Atmospheric turbulence parameter inversion method and system of two-dimensional orbital angular momentum spectrum

Abstract

The invention belongs to the technical field of atmospheric turbulence, and discloses an atmospheric turbulence parameter inversion method and system for a two-dimensional orbital angular momentum spectrum, and the method comprises the steps: generating initial light field distribution based on a preset vortex beam; generating a multi-layer random phase screen according to the target atmospheric turbulence path simulation parameters; transmitting the light field through a phase screen, and obtaining light field distribution at the tail end of the path; calculating an OAM spectrum of the vortex beam, wherein a radial index is selected; preprocessing the OAM spectrum, and inputting the preprocessed OAM spectrum into the trained support vector machine model to obtain an atmospheric turbulence parameter inversion value; the accuracy is evaluated by comparing an inversion value with a real label, topological charge and turbulence disturbance characteristics in the radial direction are captured through a radially decomposed OAM spectrum, the precision and robustness of atmospheric turbulence parameter inversion are effectively improved, and the problems that information is lost, the inversion precision is low, and the efficiency is low due to the fact that the radial distribution characteristics are ignored in traditional OAM spectrum analysis are solved. And a machine learning method is insufficient in generalization ability when the OAM spectrum width is narrow.

Description

Technical Field

[0001] This invention belongs to the field of atmospheric turbulence technology, and relates to a method and system for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum. Background Technology

[0002] Atmospheric turbulence, caused by irregular fluctuations in temperature, humidity, and pressure, is a random fluctuation in the atmospheric refractive index and a key factor affecting the propagation quality of light beams in free space. During propagation, atmospheric turbulence leads to various degradation effects on the light beam, including beam spreading, beam drift, scintillation, wavefront distortion, and reduced coherence, directly limiting the performance of optical applications that rely on atmospheric channels.

[0003] In the study of atmospheric turbulence parameters, vortex beams carrying orbital angular momentum (OAM) exhibit unique advantages. When the wavefront structure interacts with the turbulent medium, in addition to the intensity fluctuations and wavefront distortions of conventional beams, the helical phase structure is disturbed, leading to OAM state crosstalk and even vortex core splitting. This high sensitivity to turbulent disturbances makes vortex beams a potential optical probe for detecting atmospheric turbulence characteristics.

[0004] Existing atmospheric turbulence parameter inversion techniques based on vortex beams mainly fall into two categories. One category is traditional OAM spectral analysis techniques, which model the power distribution of the received beam in different OAM modes to reflect the degree of turbulence disturbance on the beam's phase structure. However, this only reflects changes in the topological charge dimension, completely ignoring the radial distribution characteristics of turbulence disturbances, leading to the loss of a large amount of valuable spatial detail information and making it difficult to accurately identify turbulence parameters in complex turbulent environments. Furthermore, under specific turbulence scales or noise influences, different turbulence states may induce similar OAM spectral broadening, further limiting the inversion accuracy and robustness.

[0005] Another type is machine learning-based inversion techniques. Among them, support vector machine (SVM) is an efficient supervised learning algorithm that can effectively handle high-dimensional and nonlinear data. However, it is highly dependent on the OAM spectrum with sufficient modal width. If the OAM spectrum is narrow, it is not possible to extract highly discriminative turbulence information from finite spectral data, and it is difficult to separate similar data in high-dimensional space, which leads to a significant decrease in its generalization ability and inversion accuracy under complex turbulence conditions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: This invention provides a method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum. The method includes: generating an initial vortex beam optical field distribution based on a preset vortex beam. ; A multi-layer random phase screen is generated based on the simulation parameters of the target atmospheric turbulence path; the initial vortex beam light field distribution The path end is obtained through multi-layer random phase screen transmission. Light field distribution at According to the light field distribution Calculate the OAM spectrum of a vortex beam The OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; the inversion values ​​are compared with the true labels to obtain the inversion accuracy.

[0008] Furthermore, the preset vortex beam is a Bessel-Gaussian beam.

[0009] in, The light field of a Bessel-Gaussian beam. In polar coordinates, The normalization constant is Category 1 Bessel function of order 1, Let be the topological charge number of the Bessel-Gaussian beam. It is the transverse wave vector. , For wavelength, It is a semi-cone angle. These are the radius coordinates in polar coordinates. It is a spiral phase. The imaginary unit, These are angular coordinates in polar coordinates. It is a Gaussian term.

