of the focal position. At the same time, the incident angle of the scanning beam is adjusted synchronously to match the beam propagation path after the numerical aperture changes, ensuring that the beam propagation path inside the medium remains stable. It can be understood that this dynamic regulation mechanism can effectively reduce the crosstalk of high-frequency components and improve the resolution and accuracy of data reading.

[0081] Further, through the synergistic effect of the multi-channel interference reference light path and the dynamic optical path compensator, the interlayer spacing error caused by the non-uniformity of the medium refractive index is compensated at the sub-micron level. As shown, Figure 3 The dynamic optical path compensator captures the transient changes in interlayer spacing through the second-order derivative term and calculates the optical path compensation amount based on the medium thermal-optic coefficient and the mechanical deformation weight factor.

[0082] Further, in the signal processing module, the aliasing signal is subjected to multi-scale decomposition. According to the frequency band suppression priority, an adaptive threshold filter is applied to the high-frequency components, and an asymmetric Gaussian function is used to reconstruct the phase of the low-frequency components to eliminate residual crosstalk. As shown, Figure 5 The phase reconstruction process of the asymmetric Gaussian function includes defining a widening coefficient of the positive and negative time axes, asymmetrically weighting the phase distribution of the low-frequency components to suppress the tailing effect caused by reflection from adjacent layers. The widening coefficient is adjusted through an iterative optimization algorithm so that the mean square error of the reconstructed waveform is less than a preset threshold. Finally, the corrected target layer data is output through inverse wavelet transform.

[0083] The benefits of this embodiment are that through accurate coherent coupling coefficient calculation, dynamic numerical aperture regulation, sub-micron level optical path compensation, and efficient signal processing technology, the dynamic interlayer crosstalk in multi-layer optical storage media is effectively suppressed, and the accuracy and reliability of data reading are improved.

[0084] Embodiment Two

[0085] To solve the problem of interlayer spacing error caused by temperature gradient and mechanical deformation in multi-layer optical storage media, this embodiment further optimizes the working mechanism of the dynamic optical path compensator. This mechanism captures the transient changes in interlayer spacing through the second-order derivative term and calculates the optical path compensation amount based on the medium thermal-optic coefficient and the mechanical deformation weight factor, thereby achieving sub-micron level optical path compensation.

[0086] First, the temperature change and wavelength drift of each partition on the medium surface are monitored in real time by a distributed temperature sensor. Specifically, the distributed temperature sensor (e.g., fiber Bragg grating sensor) is arranged along the medium surface, which can accurately measure the temperature distribution of different areas. At the same time, the wavelength drift is detected by a high-resolution spectrometer to ensure accurate monitoring of the refractive index change of the medium. It should be understood that the high sensitivity and fast response characteristics of these sensors are the basis for realizing dynamic optical path compensation.

[0087] Further, according to the thermo-optic coefficient of the medium and the mechanical deformation weight factor, the influence of the refractive index change caused by the temperature gradient on the optical path is calculated. By second-order differential operation, the transient change trend of the layer spacing error is captured, and a dynamic compensation signal is generated. By second-order differential operation, the transient change trend of the layer spacing error can be effectively captured, thereby generating a more accurate dynamic compensation signal. It should be understood that the second-order differential operation can better reflect the rapid change of the layer spacing error, improving the real-time and accuracy of the compensation.

[0088] Further, the piezoelectric ceramic displacement stage is driven to adjust the optical path length of the reference light path with sub-micron accuracy, so that the focal position of the target layer is controlled within ±50 nm from the theoretical value. By high-precision control of the piezoelectric ceramic displacement stage (e.g., Thorlabs PZT-500), sub-micron optical path adjustment can be achieved to ensure stable and accurate focal position of the target layer. It can be understood that this high-precision optical path adjustment technology can effectively eliminate the layer spacing error caused by temperature gradient and mechanical deformation, and improve the reliability and accuracy of data reading.

[0089] In order to further improve the stability of the system, a temperature control system can be integrated into the system. The system maintains the temperature fluctuation of the medium surface within ±0.1°C through a closed-loop feedback mechanism. Specifically, the temperature control system includes multiple temperature control partitions, each equipped with independent heating and cooling elements, and controlled accurately by a PID controller. This can further reduce the influence of temperature gradient on the refractive index and improve the overall performance of the system.

[0090] In addition, a real-time data processing and feedback mechanism can be introduced, which processes the collected temperature and wavelength data in real time through a high-speed data processor (e.g., FPGA or DSP), and feeds back the processing results to the piezoelectric ceramic displacement stage. This real-time feedback mechanism can further improve the response speed and accuracy of the system, ensuring the real-time and accuracy of the optical path compensation.

[0091] The embodiment has the benefits that, through accurate temperature monitoring, refractive index change calculation, second-order differential operation and high-precision optical path adjustment, the layer spacing error in the multi-layer optical storage medium is effectively suppressed, and the accuracy and reliability of data reading are improved. Through the method of the embodiment, sub-micron level optical path compensation can be realized, the focal position of the target layer is stable and accurate, and the performance of the multi-layer optical storage system is improved.

[0092] Embodiment three

[0093] In order to solve the problem of interlayer crosstalk in the multi-layer optical storage medium, the embodiment further optimizes the calculation method of the coherent coupling coefficient. Through the combination of frequency domain integration and exponential decay function, the strength and spatial correlation of interlayer crosstalk are quantified.

