Orbital angular momentum mode adaptive control method and system

Abstract

The application discloses a kind of track angular momentum mode adaptive regulation and control method and system, applied to track angular momentum mode regulation and control technical field, method includes by obtaining multiple light signals and processing as Gaussian beam array, collimation and polarization shaping are after, corresponding complex phase mask of target track angular momentum mode is applied to each Gaussian light beam, make Gaussian light beam transform into target track angular momentum mode and be coupled to the corresponding core of multicore ring core optical fiber;In receiving end, track angular momentum mode light field is demodulated as Gaussian type light spot by conjugate topological charge phase element and intensity data is collected;Based on intensity data, calculate comprehensive evaluation value, and with maximum evaluation value as target, construct optimization problem and carry out iterative optimization, finally obtain target complex phase mask, realize the efficient, stable and scalable track angular momentum mode adaptive regulation and control.

Description

Technical Field

[0001] This invention relates to the field of orbital angular momentum mode control technology, and in particular to an adaptive control method and system for orbital angular momentum modes. Background Technology

[0002] Spatial multiplexing and mode multiplexing technologies improve single-fiber transmission capacity by introducing multiple spatial and mode channels in the same optical fiber. Orbital Angular Momentum (OAM) modes, due to their helical phase structure and good orthogonality, can be used as mode multiplexing bases in ring-core fibers to build multi-core, multi-mode high-capacity optical transmission systems.

[0003] In a typical OAM mode multiplexing system, the transmitter converts a Gaussian beam from a single-mode fiber array or chip array into the target OAM mode and couples it to a ring-core fiber. The receiver then converts the target OAM mode back into a Gaussian beam using a phase element with conjugate topological charge or a spatial light modulator, so that it can be coupled into a single-mode fiber or a coherent receiver. To reduce inter-mode crosstalk, precise matching of beam position, beam size, phase leading edge, polarization state, and fiber core array geometry is required simultaneously.

[0004] Existing systems typically rely on human experience to perform multi-degree-of-freedom tuning, or perform one-time calibration of the spatial light modulator (SLM) using a static phase mask. Some solutions rely on interferometry, phase retrieval, or mode-by-mode, fiber-core calibration to estimate mode purity and phase error. For multi-core, multi-mode systems, the dimensionality of the parameters to be tuned increases rapidly with the number of fiber cores, modes, and optical components. Manual tuning is time-consuming, has poor reproducibility, and is difficult to maintain long-term stability under vibration, temperature drift, air disturbance, or device drift. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides an adaptive control method and system for orbital angular momentum modes, thereby resolving the issue of large errors in existing adaptive control methods for orbital angular momentum modes.

[0006] A first aspect of this invention provides an adaptive control method for orbital angular momentum modes, comprising: Multiple optical signals are acquired and processed to obtain multiple Gaussian beams. These Gaussian beams are arranged in a certain geometric pattern in space, with each Gaussian beam corresponding to a core position of the target multi-core ring fiber. Each Gaussian beam is collimated and polarized to obtain the corresponding shaped Gaussian beam; A composite phase mask corresponding to the target orbital angular momentum mode is applied to each shaped Gaussian beam to obtain the corresponding target OAM mode. The target OAM mode is then coupled to the corresponding core of the target multi-core ring fiber to obtain the OAM mode optical field corresponding to each core. The composite phase mask includes an OAM mode transformation phase and at least one wavefront compensation phase. The optical fields of each OAM mode are demodulated by a conjugate topological charge phase element and converted into Gaussian light spots, and the intensity data of each Gaussian light spot is obtained. Based on the intensity data of each Gaussian spot, the corresponding comprehensive evaluation value is calculated. If the comprehensive evaluation value is less than the preset evaluation value threshold, an optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is performed until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, the current composite phase mask is determined to be the target composite phase mask.

[0007] Furthermore, the wavefront compensation phase can be any one of the following: global quadratic phase, local quadratic phase, linear phase, aberration correction, etc.