[0010] Furthermore, based on the light field distribution Calculate the OAM spectrum of a vortex beam ,in and This includes: distributing the light field Complex amplitude light field distribution converted to polar coordinates According to the discrete radial index The complex amplitude optical field distribution Divided into A discrete concentric ring; with radial spacing Inner light field Projected onto a helical substrate Above, perform orbital angular momentum OAM mode decomposition to obtain the first... The topological charge number of the concentric rings is Complex amplitude of OAM mode Based on complex amplitude Calculate the first The topological charge number of the concentric rings is of spectral density ;combination OAM spectral density of concentric rings get Spectrum .

[0011] Furthermore, the complex amplitude for:

[0012] in, For the first The radius of the concentric rings, For the first The radius of the concentric rings, for The complex amplitude light field distribution at that location, For the helical base The projection, The topological charge is within the OAM spectrum range. The imaginary unit, These are angular coordinates in polar coordinates. For angular infinitesimal elements, It is a radial infinitesimal element.

[0013] Furthermore, the OAM spectrum The The topological charge number of the concentric rings is OAM spectrum for:

[0014] in, For complex amplitude, For the first The topological charge inside the layered ring is The amount, For the first The total amount of all topological loads within the annulus, from negative infinity to positive infinity. For topological charges that vary from negative infinity to positive infinity.

[0015] Furthermore, the OAM spectrum Preprocessing is performed before inputting the data into the trained SVM model, including: processing the OAM spectrum... Straightening and feature standardization are performed to form a dataset with corresponding integer labels. The data is divided into a training set and a validation set. The SVM model is trained using the training set, and the validation set is input into the trained SVM model. The output of the trained SVM model is the inversion value of the atmospheric turbulence parameters of the target atmospheric turbulence path.

[0016] Furthermore, the first step in generating the multi-layer random phase screen... A random phase screen for atmospheric turbulence The process includes: setting atmospheric turbulence phase screen parameters, which include the physical dimensions of the phase screen and the atmospheric coherence length; determining a set of spatial frequency points based on the physical dimensions of the phase screen; generating two independent complex Gaussian white noise random number matrices for each frequency point in the set; and calculating random coefficients based on the complex Gaussian white noise random number matrices, each frequency point, and the atmospheric coherence length. , No. Summing the set of spatial frequency points, the random coefficients are calculated. , No. Random coefficients for sub-low frequency harmonic compensation Performing a Fourier series expansion, we obtain the... A random phase screen for atmospheric turbulence .

[0017] Furthermore, the random coefficient for:

[0018] The first Random coefficients for sub-low frequency harmonic compensation for:

[0019] in, and Given two independent complex Gaussian white noise random number matrices, Let x be the length of the phase screen in the x and y directions. The atmospheric coherence length of a single turbulent phase screen. For external scale spatial frequency, , For the external scale of atmospheric turbulence, For internal scale spatial frequency. , For the internal scale of atmospheric turbulence, For the first along the x-direction One frequency component, , For the index of the frequency component in the x-direction, For the first along the y direction One frequency component, , This is the index of the frequency component in the y-direction. For low-frequency harmonic compensation, , For the first Second-low frequency harmonic compensation along the x-direction One frequency component, , For the first Second-low frequency harmonic compensation along the y-direction Each frequency component.

[0020] Furthermore, the initial vortex beam light field distribution is obtained at the end of the path by passing through multiple layers of random phase screens. Light field distribution at include: The angular spectrum method based on Fourier transform will be used to... The input light field of the next iteration propagates in free space. The distance is used to obtain the intermediate light field, where, Transmitting distance in free space , The number of multi-layer random phase screens; Multiplying the multilayer random phase screen with the intermediate optical field yields the optical field after applying phase perturbation; The light field after the phase perturbation is applied continues to propagate in free space. The distance is used to obtain the output light field; Repeat the above steps Next, at the transmission endpoint Light field distribution obtained at the location .