[0094] As shown in Figure 2 , the coherent coupling coefficient Γ i,j in the embodiment is used to quantify the strength and spatial correlation of crosstalk between the i-th layer and the j-th layer. The specific calculation formula is as follows:

[0095] , wherein Γ i,j represents the coherent coupling coefficient of the i-th layer and the j-th layer, S i (λ) is the spectral signal of the i-th layer, S j (λ) is the spectral signal of the j-th layer, S j ∗ (λ) is the phase conjugate spectral signal of the j-th layer, Δz i,j is the layer spacing error, σ z is the standard deviation of the refractive index fluctuation of the medium, k is the spatial sampling position number, and N is the total number of sampling points.

[0096] Further, the frequency domain integration is used to calculate the coherent integration between the spectral signals. Specifically, the coherent integration part is used to describe the degree of overlap of the two-layer spectral signals in the frequency domain. It should be understood that the larger the integral value, the stronger the coherence between the two-layer signals, resulting in greater crosstalk.

[0097] Further, the coherent integration is normalized, i.e. , the normalization factor ensures that the value of the coherent coupling coefficient is in the range of 0 to 1, which is convenient for subsequent analysis and comparison.

[0098] Further, the exponential decay function is used to describe the influence of the layer spacing error on the crosstalk. It can be understood that as the layer spacing error Δz i,j increases, the value of the exponential decay function rapidly decreases, thereby reducing the coherent coupling coefficient Γ i,j , and effectively suppressing the interlayer crosstalk.

[0099] To further improve the stability and accuracy of the system, a real-time monitoring and feedback mechanism can be integrated into the system. By using high-precision sensors (such as fiber Bragg grating sensors), the temperature and stress distribution of each layer can be monitored in real time, and the data can be fed back to the control system. In this way, the coherent coupling coefficient can be dynamically adjusted to ensure stable operation of the system.

[0100] Embodiment Four

[0101] To solve the tailing effect caused by the reflection of adjacent layers in the multi-layer optical storage medium, the phase reconstruction process of the asymmetric Gaussian function is further optimized in this embodiment. By adjusting the spreading coefficients of the positive and negative time axes and combining the iterative optimization algorithm, the asymmetric weighting of the phase distribution of low-frequency components is realized, thereby effectively suppressing the tailing effect.

[0102] As Figure 5 shown, the phase reconstruction process of the asymmetric Gaussian function in this embodiment includes defining the spreading coefficients of the positive and negative time axes according to the time-domain attenuation characteristics in the crosstalk suppression parameter. Specifically, for the negative time axis, the spreading coefficient a is used; for the positive time axis, the spreading coefficient β is used. It should be understood that these two coefficients determine the spreading degree of the asymmetric Gaussian function on different time axes, thereby affecting the effect of phase reconstruction. By adjusting the values of a and β, accurate control of low-frequency components can be achieved, thereby suppressing the tailing effect caused by the reflection of adjacent layers.

[0103] Further, the phase distribution of low-frequency components is asymmetrically weighted. Specifically, by multiplying the original phase signal with the asymmetric Gaussian function, the weighting processing of low-frequency components is realized. The formula is as follows:

[0104] wherein, ϕ(t) represents the asymmetric Gaussian function; ϕ original (t) represents the original phase signal. It can be understood that this asymmetric weighting processing can effectively suppress the tailing effect caused by the reflection of adjacent layers, and improve the clarity and stability of the signal. In this way, it can be ensured that the phase distribution of low-frequency components is reasonably adjusted in the reconstruction process, thereby reducing the tailing phenomenon.

[0105] Next, the spreading coefficients a and β are adjusted by the iterative optimization algorithm, so that the mean square error of the reconstructed waveform is less than the preset threshold. The specific iterative optimization process is as follows:

[0106] wherein, ϕ target(t) represents the target phase signal; T is the total number of time points. It needs to be understood that by constantly adjusting the values of a and β, the mean square error between the reconstructed waveform and the target waveform is minimized, so as to achieve the optimal phase reconstruction effect. This iterative optimization method can ensure that the best spreading coefficient combination is found in different application scenarios, thereby improving the adaptability of the system.

[0107] In addition, an adaptive filter can be introduced to preprocess the spectral signal to reduce noise and interference. The adaptive filter can automatically adjust the filtering parameters according to the real-time changes of the signal, thereby further improving the signal-to-noise ratio and stability of the signal. Through the adaptive filter, the noise component in the signal can be effectively removed, and the purity of the signal can be improved, thereby improving the overall performance of the system.

[0108] In order to further improve the accuracy of phase reconstruction, a multi-stage correction mechanism can be introduced. After each iteration optimization, the reconstructed waveform is corrected in multiple stages to gradually eliminate residual errors. Specifically, by multiple iterations and multi-stage correction, the target waveform can be gradually approximated, thereby improving the accuracy and stability of the reconstructed waveform. Through multi-stage correction, the phase reconstruction process can be further refined to ensure the accuracy and reliability of the final result.

[0109] Through the embodiment, the phase reconstruction method using asymmetric Gaussian function effectively suppresses the tailing effect caused by adjacent layer reflection in multi-layer optical storage medium, and improves the clarity and stability of the signal.

[0110] Embodiment Five

[0111] In order to solve the problem of reading error caused by inaccurate layer spacing calibration in multi-layer optical storage medium, the pre-calibration step is further optimized in this embodiment. By writing a reference mark array on the surface of the medium, and using a low-power probe beam to scan to generate layer spacing calibration data, as shown in Figure 6 a three-dimensional distribution map of the refractive index of the medium is constructed, so as to realize high-precision spatial refractive index deviation modeling.