[0008] Furthermore, based on the intensity data of each Gaussian spot, the corresponding comprehensive evaluation value is calculated, including: Set the target area at the corresponding position of each fiber core; Based on the intensity data within the target area, calculate the set of light spot quality evaluation indicators within the target area. The set of light spot quality evaluation indicators includes at least the integral intensity of the light spot quality evaluation value, the peak intensity, the weighted integral intensity, the ratio of the intensity of the target area to the intensity of the background area, and the normalized correlation coefficient between the target Gaussian template and the measured light spot. The comprehensive evaluation value is calculated using one or more of the spot quality evaluation indicators from the set of spot quality evaluation indicators.

[0009] Furthermore, the optimization iteration process is as follows: An optimization problem is constructed with the goal of maximizing the evaluation value, and the simulated annealing algorithm is used to solve it to obtain candidate parameters; A new composite phase mask is calculated based on the candidate parameters. The Gaussian beam after being shaped by the corresponding fiber core is re-phase-modulated, coupled, and demodulated by the conjugate topological charge phase element to obtain the updated Gaussian spot. A new comprehensive evaluation value is calculated based on the intensity data of the updated Gaussian spot. If the new comprehensive evaluation value is greater than the preset evaluation threshold, the new composite phase mask is determined as the target composite phase mask. If the new comprehensive evaluation value is less than the preset evaluation threshold, the optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is carried out until the target composite phase mask is obtained.

[0010] Furthermore, after calculating the new comprehensive evaluation value based on the updated intensity data of the Gaussian spot, it also includes: Calculate the difference between the new comprehensive evaluation value and the comprehensive evaluation value. If the difference is greater than the preset difference threshold, then accept the candidate parameter. If the difference is less than a preset difference threshold, then whether to accept the decision is determined based on a preset probability, whereby the preset probability is calculated using the following formula: In the formula, As a preset probability, For comprehensive evaluation, This is the new comprehensive evaluation value. This represents the current iteration number.

[0011] Furthermore, after determining that the current composite phase mask is the target composite phase mask, the process also includes: Select a target composite phase mask corresponding to a reference topology order in the same fiber core as the calibration mode, and continue to optimize and iterate to obtain the compensation parameters of the fiber core; When it is necessary to generate the target OAM mode of the target topology order of the fiber core, the OAM mode transformation phase in the target composite phase mask corresponding to the reference topology order is adjusted according to the compensation parameters to obtain the adjusted spiral phase term. The adjusted spiral phase term and other wavefront compensation phases in the target composite phase mask are calculated to obtain the target composite phase mask corresponding to the target topological order.

[0012] Another embodiment of the present invention provides an adaptive control system for orbital angular momentum modes, comprising: The acquisition module is used to acquire multiple optical signals and process each optical signal to obtain multiple Gaussian beams. The multiple Gaussian beams are arranged in a certain geometric arrangement in space, and each Gaussian beam corresponds to a core position of the target multi-core ring fiber. The collimation and polarization shaping module is used to collimate and polarize each Gaussian beam to obtain the corresponding shaped Gaussian beam. The control module is used to apply a composite phase mask corresponding to the target orbital angular momentum mode to each shaped Gaussian beam to obtain the corresponding target OAM mode, and to couple the target OAM mode to the corresponding core of the target multi-core ring fiber to obtain the OAM mode optical field corresponding to each core. The composite phase mask includes an OAM mode transformation phase and at least one wavefront compensation phase. The Gaussian spot conversion module is used to demodulate the optical fields of each OAM mode through a conjugate topological charge phase element, convert them into Gaussian spots, and acquire the intensity data of each Gaussian spot. The optimization module is used to calculate the corresponding comprehensive evaluation value based on the intensity data of each Gaussian spot. If the comprehensive evaluation value is less than the preset evaluation value threshold, the optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is performed until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, the current composite phase mask is determined to be the target composite phase mask.

[0013] Furthermore, the wavefront compensation phase can be any one of the following: global quadratic phase, local quadratic phase, linear phase, aberration correction, etc.

[0014] A third aspect of the present invention provides a computer device, comprising: Memory, used to store computer programs; A processor is used to implement the steps of an adaptive control method for orbital angular momentum modes, such as the first aspect, when executing a computer program.

[0015] A fourth aspect of the present invention provides a storage medium storing a computer program, which, when executed by a processor, implements the steps of the orbital angular momentum mode adaptive control method of the first aspect.