[0021] The present invention also provides an atmospheric turbulence parameter inversion system for two-dimensional orbital angular momentum spectra, comprising: a first generation module for generating an initial vortex beam optical field distribution based on a preset vortex beam. The second generation module is used to generate a multi-layer random phase screen based on the simulation parameters of the target atmospheric turbulence path; the transmission module is used for the initial vortex beam light field distribution. The path end is obtained through multi-layer random phase screen transmission. Light field distribution at ; Calculation module: used to calculate based on light field distribution Calculate the OAM spectrum of a vortex beam ,in Processing module: used to process the OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; the comparison module is used to compare the inversion values ​​with the true labels to obtain the inversion accuracy.

[0022] Compared with the prior art, the present invention has the following beneficial technical effects: This invention discloses a method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum, with light field distribution. Calculate the OAM spectrum of a vortex beam It expands the dimension and intensity of feature information, enabling a more refined characterization of the spatial modulation effect of turbulence on beam phase. It effectively overcomes the averaging effect of one-dimensional OAM spectrum on spatial information. The radial distribution combined with inner ring features significantly improves the accuracy and robustness of turbulence parameter estimation.

[0023] This invention discloses a method for inverting atmospheric turbulence parameters based on two-dimensional orbital angular momentum spectra. It employs a modified Von Kármán atmospheric phase power spectrum and subharmonic compensation techniques to construct multi-layered random phase screens, accurately simulating the propagation of vortex beams in turbulence. The beam propagation is handled using a split-step Fourier method, segmenting the path and iteratively calculating the superposition effect of free-space propagation and phase screen perturbations, realistically reproducing phenomena such as beam spreading and OAM mode dispersion caused by turbulence. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the structure of the atmospheric turbulence parameter inversion method of the two-dimensional orbital angular momentum spectrum of the present invention; Figure 2 This is a schematic diagram illustrating the principle of an atmospheric turbulence parameter inversion method based on a two-dimensional orbital angular momentum spectrum according to the present invention, where a represents the complex amplitude light field distribution. Divided into A schematic diagram of the structure of discrete concentric rings, where b represents the light field distribution. Complex amplitude light field distribution transformed to polar coordinates c is OAM spectrum of concentric ring regions ; Figure 3 This is a schematic diagram of the transmission of a vortex beam through a multi-layer turbulent phase screen structure in an embodiment of the present invention; Figure 4 This is a schematic diagram simulating the transmission and modal diffusion of a vortex beam in turbulence in an embodiment of the present invention; Figure 5 This is a schematic diagram illustrating the influence of the radial structure of the vortex beam in an embodiment of the present invention. a, b, and c represent the semi-cone angles. and Simulated intensity (left) and phase (right) distributions of the BG beam at time t, where d, e, and f represent the SVM classification accuracy and the radial segmentation number for each corresponding half-cone angle. Distribution; Figure 6 This is a schematic diagram illustrating the influence of topological charge and radial sampling index on the vortex beam in an embodiment of the present invention. a, b, and c represent the simulated intensity (left) and phase (right) distributions of the BG beams with initial topological charges of 1, 3, and 5, respectively. d, e, and f represent the SVM classification accuracy and the radial segmentation number for each corresponding topological charge. Distribution; Figure 7 This is a schematic diagram illustrating the impact of the OAM spectrum range on the recognition performance of the SVM model in an embodiment of the present invention; Figure 8 This is a schematic diagram illustrating the influence of radial sampling depth in an embodiment of the present invention, where 'a' represents the decomposition of the beam cross-section into... A schematic diagram of concentric rings, where b, c, and d are used. and The total radial segmentation calculation, where e, f, and g are the radial segmentation numbers corresponding to b, c, and d respectively. layer( ≤ SVM classification accuracy at radial sampling depth. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] Example 1 This invention discloses a method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum, comprising the following steps: The initial vortex beam light field distribution is generated based on the preset vortex beam. ; A multi-layer random phase screen is generated based on the simulation parameters of the target atmospheric turbulence path; the initial vortex beam light field distribution The path end is obtained through multi-layer random phase screen transmission. Light field distribution at According to the light field distribution Calculate the OAM spectrum of a vortex beam ,in The OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; the inversion values ​​are compared with the true labels to obtain the inversion accuracy.

[0027] This invention uses the OAM spectrum, but it can also be replaced by extracting key statistical features of the OAM spectrum as input to a machine learning model or deep learning network. For example, features of the OAM spectrum may include OAM spectrum purity, average OAM value, OAM spectrum broadening (variance), etc., which can reduce the dimensionality of the input data.