[0112] First, a reference mark array is written on the surface of the medium. These reference marks can be tiny pits or protrusions, whose position and shape are precisely designed to ensure that they can provide sufficient reference information. Then, a low-power probe beam is used to scan these reference marks to generate layer spacing calibration data. The low-power probe beam is selected to avoid thermal damage to the medium, while ensuring that high-precision reflection signals can be obtained.

[0113] According to the generated calibration data, a three-dimensional distribution map of the refractive index of the medium is constructed. Specifically, a three-dimensional sinc function interpolation algorithm is used to realize high-precision modeling of spatial refractive index deviation. The algorithm includes the following steps:

[0114] 1、Extracting coordinates and refractive index deviation coefficients, extracting the coordinates of each reference mark and its corresponding refractive index deviation coefficient from the calibration data. These coordinates and refractive index deviation coefficients will be used as the basis data for subsequent interpolation.

[0115] 2、Generating a three-dimensional sinc function kernel, generating a three-dimensional sinc function kernel centered on each reference mark. The form of the three-dimensional sinc function kernel is as follows:

[0116] where (x0, y0, z0) is the coordinate of the reference mark, and sinc(t) = sin(t) / t is the sinc function. It should be understood that the sinc function has good localization characteristics and can perform smooth interpolation of refractive index deviation in space.

[0117] 3、Weight calculation, the weight of the three-dimensional sinc function kernel is determined by the mark spacing and the thermal expansion coefficient of the medium. Specifically, the weight calculation formula is as follows:

[0118] where α is the proportionality coefficient, d is the mark spacing, d0 is the reference spacing, and σ is the standard deviation. It should be understood that by introducing the thermal expansion coefficient, the deformation of the medium at different temperatures can be considered, thereby improving the accuracy of the model.

[0119] By linearly superimposing all three-dimensional sinc function kernels, a continuous refractive index distribution model is generated, and the generated continuous refractive index distribution model is discretized into grid data for real-time calling. The discretization process can be realized by uniform sampling or adaptive sampling method to ensure sufficient resolution in key areas.

[0120] To further improve the accuracy of the refractive index distribution model, a multi-level correction mechanism can be introduced. After each iteration optimization, the refractive index distribution model is corrected at multiple levels to gradually eliminate residual errors. Specifically, by multiple iterations and multi-level correction, the actual refractive index distribution can be gradually approached, thereby improving the accuracy and stability of the model. Through multi-level correction, the refractive index distribution model can be further refined to ensure the accuracy and reliability of the final result.

[0121] Through this embodiment, a three-dimensional distribution map of the refractive index of the medium is constructed using the three-dimensional sinc function interpolation algorithm, realizing high-precision spatial refractive index deviation modeling. The data reading accuracy of the multi-layer optical storage medium is improved.

[0122] Embodiment six

[0123] To address the potential correction error issues that may arise during data reading in multi-layer optical storage media, the post-validation step is further optimized in this embodiment. By performing a hash digest calculation after data reading and correction is complete and comparing it with the pre-stored hash value, the consistency and integrity of the data can be effectively detected.

[0124] First, a hash digest calculation is performed on the corrected target layer data. Hash digest is a method of converting arbitrary length data into a fixed length hash value, commonly used for data integrity verification. Commonly used hash algorithms include SHA-256, MD5, etc. It should be understood that hash algorithms have one-way nature, i.e. from the hash value, the original data cannot be deduced, thus ensuring data security and non-tamperability.

[0125] Further, the calculated hash digest value is compared with the pre-stored hash value. The pre-stored hash value is generated and stored when the data is written, for subsequent verification. If the digest values do not match, it indicates that errors or tampering have occurred during reading or correction of the data, at which point the reading operation of the redundant data layer is triggered. The redundant data layer usually contains parity bits and error correction codes for detecting and repairing errors in the data.

[0126] Further, the abnormal data is repaired through parity bits and error correction codes. Parity bits can detect single-bit errors in the data, while error correction codes (such as Hamming codes, RS codes) can correct a certain number of error bits. It should be understood that the design of error correction codes should take into account the characteristics of the medium and possible error patterns to ensure that the data can be effectively recovered in actual application.

[0127] Further, the repaired data is re-written to the target layer and the hash value is updated. The re-writing process needs to ensure the accuracy and consistency of the data, avoiding the introduction of new errors. At the same time, updating the hash value is to maintain the latest state of the data for subsequent verification.

[0128] To further improve the fault tolerance of the system, a dynamic redundancy strategy can be introduced. According to the usage and environmental conditions of the medium, the number and distribution of redundant data layers are dynamically adjusted. For example, in a high-noise environment, the number of redundant data layers is increased to improve data reliability; in a low-noise environment, the number of redundant data layers is reduced to save storage space. It should be understood that the dynamic redundancy strategy can be flexibly adjusted according to actual needs, thus achieving optimal utilization of resources.

[0129] In addition, a multi-level verification mechanism can be introduced to improve the reliability of data through multi-level verification. For example, parity check is performed at the data block level, cyclic redundancy check (CRC) is performed at the file level, and hash check is performed at the system level. Through multi-level verification, errors in the data can be detected and corrected at different levels, thereby improving the overall data reliability.

[0130] To further improve the efficiency and security of hash digest calculation, an adaptive hash algorithm can be used. The adaptive hash algorithm can select appropriate hash functions according to the characteristics of the data and the application scenario, thereby improving the calculation efficiency while ensuring security. For example, for small data blocks, a fast hash function can be selected, and for large data blocks, a more secure but computationally intensive hash function can be selected. It should be understood that the adaptive hash algorithm can be flexibly selected according to actual needs, thereby achieving a balance between performance and security.

