Model method-based double-beam link vortex light correction system

Abstract

The invention discloses a double-beam link vortex light correction system and method based on a model method, and the system comprises a double-beam common-path light path which is used for enabling vortex light and planar light to be transmitted in a common path, to pass through a turbulent flow or optical distortion region together, and to carry the same wavefront distortion information; the wavefront correction unit is used for dynamically compensating wavefront distortion information; the image acquisition and control unit is used for acquiring a corrected far-field light spot image after the planar light passes through a distortion area, and generating a driving instruction of the wavefront correction unit according to far-field light spot information to realize closed-loop control; and the topological charge detection module is used for carrying out topological charge detection on the corrected vortex light. According to the invention, indirect correction of the turbulence aberration of the common-path vortex light is realized, so that the correction efficiency and stability of the vortex light are improved.

Description

Technical Field

[0001] This invention belongs to the field of laser communication technology, and particularly relates to a dual-beam link vortex light correction system based on the model method. Background Technology

[0002] Vortex beams, carrying orbital angular momentum (OAM) and possessing infinite degrees of freedom for manipulation, have been widely used in free-space optical communication, primarily to significantly extend channel capacity. However, vortex beams are sensitive to atmospheric turbulence and prone to wavefront distortion, which can destroy topological charge information and severely impact communication quality.

[0003] In existing technologies, the vortex wavefront correction method based on the GS (Gerchberg–Saxton) algorithm is an iterative phase retrieval method commonly used for wavefront correction. Its core idea is to utilize known amplitude information to alternately constrain the amplitude and iteratively estimate the phase in the frequency and spatial domains, thereby gradually approximating the target wavefront. This method does not require a wavefront sensor, relying only on the intensity map, and is theoretically applicable to vortex beams of arbitrary structures. However, its convergence speed is slow, especially in high-order OAM modes, where it is prone to getting trapped in local optima, limiting its correction accuracy. It is also quite sensitive to the quality and noise of the intensity map.

[0004] The vortex wavefront correction method based on the SPGD (Stochastic Parallel Gradient Descent) algorithm is a sensorless wavefront correction method applicable to scenarios involving wavefront distortion caused by vortex light propagating in atmospheric turbulence. This method uses an objective function (such as the Steller ratio or intensity correlation coefficient of the far-field spot) as a performance metric. It estimates the gradient and updates the control input by introducing positive and negative random perturbations into the deformable mirror driving voltage and comparing the changes in spot quality before and after perturbations, thus achieving iterative wavefront correction. While the SPGD algorithm also requires no wavefront sensor, has a simple structure, and is suitable for complex optical fields, it suffers from drawbacks such as a high number of perturbations, slow convergence speed, poor stability, and susceptibility to local optima.

[0005] The vortex wavefront distortion correction method based on phase diversity (PD) is a wavefront correction technique without wavefront sensors, which indirectly infers the wavefront phase distribution from the image plane light intensity information using an image sensor. However, the phase difference method requires precise focus and defocus positioning during vortex wavefront phase difference correction, and dual-channel measurement complicates system alignment. Furthermore, phase unwrapping and deconvolution are prone to 2π jumps and non-unique solutions in ring / multi-ring / strong distortion scenarios, and insufficient focal plane signal can occur in the dark region at the vortex center or during multi-ring energy splitting, affecting inversion stability.

[0006] With the development of artificial intelligence, deep learning methods have been gradually introduced into vortex wavefront correction, a sensorless, end-to-end correction technique. This method, like traditional wavefront sensors, learns the mapping relationship between wavefront information and correction commands from vortex light intensity images using a neural network. However, deep learning-based vortex wavefront distortion correction methods rely heavily on a large number of "vortex distortion images" for training when performing vortex wavefront phase aberration correction. The "phase" mapping data is labeled, and the practicality and generalization ability of the simulated data after training are limited. In addition, the deep learning prediction lacks physical constraints, and the output is unreliable in strong distortion scenarios. Summary of the Invention

[0007] To address the aforementioned technical problems, this invention provides a model-based dual-beam link vortex light correction system, comprising:

[0008] A dual-beam co-path optical path is used to enable vortex light and planar light to propagate along the same path, traversing turbulent or optically distorted regions together, and carrying the same wavefront distortion information.