[0028] Other types of beams with helical phases and multi-ring structures can also be used as probes, such as higher-order (ring) Laguerre-Gaussian beams, vector vortex beams, etc. Their interaction mode with turbulence is different from that of standard Laguerre-Gaussian vortex beams, and the characteristic changes after transmission (intensity, phase, polarization, OAM spectrum, etc.) can also be used to invert turbulence parameters.

[0029] Specifically, simulation parameters are first set to simulate the target atmospheric turbulence path. These parameters include turbulence intensity parameters and path parameters characterizing the target atmospheric turbulence path. The turbulence intensity parameters include Friedel-Crafts parameters. Turbulent outer scale Internal scale Reynolds number Path parameters include transmission distance and the number of multi-layer random phase screens .

[0030] The initial vortex beam light field distribution is generated based on the preset vortex beam. The parameters of a vortex beam include wavelength. Initial field strength Waist radius Topological load semi-cone angle .

[0031] The vortex beam is a Bessel-Gaussian beam.

[0032] in, The light field of a Bessel-Gaussian beam. In polar coordinates, The normalization constant is Category 1 Bessel function of order 1, Let be the topological charge number of the Bessel-Gaussian beam. It is the transverse wave vector. , For wavelength, It is a semi-cone angle. These are the radius coordinates in polar coordinates. It is a spiral phase. The imaginary unit, These are angular coordinates in polar coordinates. It is a Gaussian term.

[0033] Multi-layer random phase screens are generated based on simulation parameters of the target atmospheric turbulence path.

[0034] Based on turbulence intensity parameters and path parameters, a modified Von Kármán atmospheric phase power spectrum is used, and a multi-layer random phase screen is generated to simulate the atmospheric turbulence path by combining subharmonic compensation techniques.

[0035] The modified Von Kármán atmospheric phase power spectrum can be replaced by other atmospheric turbulence power spectrum models, such as the classic Kolmogorov spectrum, Tatarskii spectrum, or Hill spectrum that considers internal scale effects, to adapt to different turbulence description needs or simplify the model.

[0036] To ensure the effectiveness of numerical simulations during the generation of multi-layer random phase screens, the following constraints must be met: Constraint 1: ,in, The interval between adjacent phase screens The number of sampling points. The sampling interval is... The wavelength of the emitted light source; Constraint 2: ,in, The number of multi-layer random phase screens. This represents the total length of the atmospheric turbulence path. This represents the maximum spacing length between adjacent phase screens.

[0037] The first generation of multi-layer random phase screens A random phase screen for atmospheric turbulence The process includes the following steps: setting the atmospheric turbulence phase screen parameters, which include the physical dimensions of the phase screen. Coherence length with the atmosphere ; Based on the physical size of the phase screen Determine the set of spatial frequency points Spatial frequency point set It includes a standard Fourier transform network store and subharmonic frequency points for low-frequency compensation. This indicates the length of the phase screen in the x and y directions. Represents the first [number] along the x-direction. Each frequency component Indicates the first [number] along the y-direction Each frequency component.

[0038] Based on spatial frequency point set Each frequency point in Generate two independent complex Gaussian white noise random number matrices and ,in, Matrix dimension index; Based on complex Gaussian white noise random number matrix and Each frequency point and atmospheric coherence length Calculate random coefficients , No. , It is the number of low-frequency harmonic compensations. Indicates the first Random coefficients for sub-low frequency harmonic compensation; Random coefficients , No. The value of the modified Von Kármán atmospheric turbulence phase power spectrum at the corresponding frequency point is determined by the random coefficient. for:

[0039] No. Random coefficients for sub-low frequency harmonic compensation for:

[0040] in, and Given two independent complex Gaussian white noise random number matrices, Let x be the length of the phase screen in the x and y directions. The atmospheric coherence length of a single turbulent phase screen. For external scale spatial frequency, , For the external scale of atmospheric turbulence, For internal scale spatial frequency. , For the internal scale of atmospheric turbulence, For the first along the x-direction One frequency component, , For the index of the frequency component in the x-direction, For the first along the y direction One frequency component, , This is the index of the frequency component in the y-direction. For low-frequency harmonic compensation, , For the first Second-low frequency harmonic compensation along the x-direction One frequency component, , For the first Second-low frequency harmonic compensation along the y-direction Each frequency component.