[0131] Through this embodiment, by performing hash digest calculation on the corrected target layer data and comparing it with the pre-stored hash value, the consistency and integrity of the data can be effectively detected. At the same time, by introducing dynamic redundancy strategy, multi-level verification mechanism and adaptive hash algorithm, the fault tolerance and data reliability of the system are further improved.

[0132] Embodiment Seven

[0133] To solve the problem of interlayer crosstalk suppression in multi-layer optical storage media, as shown in Figure 7 This embodiment further optimizes the training process of deep belief network (DBN). By collecting historical crosstalk signals under different media types and interlayer spacing error conditions as training set, and optimizing network weights through contrastive divergence algorithm, the mean square error of network output crosstalk suppression parameters is reduced to below 0.1%. In addition, the noise diversity of training data is enhanced using generative adversarial network (GAN), and the generalization ability of the model to unknown media is improved.

[0134] Specifically, historical crosstalk signals under different media types and interlayer spacing error conditions are collected as training set. These historical crosstalk signals cover a variety of media types and different interlayer spacing error conditions, which can provide rich training samples. It should be understood that diversified training samples help to improve the generalization ability of the model, so that it can effectively suppress interlayer crosstalk under different conditions.

[0135] Further, the network weights are optimized using the contrastive divergence algorithm. The contrastive divergence algorithm is an effective method for unsupervised learning, which optimizes the network weights by minimizing the reconstruction error. Specifically, the contrastive divergence algorithm calculates the difference between the input data and the reconstructed data, and adjusts the network weights through gradient descent, so that the reconstruction error gradually decreases. It should be understood that the contrastive divergence algorithm has good convergence, and is suitable for complex data distribution.

[0136] Further, the mean square error of the network output crosstalk suppression parameter is reduced to below 0.1%. Mean square error (MSE) is a commonly used indicator to measure the difference between predicted values and true values. By continuously optimizing the network weights, the MSE gradually decreases, and eventually reaches the predetermined target value. It should be understood that the smaller the MSE, the higher the prediction accuracy of the network, so as to more effectively suppress interlayer crosstalk.

[0137] Application of generative adversarial network

[0138] In order to further improve the generalization ability of the model, the noise diversity of the training data is enhanced using the generative adversarial network (GAN). GAN consists of a generator and a discriminator, the generator is responsible for generating new data samples, and the discriminator is responsible for distinguishing between generated data and real data. Through continuous adversarial training, the generator can generate more realistic data samples, thereby enriching the diversity of training data. It should be understood that GAN can generate noise samples with diversity, which helps to improve the generalization ability of the model to unknown media.

[0139] In order to further improve the adaptability of the model, a dynamic weight adjustment mechanism can be introduced. According to the change of the loss function in the training process, the weights of each layer of the network are dynamically adjusted. For example, in the early stage of training, the adjustment amplitude of the bottom layer weight can be increased to quickly converge; in the later stage of training, the adjustment amplitude of the high layer weight can be reduced for fine adjustment. It should be understood that the dynamic weight adjustment mechanism can be flexibly adjusted according to the actual needs of the training process, thereby improving the training efficiency and accuracy of the model.

[0140] In addition, a multi-task learning mechanism can be introduced to improve the generalization ability of the model by simultaneously optimizing multiple related tasks. For example, while suppressing interlayer crosstalk, other related performance indicators such as data read speed, power consumption, etc. can also be optimized. Through multi-task learning, the model can achieve optimal results in multiple aspects at the same time, thereby improving overall performance.

[0141] To further improve the generalization ability of the model, a transfer learning mechanism can be introduced. By transferring the existing pre-trained model to the new task, the knowledge of the existing model can be fully utilized to accelerate the learning process of the new task. For example, an existing image recognition model can be transferred to the interlayer crosstalk suppression task, and the network weights can be fine-tuned to adapt to the new task. It should be understood that transfer learning can significantly reduce the training time and data requirements of the new task, and improve the generalization ability of the model.

[0142] Through this embodiment, by optimizing the training process of the deep belief network and introducing mechanisms such as adversarial generative network, dynamic weight adjustment, multi-task learning and transfer learning, the generalization ability and adaptability of the model are further improved, thereby effectively suppressing the interlayer crosstalk in the multi-layer optical storage medium.

[0143] Embodiment Eight

[0144] To solve the problem of controlling the variable aperture diaphragm in the multi-layer optical storage medium, this embodiment further refines the control signal generation logic of the variable aperture diaphragm. By determining the theoretical value of the target numerical aperture according to the frequency band suppression priority, and monitoring the actual numerical aperture of the objective lens in real time through a closed-loop feedback circuit, the deviation is compared with the theoretical value. If the deviation exceeds 5%, adjust the piezoelectric ceramic driving voltage until the actual aperture converges within the theoretical range. In addition, the adjustment steps of the incident angle of the scanning beam are also refined, by calculating the refraction angle and generating a matching wavefront phase distribution, using a diffractive optical element for pre-distortion compensation to ensure the axial stability of the focal position.

[0145] First, the theoretical value of the target numerical aperture is determined according to the frequency band suppression priority. Numerical aperture (NA) is an important parameter to describe the light gathering ability of an optical system, the larger the value, the higher the resolution of the system. The frequency band suppression priority refers to the different requirements for the numerical aperture at different frequency bands. For example, at high frequencies, a higher numerical aperture is required to improve resolution; at low frequencies, the numerical aperture can be appropriately reduced to reduce noise. It should be understood that the frequency band suppression priority can be flexibly adjusted according to actual needs to achieve the best imaging effect.