[0009] A wavefront correction unit is used to dynamically compensate for the wavefront distortion information;

[0010] The image acquisition and control unit is used to acquire far-field light spot images after distortion and correction of planar light. Based on the model method, it generates drive commands for the wavefront correction unit according to the far-field light spot information to achieve closed-loop control.

[0011] The topological charge detection module is used to detect the topological charge of the corrected vortex light.

[0012] Preferably, the wavefront correction unit includes an adjustable wavefront modulator and a drive control module corresponding to the adjustable wavefront modulator.

[0013] The drive control module is used to receive modulation control signals from the image acquisition and control unit, and after converting the modulation control signals into drive signals, control the wavefront modulator to dynamically adjust the incident wavefront.

[0014] Preferably, the image acquisition and control unit includes an image acquisition module, a model-based control module, and a drive module for driving the wavefront correction unit;

[0015] The image acquisition module is used to acquire far-field light spot images of planar light;

[0016] The model-based control module is used to generate drive commands for the wavefront correction unit based on the far-field spot information.

[0017] The drive controller of the wavefront correction unit is used to convert the drive command into a signal that the wavefront correction unit can execute.

[0018] Preferably, the image acquisition and control unit adopts physical modeling based on the model method, establishes a linear relationship between the light intensity distribution acquired by the image sensor and the wavefront distortion, and performs closed-loop control by applying basis function mode perturbation and performing T+1 spot measurements in the T-order basis function mode.

[0019] Preferably, the process of vortex light correction based on the model method in the image acquisition and control unit includes:

[0020] Generate preset parameters, including generating a unit circular pupil for far-field light intensity processing, generating a pupil with a size 0.99 times that of the unit circular pupil for gradient processing, and generating an Airy disk weight mask, defined as a far-field light spot correction matrix mask;

[0021] Select wavefront basis functions, calculate the gradient inner product for each pair of basis functions to obtain the second-order moment matrix, and calculate the pseudo-inverse of the second-order moment matrix;

[0022] Measure and obtain the influence function of the wavefront corrector, calculate the coupling matrix between the drivers, and the relationship matrix between the basis function modes and the wavefront corrector drivers;

[0023] Establish the relationship between the basis function coefficients and the driving voltage;

[0024] The image acquisition device is used to acquire the far-field spot image of the plane light wavefront to be corrected, and the region centered on the centroid of the spot is extracted.

[0025] Based on the single-order basis function mode, the control voltage of the wavefront corrector corresponding to the single-order basis function mode is calculated and applied to the wavefront corrector to generate a perturbation wavefront, which is superimposed on the current wavefront. The perturbation of the far-field spot image is acquired by the image acquisition device and the light intensity is calculated.

[0026] The difference between the far-field light intensity after superimposing single-order basis function modes and the light intensity corresponding to the wavefront under test is calculated to obtain the correspondence between light intensity changes and mode perturbations.

[0027] Based on the model formula, the basis function coefficients of the wavefront to be corrected are obtained;

[0028] The optimal control voltage of the corrected wavefront corrector is calculated based on the basis function coefficients and applied to the wavefront corrector for wavefront correction. This process is repeated iteratively until the preset termination condition is reached, thus completing the vortex light correction.

[0029] Preferably, the value range of the weight mask is 5 times that of the unit circular pupil.

[0030] Preferably, in the process of selecting wavefront basis functions, Zernike polynomials, KL modes, deformable mirror eigenmodes, or other orthogonal modes are used as basis functions and range normalization is performed.

[0031] Preferably, the iteration is repeated until a preset termination condition is reached:

[0032] Determine whether the similarity evaluation index between the corrected image and the reference image meets the preset evaluation index threshold. If it does, terminate the iteration process.

[0033] The similarity evaluation metrics include structural similarity, peak signal-to-noise ratio, and other image quality metrics.

[0034] Preferably, the wavefront corrector is a controllable optical element capable of dynamically modulating the wavefront of a light wave;

[0035] The wavefront corrector includes a deformable mirror, a liquid crystal spatial light modulator, and other devices with phase modulation functions.

[0036] Preferably, the model formula is established based on the approximate linear relationship between the second moment of the wavefront gradient and the corrected far-field intensity distribution.