[0016] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following: By acquiring multiple optical signals and processing them into a Gaussian beam array, and after collimation and polarization shaping, a composite phase mask corresponding to the target orbital angular momentum mode is applied to each Gaussian beam, transforming the Gaussian beam into the target orbital angular momentum mode and coupling it to the corresponding core of a multi-core ring fiber. At the receiving end, the orbital angular momentum mode optical field is demodulated into a Gaussian spot using a conjugate topological charge phase element, and intensity data is acquired. Based on the intensity data, a comprehensive evaluation value is calculated, and an optimization problem is constructed with the goal of maximizing the evaluation value for iterative optimization. Finally, the optimal composite phase mask is obtained, realizing efficient, stable, and scalable adaptive control of the orbital angular momentum mode. Attached Figure Description

[0017] Figure 1 This is a flowchart of an adaptive control method for orbital angular momentum modes in one embodiment of the present invention; Figure 2 This is a diagram showing the overall structure of the control system for the adaptive control method of orbital angular momentum mode in one embodiment of the present invention. Figure 3 This is a schematic diagram of the composite phase mask parameterization of the adaptive control method for orbital angular momentum modes in one embodiment of the present invention; Figure 4 This is a flowchart of the adaptive mode control method based on intensity feedback in one embodiment of the orbital angular momentum mode adaptive control method of the present invention. Figure 5 This is a schematic diagram of OAM cross-order parameter migration and parallel control in an adaptive control method for orbital angular momentum modes according to one embodiment of the present invention. Figure 6 This is a system block diagram of an adaptive control system for orbital angular momentum modes in one embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0020] One embodiment of the present invention provides an adaptive control method for orbital angular momentum modes. For details, please refer to [link to relevant documentation]. Figure 1 , Figure 1 The flowchart shown is a method for adaptive control of orbital angular momentum modes according to one embodiment of the present invention, including steps S101 to S105, the specific steps of which are as follows: S101. Acquire multiple optical signals and process each optical signal to obtain multiple Gaussian beams. The multiple Gaussian beams are arranged in a certain geometric pattern in space, and each Gaussian beam corresponds to a core position of the target multi-core ring fiber.

[0021] In this embodiment, a light source, a modulator, and a single-mode fiber array output one or more Gaussian beams, each beam corresponding to a core or spatial channel of the target multi-core fiber. The light source in this embodiment is a device that generates optical signals. This embodiment does not limit the specific type of light source. For example, the light source in this embodiment can be a continuous-wave laser; or it can be a modulated data optical transmitter; or it can be a pilot light generator. The single-mode fiber array in this embodiment is used to distribute the optical signal output by the light source into multiple single-mode fibers, with each single-mode fiber outputting an independent Gaussian beam. The Gaussian beam in this embodiment has a transverse field distribution approximately equal to the fundamental mode Gaussian distribution. When propagating in free space, the beam spot is symmetrically distributed in a circle, with the highest intensity at the center and gradually decreasing towards the edges. The multiple Gaussian beams in this embodiment are arranged in a certain geometric pattern in space, such as a hexagonal array or a rectangular array, with each Gaussian beam corresponding to a core position of the target multi-core ring-core fiber.

[0022] This embodiment does not limit the specific method of generating the input beam array. For example, the input beam array can be generated by: a light source generating continuous light, which is split into multiple paths by a beam splitter and then enters each fiber of the single-mode fiber array, with each fiber outputting a Gaussian beam; or the input beam array can be generated by: multiple independent light sources generating optical signals, which are coupled to each fiber of the single-mode fiber array, with each fiber outputting a Gaussian beam.

[0023] S102. Collimate and polarize each Gaussian beam to obtain the corresponding shaped Gaussian beam.