[0146] Further, the actual numerical aperture of the objective lens is monitored in real time through a closed-loop feedback circuit, and compared with the theoretical value. The closed-loop feedback circuit is a technology that adjusts system parameters through real-time monitoring and feedback. Specifically, the closed-loop feedback circuit monitors the actual numerical aperture of the objective lens in real time through a sensor, and compares it with the theoretical value. If the deviation exceeds 5%, adjust the piezoelectric ceramic driving voltage to change the aperture size of the diaphragm until the actual aperture converges within the theoretical range. It should be understood that the closed-loop feedback circuit can realize high-precision real-time control to ensure the stability and accuracy of the system.

[0147] Further, the refraction angle of the light beam inside the medium is calculated according to the change of the numerical aperture of the objective lens. The refraction angle refers to the change of the direction of the light beam when it propagates inside the medium. Specifically, the refraction angle is calculated by Snell's law. It should be understood that by calculating the refraction angle, the propagation direction of the light beam inside the medium can be accurately controlled.

[0148] Further, a wavefront phase distribution matching the refraction angle is generated by a liquid crystal phase modulator. The liquid crystal phase modulator is a device that can dynamically adjust the phase distribution of the light beam, and by applying different voltages, the required wavefront phase distribution can be generated. Specifically, by controlling the voltage of the liquid crystal phase modulator, a wavefront phase distribution matching the refraction angle is generated, thereby achieving accurate control of the light beam. It should be understood that the liquid crystal phase modulator has the characteristics of high resolution and fast response, and is suitable for dynamic light beam control.

[0149] Further, a diffractive optical element is used to compensate for the pre-distortion of the light beam to ensure the axial stability of the focal point position. The diffractive optical element is a device that controls the propagation of the light beam through the diffraction effect, and by designing a specific diffraction pattern, the pre-distortion compensation of the light beam can be achieved. Specifically, by designing the diffraction pattern of the diffractive optical element, the pre-distortion compensation of the light beam is achieved, thereby eliminating the distortion of the light beam when it propagates inside the medium, ensuring the axial stability of the focal point position. It should be understood that the diffractive optical element has the characteristics of high precision and flexibility, and is suitable for complex light beam control.

[0150] To further improve the adaptability of the system, a dynamic aperture adjustment mechanism can be introduced. According to the characteristics of the medium and the environmental conditions, the aperture size of the diaphragm is dynamically adjusted. For example, in a high-noise environment, the aperture size can be appropriately reduced to reduce noise; in a low-noise environment, the aperture size can be increased to improve the resolution. It should be understood that the dynamic aperture adjustment mechanism can be flexibly adjusted according to actual needs, thereby achieving the best imaging effect.

[0151] In addition, an adaptive phase modulation mechanism can be introduced, which monitors the phase distribution of the light beam in real time and dynamically adjusts the voltage of the liquid crystal phase modulator according to the monitoring results to achieve accurate control of the light beam. Specifically, the phase distribution of the light beam is monitored in real time by a sensor, and the monitoring results are fed back to the control system, and the voltage of the liquid crystal phase modulator is adjusted by an adaptive algorithm, thereby achieving dynamic adjustment of the light beam. It should be understood that the adaptive phase modulation mechanism can achieve high-precision real-time control, improving the stability and accuracy of the system.

[0152] To further improve the stability of the system, a multi-level feedback control mechanism can be introduced. By setting feedback control loops at different levels, multi-level control of the system is achieved. For example, a first level feedback control loop is set at the aperture control level, and a second level feedback control loop is set at the phase modulation level. Through multi-level feedback control, the error of the system can be gradually eliminated, and the stability and accuracy of the system can be improved. It should be understood that the multi-level feedback control mechanism can achieve multi-level control and improve the overall performance of the system.

[0153] Through this embodiment, by refining the control signal generation logic of the variable aperture diaphragm and the adjustment steps of the incident angle of the scanning light beam, and introducing dynamic aperture adjustment, adaptive phase modulation, and multi-level feedback control mechanisms, the adaptability and stability of the system are further improved, thereby effectively controlling the propagation of the light beam in the multi-layer optical storage medium.

[0154] Embodiment Nine

[0155] To solve the problem of dynamic interlayer crosstalk suppression in multi-layer optical storage media, this embodiment further optimizes the multi-layer optical storage data reading system for dynamic interlayer crosstalk suppression. The system includes a tomographic signal acquisition module, a dynamic crosstalk prediction module, an optical parameter control module, an optical path compensation module, and a signal processing module. Through the coordinated work of these modules, the collection of target layer and adjacent layer interference signals, the generation of crosstalk suppression parameters, the dynamic adjustment of numerical aperture and wavefront phase, the compensation of sub-micron interlayer spacing error, and the separation, correction and data integrity verification of signals are realized.

[0156] Firstly, the tomographic signal acquisition module includes a low coherence interferometer and a high-speed photodetector, which is used to collect the interference signals of the target layer and the adjacent layers. The low coherence interferometer is a technology that uses low coherence light source for interference measurement, with high resolution and high sensitivity. The high-speed photodetector can capture the interference signal in real time, ensuring the accuracy and real-time of the data. It should be understood that the combination of low coherence interferometer and high-speed photodetector can realize accurate measurement of different layers in multi-layer medium.