[0037] Compared with the prior art, the present invention has the following advantages and technical effects:

[0038] This invention employs a co-path dual-wavelength linkage scheme, measuring only planar light with a different wavelength but the same optical path as the vortex light. Within the same model-based wavefront-free sensorless correction system framework, it achieves indirect correction of turbulent aberrations in co-path vortex light, thereby improving the efficiency and stability of the correction.

[0039] This invention eliminates the reliance on vortex light annular intensity maps for GS iteration, employing a simpler structure, planar light imaging feedback, a concentrated structure, and a clearly defined single target peak, thus avoiding insufficient constraints caused by annular dark areas. A common path ensures that planar light and vortex light carry the same source of turbulent aberrations; phase compensation obtained on the planar light is directly applied to the vortex light, avoiding independent corrections for complex modes. This avoids poor GS convergence and the tendency to get trapped in local optima, improving robustness.

[0040] This invention addresses the issue of slow convergence speed in the SPGD method by inversely calculating the gradient direction based on a model for wavefront-distorted planar light, solving for aberrations, and directly applying the solution to a deformable mirror. This significantly reduces the number of convergence steps, improves control stability, and solves the problem of slow convergence speed in the SPGD method.

[0041] This invention does not rely on vortex focal length and defocus intensity inversion; instead, it uses planar light wavefront detection, common path to ensure compensation consistency, dual-wavelength spectroscopic detection, common path dual-wavelength and single CCD channel imaging, and planar light for feedback, avoiding inversion deviation and unsolvable encapsulation problems caused by multi-channel measurement errors, achieving single-channel accurate correction, and making the system more simplified.

[0042] This invention requires no training set, relying solely on physical modeling and system calibration. It measures the wavefront error of the planar optical path, calculates it, and maps it to a wavefront corrector for compensation. Aberration calculation and compensation strictly adhere to physical constraints; no large-scale training dataset is needed. The system's common-path mechanism makes compensation independent of OAM order, suitable for multi-mode, multi-ring vortex optical correction, significantly reducing training costs and demonstrating strong adaptability.

[0043] This invention supports a general framework for vortex beam correction of arbitrary structures. The optical path structure and control model do not depend on the specific form of OAM, and it has good scalability. It can support adaptive correction of high-dimensional encoded optical fields such as perfect vortex beam POVB, multi-ring perfect vortex beam MR-POVB, and multi-mode superposition.

[0044] The modular decoupled system structure design of the present invention consists of a wavefront correction unit, an image acquisition and control unit, a model method control module, and a topological charge detection module. The structure is clear and easy to expand and integrate. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0046] Figure 1 This is a schematic diagram of the system structure according to an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the model-based dual-beam link vortex light correction algorithm according to an embodiment of the present invention;

[0048] Figure 3 This is a diagram illustrating the perfect vortex beam correction effect of a dual-beam link based on the model method in an embodiment of the present invention.

[0049] Figure 4 The image shows the distribution of indicators before and after correction of a dual-beam link vortex light with a turbulence intensity D / r0 = 15 for 100 frames, as shown in this embodiment of the invention. Detailed Implementation

[0050] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0051] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0052] like Figure 1 As shown, this embodiment provides a model-based dual-beam link vortex light correction system, including:

[0053] A dual-beam co-path optical path is used to enable vortex light and planar light to propagate along the same path, traversing turbulent or optically distorted regions together, and carrying the same wavefront distortion information.

[0054] The wavefront correction unit is used to dynamically compensate for wavefront distortion information.

[0055] The image acquisition and control unit is used to acquire far-field light spot images after distortion and correction of planar light. Based on the model method, it generates drive commands for the wavefront correction unit according to the far-field light spot information to achieve closed-loop control.

[0056] The topological charge detection module is used to detect the topological charge of the corrected vortex light.

[0057] Specifically, this embodiment proposes a novel structure and fast correction algorithm for vortex beam wavefront correction based on the approximate linear relationship between the second moment of the wavefront gradient and the corrected far-field intensity distribution. By constructing a dual-beam common-path link (a vortex beam and a plane beam carrying the same turbulence disturbance information), the phase distortion carried by the vortex beam is indirectly characterized by the change in the plane beam wavefront. Combined with a model-based wavefront estimation and driving control algorithm, effective, indirect, and wavefront-detector-free adaptive compensation of the vortex beam wavefront is achieved without directly measuring the wavefront. Figure 1 As shown.