[0024] In this embodiment, each input Gaussian beam is sequentially collimated and polarized. For example, as... Figure 2 As shown, each Gaussian beam emitted from the single-mode fiber array is collimated using a collimating lens group, making each beam approximately parallel to the others. This controls the beam divergence angle and satisfies the parallelism requirements of the subsequent phase modulation device. After collimation, each Gaussian beam maintains a spatial geometric arrangement corresponding to the cores of the target multi-core ring fiber. Then, the collimated Gaussian beams are sequentially polarized using polarization control elements, setting their polarization states to a predetermined state that matches the generation conditions of the target orbital angular momentum mode. After this collimation and polarization shaping process, each Gaussian beam is incident on the programmable phase modulator in a parallel state with the predetermined polarization state.

[0025] S103. Apply a composite phase mask corresponding to the target orbital angular momentum mode to each shaped Gaussian beam to obtain the corresponding target OAM mode, and couple the target OAM mode to the corresponding core of the target multi-core ring fiber to obtain the OAM mode optical field corresponding to each core. The composite phase mask includes an OAM mode transformation phase and at least one wavefront compensation phase.

[0026] In this embodiment, the controller generates an initial composite phase mask based on the target orbital angular momentum order, fiber core position, imaging magnification, and preset compensation parameters, and loads it onto a spatial light modulator or other programmable phase modulator. The input Gaussian beam, after passing through the phase modulator, forms a target orbital angular momentum mode or an array of orbital angular momentum modes, and is coupled to a ring-core fiber or target waveguide via the imaging optical path. In this embodiment, mode transformation refers to transforming the planar wavefront of the Gaussian beam into a wavefront with a helical phase structure, enabling the beam to carry orbital angular momentum. Coupling in this embodiment refers to projecting the phase-modulated OAM mode beam onto the end face of the target fiber, allowing it to enter the fiber core and propagate forward along the fiber.

[0027] This embodiment does not limit the specific implementation of the mode transformation. For example, the mode transformation can be implemented by loading a composite phase mask containing a spiral phase onto a spatial light modulator. After the Gaussian beam is reflected or transmitted through the spatial light modulator, its wavefront is modulated into a spiral phase distribution, thereby transforming it into an orbital angular momentum mode. This embodiment is preferably an OAM mode, but it can also be extended to ring vector mode, LP mode, Bessel mode, or other structured light modes if the target mode has a reversible phase transformation and its demodulated intensity can be used as a mode matching index. For the LP mode, due to its weak rotation invariance and cross-order migration capability, independent mode optimization or the addition of rotation angle parameters can be used.

[0028] It should be noted that in multi-wavelength WDM systems, the phase response of the SLM phase mask may differ for different wavelengths, and the lens and fiber interface may also have wavelength-dependent aberrations. This method can select one or more representative wavelengths for calibration and apply the optimized parameters to adjacent wavelengths; alternatively, different parameter sets can be saved for different bands, for example, saving a set of global focal length and local compensation parameters for the S, C, and L bands respectively. In bidirectional transmission systems, the forward and reverse optical paths may share some free-space components or may use independent coupling paths. The controller can separately acquire the feedback intensity of the forward and reverse target modes to establish two evaluation functions; if the two directions share the same SLM, a weighted objective function can be constructed to simultaneously consider the coupling efficiency of both directions for the same phase mask. When the system has entered the high-speed data transmission stage, the feedback signal can come from a low-power pilot channel, idle wavelength, monitoring wavelength, inter-frame training sequence, or service signal bypass. This can maintain the stability of the OAM mode interface without interrupting the main transmission.

[0029] In one specific embodiment, in the integrated photonics implementation, the programmable phase modulator consists of a two-dimensional grating coupler array and a phase shifter array. Each fiber core corresponds to a set of grating couplers or a programmable emission aperture. The controller generates an approximate OAM spiral phase and wavefront-compensated phase by adjusting the phase of each phase shifter within the aperture. The receiver can convert the target mode into the fundamental mode via an integrated mode demodulator or an off-chip VPP, and then the feedback is acquired by an on-chip detector or an external camera. Since the number of phase units in an on-chip phased array is typically less than the number of pixels in an SLM, the low-resolution phase control strategy of this invention can be directly applied. The controller optimizes low-dimensional parameters such as phase of the phased array units, array tilt, and array focal length, rather than a high-resolution two-dimensional phase map. This embodiment illustrates that the invention is not limited to laboratory free-space optical paths but can also be extended to engineered optical transceiver modules.