[0157] Further, the dynamic crosstalk prediction module integrates a deep belief network (DBN) and a frequency domain convolution unit to generate crosstalk suppression parameters based on phase distribution parameters. Deep belief network is an unsupervised learning method based on neural network, which can effectively predict and suppress interlayer crosstalk through multi-level feature extraction and reconstruction. The frequency domain convolution unit further improves the accuracy of crosstalk suppression by processing the signal in the frequency domain. It should be understood that the combination of deep belief network and frequency domain convolution unit can efficiently suppress crosstalk in time and frequency dimensions.

[0158] Further, the optical parameter adjustment module is composed of a variable aperture objective lens, a liquid crystal phase modulator, and a piezoelectric ceramic driver, which is used to dynamically adjust the numerical aperture and wavefront phase. The variable aperture objective lens can adjust its aperture size according to the demand, thereby changing the numerical aperture of the system. The liquid crystal phase modulator generates the required wavefront phase distribution by applying different voltages, achieving precise control of the light beam. The piezoelectric ceramic driver is used to drive the movement of the objective lens, achieving high-precision position adjustment. It should be understood that through the cooperative work of these components, dynamic adjustment of the optical parameters can be achieved, thereby effectively suppressing interlayer crosstalk.

[0159] Further, the optical path compensation module includes a multi-channel interference reference light path, a piezoelectric ceramic displacement stage, and a temperature control subsystem, which is used to achieve sub-micron interlayer spacing error compensation. The multi-channel interference reference light path realizes high-precision detection of interlayer spacing error through interference measurement of multiple reference light paths. The piezoelectric ceramic displacement stage is used to accurately adjust the optical path, achieving sub-micron error compensation. The temperature control subsystem maintains the stability of the medium surface temperature through temperature control, reducing the influence of temperature changes on the interlayer spacing. It should be understood that through the cooperative work of these components, high-precision compensation of the interlayer spacing error can be achieved, thereby improving the accuracy of data reading.

[0160] Further, the signal processing module embeds an asymmetric wavelet transform algorithm and a hash verification unit, which is used for signal separation, correction, and data integrity verification. The asymmetric wavelet transform algorithm is an efficient data processing method that can effectively separate and correct noise and distortion in the signal through multi-scale decomposition. The hash verification unit calculates the hash digest of the corrected data and compares it with the pre-stored hash value to verify the data integrity. It should be understood that through the combination of these algorithms, efficient processing of the signal and verification of data integrity can be achieved.

[0161] In order to further improve the anti-interference ability of the system, an adaptive filter can be introduced. The adaptive filter can automatically adjust the filtering parameters according to the characteristics of the input signal, thereby achieving effective suppression of noise. For example, adaptive filtering can be achieved through the least mean square error (LMS) algorithm or the recursive least squares (RLS) algorithm. It should be understood that the adaptive filter can be flexibly adjusted according to actual needs, thereby improving the anti-interference ability of the system.

[0162] In addition, a multi-modal fusion mechanism can be introduced to improve the accuracy of the system by fusing data from different modalities. For example, optical signals and acoustic signals can be fused to achieve more comprehensive suppression of interlayer crosstalk through the complementarity of multi-modal data. It should be understood that the multi-modal fusion mechanism can fully utilize the advantages of different modal data to improve the overall performance of the system.

[0163] To further improve the intelligent level of the system, an intelligent control system can be introduced. The intelligent control system realizes adaptive control of the system through machine learning and artificial intelligence technology. For example, intelligent control of the optical parameter regulation module and the optical path compensation module can be realized through reinforcement learning algorithm, thereby improving the automation degree and stability of the system. It should be understood that the intelligent control system can automatically adjust the system parameters according to the actual needs, thereby improving the intelligent level of the system.

[0164] Through this embodiment, by optimizing and refining each module of the multi-layer optical storage data reading system for dynamic interlayer crosstalk suppression, and introducing adaptive filter, multi-modal fusion and intelligent control system mechanisms, the anti-interference ability and intelligent level of the system are further improved, thereby effectively suppressing the interlayer crosstalk in the multi-layer optical storage medium.

[0165] Embodiment Ten

[0166] To solve the problem of accurate control of optical path compensation in the multi-layer optical storage medium, the optical path compensation module is further optimized in this embodiment. The module includes a dynamic optical path compensator, a wavefront phase modulator and a temperature control subsystem. Through the cooperative work of these components, accurate calculation of the optical path compensation amount, dynamic modulation of the wavefront phase and stable control of the medium surface temperature are realized.

[0167] Specifically, the compensation amount calculation method of the dynamic optical path compensator is:

[0168] where ΔL is the optical path compensation amount, α m and β are the medium thermal-optical coefficient and the mechanical deformation weight factor, ΔT m is the temperature change amount of the mth temperature control partition, Δλ m is the wavelength drift amount, Δz is the axial deformation amount, n is the medium refractive index, and T is the absolute temperature of the medium. It should be understood that through the above formula, the temperature change, wavelength drift and mechanical deformation can be considered comprehensively to realize accurate calculation of the optical path compensation amount.

[0169] Further, the wavefront phase modulator is configured to generate a conjugate phase field according to the crosstalk suppression parameters, for canceling the wavefront distortion caused by interlayer crosstalk. The wavefront phase modulator generates the required wavefront phase distribution by applying different voltages, thereby realizing accurate control of the light beam. Specifically, by generating a conjugate phase field matched with the crosstalk suppression parameters, the wavefront distortion caused by interlayer crosstalk can be effectively canceled, thereby improving the accuracy of data reading. It should be understood that the wavefront phase modulator has high resolution and fast response characteristics, and is suitable for dynamic beam control.