[0058] The system employs two beams of light with different wavelengths but propagating along the same path. λ1 (e.g., 850nm) is a perfectly vortex beam carrying information and possessing multiple topological charges, while λ2 (e.g., 632nm) is a plane beam with a simple structure and stable intensity distribution. The vortex and plane beams are combined and propagate along the same path, traversing atmospheric turbulence and carrying the same wavefront distortion. The distorted beam is reflected by a wavefront corrector (e.g., a deformable mirror or a liquid crystal spatial light modulator), which performs wavefront correction. Unlike traditional methods, this system does not require measuring the wavefront of the vortex beam. Instead, it uses an image sensor to acquire a far-field image of the plane beam (without acquiring the vortex beam) to complete closed-loop control. The far-field image has high structural clarity and is easily identifiable. After the image information is transmitted to the control algorithm module, the model method generates a wavefront corrector drive signal based on the beam information. The vortex beam, after closed-loop correction, is separated by a filter and then enters the interferometry module for measuring the topological charges.

[0059] Furthermore, the system in this embodiment has two beams of light: a vortex beam and a planar beam. The former is mainly used to carry information and has a vortex phase structure, while the latter has a simple structure and a clear beam pattern, and is used as a wavefront correction reference. The two beams of light propagate along a common path, traversing turbulent or optically distorted regions together, thereby carrying the same wavefront distortion information.

[0060] This embodiment employs a dual-wavelength, dual-beam co-path propagation system, using dual light sources of different wavelengths (e.g., 850nm and 632nm) to achieve co-path propagation of perfect vortex light and plane light in space. This ensures they experience the same atmospheric turbulent wavefront disturbances, enabling indirect wavefront sensing and correction of the vortex light. This solves the problem of directly detecting the distorted wavefront of vortex light. By using plane light instead of vortex light for wavefront measurement, instead of directly acquiring the far-field spot of the vortex light, the high contrast and clear structure of the far-field spot of the plane light are utilized to indirectly infer the phase distortion of the vortex light through its wavefront changes, achieving sensorless measurement.

[0061] Furthermore, the wavefront correction unit in this example includes a 61-element deformable mirror and its corresponding drive control module;

[0062] The drive control module receives the modulation control signal from the image acquisition and control unit, converts the modulation control signal into a drive signal, and controls the deformable mirror to dynamically adjust the incident wavefront.

[0063] Furthermore, the image sensor in this example, typically a CCD or CMOS, only acquires far-field spot images of planar light after distortion and correction, and does not directly acquire vortex light.

[0064] Furthermore, the image acquisition and control unit includes an image acquisition module, a model-based control module, and a drive module for driving the wavefront correction unit;

[0065] The image acquisition module is used to acquire far-field spot images of planar light;

[0066] The model-based control module is used to generate drive commands for the wavefront correction unit based on far-field spot information;

[0067] The high-voltage drive controller of the deformable mirror is used to convert drive commands into signals that the deformable mirror can execute.

[0068] Furthermore, the image acquisition and control unit consists of an image acquisition module, a wavefront estimation algorithm module based on the model method, and a high-voltage drive controller (D / A) for the deformable mirror. It achieves closed-loop control by acquiring far-field light spot images of the planar beam and calculating and generating drive commands for the deformable mirror. The compensated beam is filtered to separate vortex light, which then enters the interferometry module to complete topological charge detection.

[0069] Furthermore, the image acquisition and control unit employs model-based physical modeling to establish a linear relationship between the light intensity distribution acquired by the image sensor and wavefront distortion. By applying mode perturbations (such as Zernike basis functions), wavefront correction is performed through T+1 spot measurements in T-order mode. Based on the measurement results, deformable mirror drive commands are generated to achieve closed-loop control.

[0070] Furthermore, such as Figure 2 As shown, the process of vortex light correction based on the model method in the image acquisition and control unit includes:

[0071] Generate preset parameters, including generating a unit circular pupil for far-field light intensity processing, generating a gradient pupil with 0.99 times the unit circular pupil for gradient processing, and generating an Airy disk weight mask, which is defined as a far-field spot correction matrix mask.