[0041] For the set of spatial frequency points Summation is performed using the random coefficients of the Fourier series. and Performing a Fourier series expansion, we obtain the... A random phase screen for atmospheric turbulence .

[0042] Fourier transform can be replaced by other numerical calculation methods, such as the parabolic propagation method, which may be necessary, especially when dealing with strong turbulence or large-angle propagation problems.

[0043] The initial vortex beam light field distribution The path end is obtained through multi-layer random phase screen transmission. Light field distribution at ,like Figure 3 He Ru Figure 4 As shown.

[0044] Numerical methods were used to simulate the light field distribution of the initial vortex beam. The split-step Fourier method is used to divide the total transmission path L into... There are 1 sub-paths, and the length of each sub-path is 1. , corresponding to a phase screen The simulation process is executed iteratively. The next, of which The simulation of each sub-path includes: The angular spectrum method based on Fourier transform will be used to... The input light field of the next iteration Transmission in free space The distance is used to obtain the intermediate light field. .

[0045]

[0046] in, and These represent the two-dimensional Fourier transform and the inverse Fourier transform, respectively. For transmission distance The corresponding free space transfer function, Transmitting distance in free space;

[0047] The wavelength of the light beam. These are spatial frequency coordinates.

[0048] The first of the multi-layer random phase screen A random phase screen for atmospheric turbulence With intermediate light field Multiplying yields the optical field after applying phase perturbation. ;

[0049] in, For the applied phase perturbation, It is the first Each phase screen.

[0050] The light field after applying phase perturbation Continue transmission in free space The distance was used to calculate the output light field using the angular spectrum method. ;

[0051] Repeat the above steps Next, it can simulate the beam passing through in sequence Each of the following transmission sub-paths contains [number] transmission sub-paths. Free space transmission, phase screen disturbance, The total turbulent path represented by free space transport The cumulative transmission effect eventually yields the light field distribution at the transmission endpoint z=L. .

[0052] According to the light field distribution Calculate the vortex beam Spectrum ,in and .

[0053] Light field distribution Complex amplitude light field distribution transformed to polar coordinates ; According to the discrete radial index , and The complex amplitude optical field distribution Divided into A discrete concentric ring, such as Figure 2 As shown; radial spacing Inner light field Projected onto a helical substrate Above, orbital angular momentum is measured. Mode decomposition yields the first... The topological charge number of the concentric rings is of Complex amplitude of the mode ;

[0054] in, For the first The radius of the concentric rings, For the first The radius of the concentric rings, for Complex amplitude light field distribution at the location , For the helical base The projection, and These are the radius coordinates and coordinates in polar coordinates. For angular infinitesimal elements, It is a radial infinitesimal element.

[0055] Based on complex amplitude Calculate the first The topological charge number of the concentric rings is OAM spectrum ;

[0056] in, For complex amplitude, For the first The topological charge inside the layered ring is The amount, For the first The total amount of all topological loads within the annulus, from negative infinity to positive infinity. For topological charges that vary from negative infinity to positive infinity.

[0057] combination OAM spectral density of concentric rings Obtain the OAM spectrum .

[0058] The OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; OAM spectrum Perform straightening and feature standardization preprocessing operations, and combine them with the corresponding integer labels to form a dataset, for example, " =1574, =0.02” corresponds to the integer label “0” =1574, =0.015” corresponds to the integer label “1”, and so on. The dataset is then divided into training and validation sets in a 7:3 ratio. The training set is fed into the SVM model for training, and the validation set is then fed into the trained SVM model. The straightening preprocessing operation involves converting a set of data of size [missing information]. 2D Spectral processing yields 1D vectors suitable for SVM models. Feature normalization preprocessing involves standardizing the features of the straightened 1D vectors according to the core formula. Transform the data into a standard normal distribution (mean 0, variance 1), where This is the original data. For the data after feature standardization, The characteristic mean, The standard deviation is the standardization preprocessing operation for this feature, which can be implemented using the StandardScale function in the scikit-learn library in Python.

[0059] The output of the SVM model is the inversion value of the atmospheric turbulence parameters of the target atmospheric turbulence path. The inversion value is compared with the true label to obtain the inversion accuracy.

[0060] The dataset was randomly and non-repeatingly divided into training and validation sets 30 times (each time the dataset and validation set were different), and the straightening and standardization preprocessing steps were independently repeated 30 times. The results of each step were recorded and the fluctuation range of the inversion accuracy was observed.