[0170] Further, a temperature control subsystem is used to maintain the medium surface temperature fluctuation within ±0.1℃. The temperature control subsystem ensures the stability of the medium surface temperature through precise temperature control technology, thereby reducing the impact of temperature changes on the layer spacing. Specifically, the temperature control subsystem includes temperature sensors, heating elements, and cooling elements, which realize precise control of the medium surface temperature through real-time monitoring and feedback control. It should be understood that the introduction of the temperature control subsystem can significantly improve the stability and reliability of the system.

[0171] To further improve the accuracy of temperature control, a multi-channel temperature control mechanism can be introduced. By dividing the medium surface into multiple temperature control zones and independently controlling the temperature of each zone, fine adjustment of the temperature distribution can be achieved. For example, multiple temperature sensors and heating elements can be arranged on the medium surface, and a multi-channel temperature controller can be used to independently control the temperature of each zone. It should be understood that the multi-channel temperature control mechanism can achieve fine adjustment of the temperature distribution and improve the accuracy of temperature control of the system.

[0172] To further improve the stability of the system, a multi-level feedback control mechanism can be introduced. By setting feedback control loops at different levels, multi-level control of the system can be achieved. For example, a first-level feedback control loop can be set at the temperature control level, and a second-level feedback control loop can be set at the optical path compensation level. Through multi-level feedback control, the error of the system can be gradually eliminated, and the stability and accuracy of the system can be improved. It should be understood that the multi-level feedback control mechanism can achieve multi-level control and improve the overall performance of the system.

[0173] To further improve the fault diagnosis and maintenance capability of the system, an intelligent diagnosis system can be introduced. The intelligent diagnosis system realizes real-time monitoring and fault diagnosis of the system state through machine learning and data analysis technology. For example, an anomaly detection algorithm can be used to monitor key parameters of the system in real time, and an alarm can be sent in a timely manner when an anomaly is found. It should be understood that the intelligent diagnosis system can realize real-time monitoring and fault diagnosis of the system state, and improve the reliability and maintenance efficiency of the system.

[0174] Through this embodiment, by optimizing and refining each component of the optical path compensation module, and introducing mechanisms such as multi-channel temperature control, adaptive temperature compensation, multi-level feedback control, and intelligent diagnosis system, the temperature control accuracy, stability, and intelligent level of the system are further improved, thereby effectively realizing compensation of sub-micron level layer spacing error.

[0175] While the application has been particularly shown and described with reference to the preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details can be made therein without departing from the spirit and scope of the application, which is defined by the appended claims and their equivalents.

Claims

1. A method for reading data from a multi-layer optical storage with dynamic interlayer crosstalk suppression, characterized in that, The method comprises the following steps: A tomographic interference signal of a target data layer in a multi-layer optical storage medium is collected by a low-coherence interferometer, and a spatial frequency characteristic of an adjacent data layer is synchronously acquired, the spatial frequency characteristic including a power spectral density, a frequency bandwidth, and a phase distribution parameter; A coherent coupling coefficient of the adjacent layer and the target layer is calculated based on the phase distribution parameter, an initial cross-talk weight matrix is generated by a frequency domain convolution operation, and the matrix is input into a deep confidence network to generate a cross-talk suppression parameter including a frequency band suppression priority and a phase compensation amount; A control signal of a variable aperture diaphragm is generated according to the frequency band suppression priority, a numerical aperture of an objective lens is adjusted in real time by a piezoelectric ceramic driver to continuously change in a range of 0.45 microns to 0.85 microns, and an incident angle of a scanning light beam is synchronously adjusted to match a light beam propagation path after the numerical aperture is changed; An interlayer spacing error caused by non-uniformity of a refractive index of the medium is compensated at a sub-micron level through a synergistic effect of a multi-channel interference reference light path and a dynamic optical path compensator; In a signal processing module, an aliasing signal is subjected to multi-scale decomposition, an adaptive threshold filter is applied to a high-frequency component according to the frequency band suppression priority, a phase reconstruction is performed on a low-frequency component by an asymmetric Gaussian function to eliminate residual cross-talk, and finally corrected target layer data is output through an inverse wavelet transform.

2. The method for dynamic interlayer crosstalk suppression for multilayer optical storage data reading as claimed in claim 1, wherein, The dynamic optical path compensator captures transient changes in the interlayer spacing through a second-order differential term, and calculates an optical path compensation amount based on a thermal-optic coefficient of the medium and a mechanical deformation weight factor, and the steps include: A distributed temperature sensor is used to monitor temperature changes and wavelength drifts of each partition on the surface of the medium in real time; Based on the thermal-optic coefficient of the medium and the mechanical deformation weight factor, an influence amount of a refractive index change caused by a temperature gradient on an optical path is calculated; A second-order differential operation is used to capture transient change trends of the interlayer spacing error to generate a dynamic compensation signal; A piezoelectric ceramic displacement stage is driven to adjust the optical path length of the reference light path with sub-micron level precision, and the deviation of the focal point position of the target layer from the theoretical value is controlled within ± 50 nm.