[0072] Select wavefront basis functions, calculate the gradient inner product for each pair of basis functions to obtain the second-order moment matrix, and calculate the pseudo-inverse of the second-order moment matrix;

[0073] Measure and obtain the influence function of the wavefront corrector, and calculate the coupling matrix C between the drivers. v And the relationship matrix C between the basis function modes and the wavefront corrector driver. z2v ;

[0074] The relationship between the basis function coefficients and the driving voltage is established by the following formula:

[0075]

[0076] Among them, Z i For i-th order Zernike coefficients.

[0077] The image of the far-field spot of the planar light wavefront to be corrected is acquired by an image acquisition device. The region centered on the centroid of the spot is cropped and I0 is obtained by mask weighted normalization.

[0078] Based on the single-order basis function modes (a total of N modes), the perturbation coefficient α (ranging from 0.1 to 0.5) is used to calculate the wavefront corrector control voltage corresponding to each single-order basis function mode. This voltage is then applied to the wavefront corrector to generate a perturbation wavefront, which is superimposed on the current wavefront. The perturbation is then captured by an image acquisition device to obtain a far-field spot image and calculate the light intensity I. i (i = 1, ..., T);

[0079] Calculate the difference Q between the far-field light intensity after superimposing single-order basis function modes and the light intensity corresponding to the wavefront to be measured. i (i=1,…T), thus obtaining the correspondence between light intensity changes and mode perturbations;

[0080]

[0081] Based on the model formula, the basis function coefficients Z of the wavefront to be corrected are calculated. i (where c0 is an adjustable parameter).

[0082] Z i =pinv_rr·c0·Q i / (2α) (i=1,…T) (3)

[0083] Based on the basis function coefficients Z i The optimal control voltage v of the corrected wavefront corrector is obtained by calculating using formula (1). * The wavefront correction is applied to the wavefront corrector and repeated iteratively until the preset termination condition is met, thus completing the vortex light correction.

[0084] The aforementioned model-based wavefront-free correction algorithm is a physical modeling-based adaptive optics correction method that does not rely on a wavefront sensor. Instead, it constructs a physical relationship model between imaging light intensity and wavefront distortion. Wavefront distortion information is extracted directly from multiple far-field spot images acquired by the image sensor, thereby achieving phase compensation. For example, in T-order mode aberration correction, the system only needs to perform T+1 far-field spot measurements. Compared with existing wavefront-free adaptive optics system control algorithms, the model-based wavefront-free adaptive optics system requires significantly fewer measurements.

[0085] This embodiment utilizes a model-based adaptive correction mechanism for wavefront-sensorless operation. Leveraging the linear relationship between the second moment of the wavefront gradient and the corrected far-field intensity distribution, a model-based dual-beam link vortex light correction method and system are designed, achieving wavefront distortion correction for vortex light without wavefront sensors. This avoids the slow iteration and poor convergence problems of traditional GS and SPGD methods. This method belongs to wavefront correction using physical modeling plus mode (T+1 times, where T is the number of aberration modes) perturbation measurements, resulting in stronger stability and fewer iterations.

[0086] This embodiment converts the wavefront correction result of planar light into driving commands for a wavefront corrector, achieving real-time closed-loop compensation for vortex light. After compensation, the beam is separated by a filter and enters an interferometer system to measure the topological charge of the vortex light, realizing the integration of compensation and detection. For beams with complex structures such as POVB and MR-POVB, which are difficult to detect using traditional wavefront methods, high-precision and fast compensation can be achieved without directly measuring their wavefront; suitable for dynamic communication systems.

[0087] Furthermore, the weight mask takes values ​​that are 5 times the unit circular pupil, and is used to weight the subsequently captured spot images to reduce noise and enhance the linear relationship with the second moment of the wavefront gradient.

[0088] Furthermore, in the process of selecting wavefront basis functions, Zernike polynomials are used as basis functions and range normalization is performed. Then, the gradient inner product is calculated for each pair of basis functions, that is, the first-order difference is performed in the x and y directions and integrated on the pupil to obtain the second-order moment matrix, and its pseudo-inverse is calculated.

[0089] Furthermore, the iterations are repeated until a preset termination condition is met:

[0090] Determine whether the similarity evaluation index between the corrected image and the reference image meets the preset evaluation index threshold. If it does, terminate the iteration process.