[0061] like Figure 5 As shown, the influence of the radial structure of the vortex beam is investigated: changing the semi-cone angle of the vortex beam... They are respectively , and Radial sampling index Set the values ​​to 1, 4, 8, 16, and 32 to obtain new samples and datasets. Feed the new datasets into the SVM model and perform cross-validation to compare the accuracy and stability of the inversion under different half-cone angles.

[0062] like Figure 6 As shown, the effects of topological charge and radial sampling index on the vortex beam are investigated: changing the topological charge of the vortex beam... The radial sampling indices are 1, 3, and 5, respectively. OAM spectra set to 1, 4, 8, 16, and 32. Spectral range (i.e. The range of values ​​is set to [-20, 20]. New samples and datasets are obtained. The new datasets are fed into the SVM model and cross-validated to compare the accuracy and stability of the inversion under different topological charges and radial sampling exponents.

[0063] like Figure 7 As shown, the influence of the OAM spectral range is investigated: the topological charge of the vortex beam is set. =1, half cone angle = Radial sampling index Set to 1 and 8, OAM spectrum Spectral range (i.e. The range of values ​​is set to [-5, 5], [-10, 10], [-15, 15] and [-20, 20]. New samples and datasets are obtained, and the new datasets are fed into the SVM model and cross-validated to compare the accuracy and stability of the inversion under different OAM spectrum ranges.

[0064] like Figure 8 As shown, the influence of radial sampling depth is investigated: the topological charge of the vortex beam is set. =1, half cone angle = OAM spectrum Spectral range (i.e. The value range is set to [-20, 20], and the radial sampling index is... Set to 8, 16, and 32 to pre-OAM spectrum ( ≤ After the layers are truncated, straightened, and feature standardized, the resulting dataset is fed into an SVM for cross-validation to compare the accuracy and stability of the inversion at different radial sampling depths.

[0065] Example 2 The present invention also provides an atmospheric turbulence parameter inversion system for two-dimensional orbital angular momentum spectrum, comprising: a first generation module, a second generation module, a transmission module, a calculation module, a processing module, and a comparison module.

[0066] First generation module: used to generate the initial vortex beam light field distribution based on a preset vortex beam. The second generation module is used to generate a multi-layer random phase screen based on the simulation parameters of the target atmospheric turbulence path; the transmission module is used for the initial vortex beam light field distribution. The path end is obtained through multi-layer random phase screen transmission. Light field distribution at ; Calculation module: used to calculate based on light field distribution Calculate the OAM spectrum of a vortex beam ,in Processing module: used to process the OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; the comparison module is used to compare the inversion values ​​with the true labels to obtain the inversion accuracy.

[0067] The atmospheric turbulence parameter inversion system of the two-dimensional orbital angular momentum spectrum of the present invention can implement the same method steps as the above method, so it will not be described again.

[0068] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

Claims

1. A method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum, characterized in that, Includes the following steps: The initial vortex beam light field distribution is generated based on the preset vortex beam. ; Multi-layer random phase screens are generated based on simulation parameters of the target atmospheric turbulence path; The initial vortex beam light field distribution The path end is obtained through multi-layer random phase screen transmission. Light field distribution at ; According to the light field distribution Calculate the OAM spectrum of a vortex beam ; The OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; The inversion values ​​are compared with the true labels to obtain the inversion accuracy.

2. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 1, characterized in that: The preset vortex beam is a Bessel-Gaussian beam. in, The light field of a Bessel-Gaussian beam. In polar coordinates, The normalization constant is Category 1 Bessel function of order 1, Let be the topological charge number of the Bessel-Gaussian beam. It is the transverse wave vector. , For wavelength, It is a semi-cone angle. These are the radius coordinates in polar coordinates. It is a spiral phase. The imaginary unit, These are angular coordinates in polar coordinates. It is a Gaussian term.

3. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 1, characterized in that: According to the light field distribution Calculate the OAM spectrum of a vortex beam ,in and ,include: The light field distribution Complex amplitude light field distribution converted to polar coordinates ; According to the discrete radial index The complex amplitude optical field distribution Divided into A discrete concentric ring; radial spacing Inner light field Projected onto a helical substrate Above, perform orbital angular momentum OAM mode decomposition to obtain the first... The topological charge number of the concentric rings is Complex amplitude of OAM mode ; Based on complex amplitude Calculate the first The topological charge number of the concentric rings is of spectral density ; combination OAM spectral density of concentric rings get Spectrum .

4. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 3, characterized in that: The complex amplitude for: in, For the first The radius of the concentric rings, For the first The radius of the concentric rings, for The complex amplitude light field distribution at that location, For the helical base The projection, The topological charge is within the OAM spectrum range. The imaginary unit, These are angular coordinates in polar coordinates. For angular infinitesimal elements, It is a radial infinitesimal element.

5. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 4, characterized in that: The OAM spectrum The The topological charge number of the concentric rings is OAM spectrum for: in, For complex amplitude, For the first The topological charge inside the layered ring is The amount, For the first The total amount of all topological loads within the annulus, from negative infinity to positive infinity. For topological charges that vary from negative infinity to positive infinity.

6. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 1, characterized in that, The OAM spectrum Preprocessing is performed before inputting into the trained SVM model, including: The OAM spectrum Straightening and feature standardization are performed to form a dataset with corresponding integer labels. The data is divided into a training set and a validation set. The SVM model is trained using the training set, and the validation set is input into the trained SVM model. The output of the trained SVM model is the inversion value of the atmospheric turbulence parameters of the target atmospheric turbulence path.

7. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 1, characterized in that: The first step in generating multi-layer random phase screens A random phase screen for atmospheric turbulence ,include: Set the atmospheric turbulence phase screen parameters, which include the phase screen physical dimensions and atmospheric coherence length; Determine the set of spatial frequency points based on the physical dimensions of the phase screen; Generate two independent complex Gaussian white noise random number matrices for each frequency point in the spatial frequency point set; Random coefficients are calculated based on the complex Gaussian white noise random number matrix, each frequency point, and the atmospheric coherence length. , No. ; Summing the set of spatial frequency points, random coefficients , No. Random coefficients for sub-low frequency harmonic compensation Performing a Fourier series expansion, we obtain the... A random phase screen for atmospheric turbulence .

8. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 7, characterized in that: The random coefficient for: The first Random coefficients for sub-low frequency harmonic compensation for: in, and Given two independent complex Gaussian white noise random number matrices, Let x be the length of the phase screen in the x and y directions. The atmospheric coherence length of a single turbulent phase screen. For external scale spatial frequency, , For the external scale of atmospheric turbulence, For internal scale spatial frequency. , For the internal scale of atmospheric turbulence, For the first along the x-direction One frequency component, , For the index of the frequency component in the x-direction, For the first along the y direction One frequency component, , This is the index of the frequency component in the y-direction. For low-frequency harmonic compensation, , For the first Second-low frequency harmonic compensation along the x-direction One frequency component, , For the first Second-low frequency harmonic compensation along the y-direction Each frequency component.

9. The method for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum according to claim 1, characterized in that: The initial vortex beam light field distribution is obtained by passing through multiple layers of random phase screens at the end of the path. Light field distribution at include: The angular spectrum method based on Fourier transform will be used to... The input light field of the next iteration propagates in free space. The distance is used to obtain the intermediate light field, where, Transmitting distance in free space , The number of multi-layer random phase screens; Multiplying the multilayer random phase screen with the intermediate optical field yields the optical field after applying phase perturbation; The light field after the phase perturbation is applied continues to propagate in free space. The distance is used to obtain the output light field; Repeat the above steps Next, at the transmission endpoint Light field distribution obtained at the location .

10. A system for inverting atmospheric turbulence parameters from a two-dimensional orbital angular momentum spectrum, characterized in that, include: First generation module: used to generate the initial vortex beam light field distribution based on a preset vortex beam. ; The second generation module is used to generate multi-layer random phase screens based on the simulation parameters of the target atmospheric turbulence path. Transmission module: used for the initial vortex beam light field distribution The path end is obtained through multi-layer random phase screen transmission. Light field distribution at ; Calculation module: used to calculate based on light field distribution Calculate the OAM spectrum of a vortex beam ,in ; Processing module: used to process the OAM spectrum Preprocessing is performed and the data is input into the trained SVM model to obtain the inversion values ​​of atmospheric turbulence parameters of the target atmospheric turbulence path; Comparison module: used to compare the inverted values ​​with the real labels to obtain the inversion accuracy.

Images