3. The method for dynamic interlayer crosstalk suppression for multilayer optical storage data reading as recited in claim 1, wherein, The coherent coupling coefficient quantifies the strength and spatial correlation of interlayer cross-talk through the combination of frequency domain integration and an exponential decay function, and the coherent coupling coefficient is calculated in the following manner: where Γ i,j denotes the coherent coupling coefficient between the i-th and j-th layers, S i (λ) is the spectral signal of the i-th layer, S j (λ) is the spectral signal of the j-th layer, S j ∗ (λ) is the phase-conjugated spectral signal of the j-th layer, Δz i,j is the layer spacing error, σ z is the standard deviation of the refractive index fluctuation of the medium, k is the index of the spatial sampling position, and N is the total number of sampling points.

4. The method for dynamic interlayer crosstalk suppression for multilayer optical storage data reading as recited in claim 1, wherein, The phase reconstruction process of the asymmetric Gaussian function includes: According to the time domain attenuation characteristics in the cross-talk suppression parameter, the widening coefficients of the positive and negative time axes are defined respectively; The phase distribution of the low-frequency component is asymmetrically weighted to suppress the tailing effect caused by reflection from the adjacent layer; The widening coefficients are adjusted by an iterative optimization algorithm so that the mean square error of the reconstructed waveform is less than a preset threshold.

5. The method for dynamic interlayer crosstalk suppression for multilayer optical storage data reading as recited in claim 1, wherein, The data reading method further comprises a pre-calibration step: A reference mark array is written on the surface of the medium, and interlayer spacing calibration data is generated by scanning with a low-power probe light beam; A three-dimensional distribution map of the refractive index of the medium is constructed based on the calibration data, the three-dimensional distribution map realizes high-precision modeling of the spatial refractive index deviation through a three-dimensional sinc function interpolation algorithm, and is pre-stored in the signal processing module; The three-dimensional sinc function interpolation algorithm includes: The coordinates of the reference marks and the corresponding refractive index deviation coefficients are extracted; A three-dimensional sinc function kernel is generated with each mark as the center, and the weight of the kernel is determined by the mark spacing and the thermal expansion coefficient of the medium; A continuous refractive index distribution model is generated by linearly superimposing all kernel functions, and is discretized into grid data for real-time calling.

6. The method for dynamic interlayer crosstalk mitigation for multilayer optical storage data reading as recited in claim 1, wherein, The data reading method further includes a post-validation step: A hash digest is calculated for the corrected target layer data and compared with a pre-stored hash value; If the digest values do not match, a read operation of the redundant data layer is triggered, and the abnormal data is repaired through the parity check bit and error correction code; The repaired data is written back to the target layer and the hash value is updated.

7. The method for dynamic interlayer crosstalk suppression for multilayer optical storage data reading as recited in claim 1, wherein, The training process of the deep belief network includes: Collecting historical crosstalk signals under different medium types and interlayer spacing error conditions as training sets; Optimizing network weights through contrast divergence algorithm to reduce the mean square error of network output crosstalk suppression parameters to below 0.1%; Using a generative adversarial network to enhance the noise diversity of training data to improve the model's generalization ability to unknown media.

8. The method for dynamic interlayer crosstalk suppression for multilayer optical storage data reading of claim 1, wherein, The control signal generation logic of the variable aperture diaphragm includes: Determine the theoretical value of the target numerical aperture according to the frequency band suppression priority; Real-time monitor the actual numerical aperture of the objective lens through a closed-loop feedback circuit and compare it with the theoretical value; If the deviation exceeds 5%, adjust the piezoelectric ceramic driving voltage until the actual aperture converges within the theoretical range; The adjustment step of the incident angle of the scanning beam includes: Calculate the refraction angle of the beam inside the medium according to the change amount of the numerical aperture of the objective lens; Generate a wavefront phase distribution matching the refraction angle through a liquid crystal phase modulator; Use a diffractive optical element to pre-distort the beam to ensure the axial stability of the focal point position.

9. A dynamic interlayer crosstalk suppressed multi-layer optical storage data reading system for implementing the dynamic interlayer crosstalk suppressed multi-layer optical storage data reading method according to any one of claims 1 to 8, characterized in that, The data reading system includes: A tomographic signal acquisition module containing a low-coherence interferometer and a high-speed photodetector for acquiring interference signals of the target layer and adjacent layers; A dynamic crosstalk prediction module integrating a deep belief network and a frequency domain convolution unit to generate crosstalk suppression parameters based on phase distribution parameters; An optical parameter control module composed of a variable aperture objective lens, a liquid crystal phase modulator, and a piezoelectric ceramic driver for dynamically adjusting the numerical aperture and wavefront phase; An optical path compensation module including a multi-channel interference reference light path, a piezoelectric ceramic displacement stage, and a temperature control subsystem for sub-micron interlayer spacing error compensation; A signal processing module embedded with an asymmetric wavelet transform algorithm and a hash verification unit for signal separation, correction, and data integrity verification.

10. The multi-layer optical memory data reading system with dynamic suppression of interlayer crosstalk of claim 9, wherein, The optical path compensation module further includes: The calculation method of the compensation amount of the dynamic optical path compensator is: wherein, AL is the optical path compensation, a m and β are the thermal-optical coefficient of the medium and the mechanical deformation weight factor, ΔT m is the temperature change of the mth temperature control zone, M is the total number of temperature control zones, Δλ m is the wavelength shift, Δz is the axial deformation, n is the refractive index of the medium, and T is the absolute temperature of the medium. A wavefront phase modulator configured to generate a conjugate phase field based on the crosstalk suppression parameters to counteract the wavefront distortion of interlayer crosstalk; A temperature control subsystem for maintaining the surface temperature of the medium within ±0.1℃.

Images