[0091] The similarity evaluation metrics include structural similarity, peak signal-to-noise ratio, and other image quality metrics.

[0092] Furthermore, the wavefront corrector is a controllable optical element capable of dynamically modulating the wavefront of a light wave;

[0093] Wavefront correctors include deformable mirrors, liquid crystal spatial light modulators, and other devices with phase modulation capabilities.

[0094] This embodiment addresses the problems of high difficulty, low contrast, and slow iteration efficiency in traditional wavefront measurements of vortex beams or complex spatial patterns (such as perfect vortex beam POVB and multi-ring MR-POVB). In this system, the model method does not directly apply to the vortex beam; instead, it calculates the wavefront distortion by acquiring far-field spot images using planar light propagating along a common path, and then controls a wavefront corrector for feedback compensation. Because planar light imaging is more concentrated and has higher contrast, and shares the same turbulent wavefront as the vortex beam, this embodiment can achieve effective correction of the vortex beam without measuring its complex structure, while balancing accuracy, speed, and ease of engineering implementation, making it highly suitable for dynamic communication systems.

[0095] In practical applications, the wavefront corrector can be a deformable mirror, with the number of control units typically being 37, 61, or 97. The following description uses a deformable mirror with 61 control units as an example, along with the accompanying drawings, to further illustrate the invention. In this embodiment, Zernike polynomials are used as basis functions, and two double perfect vortex light fields with different topological charges are used as the targets to be corrected. The topological charges of light field 1 are l1 = 2 and l2 = 4, while those of light field 2 are l1 = -2 and l2 = -4. In practice, the implementation is not limited to Zernike polynomials or perfect vortex light. The algorithm termination is not limited to the number of iterations, but rather uses the SSIM (structural similarity) between the corrected vortex light interference image and the ideal image as the objective function controlling the algorithm. The smaller the aberration, the larger the SSIM. When SSIM > 0.9, the iteration terminates. In practice, the objective function is not limited to SSIM. Specifically, the following steps are included:

[0096] Step 1: Preset, generate a unit circular pupil for far-field light intensity processing, generate a gradient pupil of 0.99 times the unit circular pupil for gradient processing, generate an Airy disk weight mask (value range is 5 times the unit circular pupil), and define it as the far-field spot correction matrix mask.

[0097] Step 2: Calculate the pseudo-inverse using Zernike polynomials as basis functions.

[0098] Step 3: Calculate the coupling matrix between the deformable mirror actuators, and the relationship matrix between the Zernike mode and the deformable mirror actuators.

[0099] Step 4: Establish the relationship between the Zernike coefficient and the driving voltage.

[0100] Step 5: The image acquisition device acquires the far-field spot image of the plane light wavefront to be corrected, extracts the region centered on the centroid of the spot, and performs mask weighted normalization.

[0101] Step 6: Multiply the single-order Zernike mode (T-order mode in total) by the perturbation coefficient in sequence, calculate the deformable mirror control voltage corresponding to the single-order Zernike mode, apply it to the deformable mirror to generate a perturbation wavefront, superimpose it on the current wavefront, and the image acquisition device acquires the far-field spot image after perturbation and calculates the light intensity.

[0102] Step 7: Calculate the difference between the far-field light intensity after superimposing the Zernike mode and the light intensity corresponding to the wavefront under test, and obtain the correspondence between light intensity change and mode perturbation.

[0103] Step 8: Calculate the Zernike coefficients of the wavefront to be corrected using the model formula. The model formula is based on the approximate linear relationship between the second moment of the wavefront gradient and the corrected far-field intensity distribution.

[0104] Step Nine: Apply the optimal control voltage of the deformable mirror obtained in Step Eight (based on Step Four) to the deformable mirror for wavefront correction. Then, perform Gaussian mode interferometry on the corrected vortex light. Repeat the steps until the preset termination condition is met, i.e., the SSIM of the vortex light interferogram and the ideal image is > 0.9. The correction is then complete, and the result is as follows: Figure 3 and Figure 4 As shown.

[0105] This embodiment supports a general framework for vortex optical correction of arbitrary structures. The optical path structure and control model do not depend on the specific form of OAM, and it has good scalability. It can support adaptive correction of high-dimensional encoded optical fields such as MR-POVB and multi-mode superposition.

[0106] The modular and decoupled system architecture design in this embodiment consists of a wavefront correction module, an image acquisition module, a model-based algorithm unit, and an electrical signal controller. The structure is clear and easy to expand and integrate.

[0107] In this embodiment, the vortex light and the plane light use different wavelengths, which facilitates optical separation and back-end measurement, and avoids interference between the signal light and the reference light at the detection end; the wavelength difference can be split by a filter.

[0108] In this embodiment, the wavefront corrector can be either a deformable mirror or a liquid crystal spatial light modulator.

[0109] In this embodiment, the basis function selection for the wavefront correction model method can use the eigenmodes of the deformable mirror, Zernike polynomials, or other orthogonal modes (KL mode, Fourier mode, Chebyshev polynomial mode, Legendre polynomial mode) as basis functions.

[0110] The mode perturbation strategy in this embodiment adopts a T-order mode, which requires only T+1 far-field spot measurements during aberration correction. T can be dynamically adjusted according to different turbulence intensities or wavefront residuals, thereby flexibly controlling accuracy and computation time.

[0111] In this embodiment, the image sensor extracts a rectangular region centered on the centroid of the light spot and multiplies it with a mask to retain only the main energy region of the far-field light spot of the planar light, thereby reducing the interference of background noise on wavefront estimation, reducing the amount of data, and speeding up the calculation.

[0112] As a supplementary embodiment, this embodiment also provides a model-based method for correcting vortex beams in a dual-beam link, characterized in that...

[0113] Specifically, the following steps are included:

[0114] Step 1: Define a set of basis functions to characterize wavefront aberrations, calculate the second-order moments of the gradients of the x and y components of each basis function, and inverse them, denoted as P.

[0115] Step 2: Measure and obtain the influence function of the wavefront corrector, and calculate the coupling matrix C between the drivers. v The control signal C for each wavefront corrector corresponding to each mode is obtained sequentially based on the relationship between each mode of the basis function and the influence function of the wavefront corrector. z2v Establish the relationship between the basis function coefficients and the driving voltage:

[0116]

[0117] Among them, Z i These are the coefficients of the i-th order basis functions.

[0118] Step 3: The image acquisition device acquires the far-field light intensity of the plane light wavefront to be corrected, and extracts an M×M region centered on the centroid of the light spot, which is then compared with the correction matrix M. ask After multiplying, calculate the sum of the light intensities and denote it as I0(x,y).

[0119] Step 4: Apply the control voltage of the wavefront corrector corresponding to each order basis function mode to the wavefront corrector to generate a disturbance wavefront, and denote the coefficient vector of each order basis function mode as α;

[0120] Step 5: The perturbation wavefronts of each order basis function mode are superimposed onto the current wavefront, focused onto the image sensor through a focusing lens, and their corresponding far-field spots are sequentially acquired. An M×M region is cropped centered on the centroid of the spot, and the sum of the image plane light intensities of the cropped region is calculated and denoted as I. T (x,y), where T is the order of the basis functions.

[0121] Step 6: Calculate I1(x,y), I2(x,y)...I T The difference between (x,y) and I0(x,y) yields the 1×T dimensional vector Q, representing the correspondence between light intensity variation and mode perturbation. i (i = 1, ..., T):

[0122]

[0123] The basis function coefficients Z of the wavefront to be corrected are obtained using the model method. i (where c0 is an adjustable parameter).

[0124] Z i =P·c0·Q i / (2α)(i=1,…T)

[0125] Step 7: According to the formula Obtain the wavefront corrector drive signal and apply it to the wavefront corrector for wavefront correction. Repeat steps three to six iteratively until the preset termination condition is reached to complete the vortex light correction.

[0126] Wherein, the correction matrix M ask Its function is to make I i It has an approximately linear relationship with the second moment of the wavefront gradient.

[0127] Basis functions include Zernike polynomials, KL polynomials, deformable mirror eigenmodes, and other basis functions characterizing wavefront aberrations.

[0128] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A model-based dual-beam link vortex beam correction system, characterized in that, include: A dual-beam co-path optical path is used to enable vortex light and planar light to propagate along the same path, traversing turbulent or optically distorted regions together, and carrying the same wavefront distortion information. A wavefront correction unit is used to dynamically compensate for the wavefront distortion information; The image acquisition and control unit is used to acquire images of the plane light passing through the distortion region and the corrected far-field light spot. Based on the model method, it generates drive commands for the wavefront correction unit according to the far-field light spot information to achieve closed-loop control. The topological charge detection module is used to detect the topological charge of the corrected vortex light.

2. The system according to claim 1, characterized in that, The wavefront correction unit includes an adjustable wavefront modulator and a drive control module corresponding to the adjustable wavefront modulator. The drive control module is used to receive the modulation control signal from the image acquisition and control unit, and after converting the modulation control signal into a drive signal, control the wavefront modulator to dynamically adjust the incident wavefront.

3. The system according to claim 1, characterized in that, The image acquisition and control unit includes an image acquisition module, a model-based control module, and a drive module for driving the wavefront correction unit. The image acquisition module is used to acquire far-field light spot images of planar light; The model-based control module is used to generate drive commands for the wavefront correction unit based on the far-field spot information. The drive controller of the wavefront correction unit is used to convert the drive command into a signal that the wavefront correction unit can execute.

4. The system according to claim 1, characterized in that, The image acquisition and control unit adopts physical modeling based on the model method, establishes a linear relationship between the light intensity distribution acquired by the image sensor and the wavefront distortion, and performs closed-loop control by applying basis function mode perturbation and performing T+1 spot measurements in the T-order basis function mode.

5. The system according to claim 1, characterized in that, The process of vortex light correction based on the model method in the image acquisition and control unit includes: Generate preset parameters, including generating a unit circular pupil for far-field light intensity processing, generating a gradient pupil with 0.99 times the unit circular pupil for gradient processing, and generating an Airy disk weight mask, which is defined as a far-field spot correction matrix mask. Select wavefront basis functions, calculate the gradient inner product for each pair of basis functions to obtain the second-order moment matrix, and calculate the pseudo-inverse of the second-order moment matrix; Measure and obtain the influence function of the wavefront corrector, calculate the coupling matrix between the drivers, and the relationship matrix between the basis function modes and the wavefront corrector drivers; Establish the relationship between the basis function coefficients and the driving voltage; The image acquisition device is used to acquire the far-field spot image of the plane light wavefront to be corrected, and the region centered on the centroid of the spot is extracted. Based on the single-order basis function mode, the control voltage of the wavefront corrector corresponding to the single-order basis function mode is calculated and applied to the wavefront corrector to generate a perturbation wavefront, which is superimposed on the current wavefront. The perturbation of the far-field spot image is acquired by the image acquisition device and the light intensity is calculated. The difference between the far-field light intensity after superimposing single-order basis function modes and the light intensity corresponding to the wavefront under test is calculated to obtain the correspondence between light intensity changes and mode perturbations. Based on the model formula, the basis function coefficients of the wavefront to be corrected are obtained; The optimal control voltage of the corrected wavefront corrector is calculated based on the basis function coefficients and applied to the wavefront corrector for wavefront correction. This process is repeated iteratively until the preset termination condition is reached, thus completing the vortex light correction.

6. The system according to claim 5, characterized in that, s The value range of the weight mask is 5 times that of the unit circular pupil.

7. The system according to claim 5, characterized in that, In the process of selecting wavefront basis functions, Zernike polynomials, KL modes, deformable mirror eigenmodes, or other orthogonal modes are used as basis functions and range normalization is performed.

8. The system according to claim 5, characterized in that, The iteration repeats until the preset termination condition is met: Determine whether the similarity evaluation index between the corrected image and the reference image meets the preset evaluation index threshold. If it does, terminate the iteration process. The similarity evaluation metrics include structural similarity, peak signal-to-noise ratio, and other image quality metrics.

9. The system according to claim 5, characterized in that, The wavefront corrector is a controllable optical element capable of dynamically modulating the wavefront of a light wave. The wavefront corrector includes a deformable mirror, a liquid crystal spatial light modulator, and other devices with phase modulation functions.

10. The system according to claim 5, characterized in that, The model formula is established based on the approximate linear relationship between the second moment of the wavefront gradient and the corrected far-field intensity distribution.

Images