[0030] In some embodiments, the wavefront compensation phase is any one of global quadratic phase, local quadratic phase, linear phase, aberration correction, etc.

[0031] In this embodiment, in a multi-core OAM mode multiplexing scenario, the core spacing of the input single-mode fiber array is typically different from that of the target multi-core ring fiber, and there are assembly errors in rotation angle, end-face position, magnification, and focal plane position. This invention can establish a "common phase reference plane" using the imaging lens group and the SLM plane, ensuring that the position of the input Gaussian beam array on this plane corresponds to the projected position of each core of the target ring fiber. Let the coordinates of the c-th beam center of the input beam array on the SLM plane be... The target coordinates of the c-th fiber core on the same plane are: By selecting the lens focal length, object distance, image distance, and magnification, the two sets of coordinates are matched as closely as possible on an overall scale; the remaining error is compensated by linear phase terms, local quadratic phase terms, and local correction terms. For a 19-core structure, the axial alignment of the central core can be completed first, and then the inner and outer cores can be matched using a hexagonal / triangular lattice arrangement as a constraint. This scheme does not require each beam to be precisely aligned mechanically. As long as the input array and the target array reach a couplerable initial state, subsequent closed-loop optimization can compensate for lateral offsets, angular deviations, focal length deviations, and local aberrations within a certain range by adjusting the phase mask parameters. This feature makes the system suitable for large-scale multi-core, multi-mode parallel interfaces, eliminating the need for tedious manual adjustments for each core and each mode.

[0032] like Figure 3 As shown, Figure 3 The following is a schematic diagram of the parameterization of the composite phase mask. Its construction process is as follows: For example, for the c-th core region and the l-th order OAM mode, the composite phase mask loaded by the programmable phase modulator 105 can be represented as: in, This is the spiral phase term, i.e., the OAM mode transformation phase, used to convert the incident Gaussian beam into a topological charge. OAM mode; It is a global secondary phase or Fresnel lens phase, used to compensate for the wavefront curvature and imaging mismatch of the entire beam array; This is a local secondary phase for the c-th fiber core region, used to compensate for focal length deviations in a single beam or a single fiber core. This is a linear phase term used to compensate for lateral displacement, incident angle deviation, or beam pointing error; Local aberration correction terms may include partition constant phase, higher-order Zernike terms, piecewise polynomial phase, or pixel-level phase. This is a constant phase term.

[0033] It should be noted that not all of the above phase terms must be used simultaneously. In simpler systems, only the spiral phase term, the global quadratic phase term, and the linear phase term may be used; in large-scale multi-core systems or when significant local distortions exist, local quadratic phase terms and partition aberration terms may be added.

[0034] In multi-core systems, the effective surface of an SLM can be divided into multiple sub-apertures, each corresponding to a fiber core or an input beam. Each sub-aperture can be loaded with a helical phase of the same topological charge to form a multi-core array of the same OAM order, or with different topological charges to form a multi-mode parallel multiplexing array. For systems requiring simultaneous generation of ±1 OAM directions or dual polarization modes, this can be achieved through beam splitting, polarization combining, dual SLMs, dual-path systems, or on-chip multi-port phased arrays.

[0035] S104. Demodulate the optical fields of each OAM mode using a conjugate topological charge phase element, convert them into Gaussian light spots, and obtain the intensity data of each Gaussian light spot.

[0036] In this embodiment, the transmitted output light field is demodulated by a conjugate topological charge phase element, converting the light fields of each target OAM mode into Gaussian spots and forming an intensity distribution on the feedback detector. The conjugate topological charge phase element in this embodiment refers to an element loaded with a spiral phase opposite to the topological charge of the target orbital angular momentum mode. For example, when the target OAM mode is topological charge +1, the demodulation element is loaded with a spiral phase of -1; when the target OAM mode is -1, the demodulation element is loaded with a spiral phase of +1. If the transmitted light field matches the target mode, the conjugate phase cancels out the spiral phase, and the demodulated light field is concentrated into a fundamental mode or approximately fundamental mode Gaussian spot. If the transmitted light field contains non-target OAM orders, tilt, defocus, astigmatism, coma, or coupling distortion, the intensity of the target region decreases.

[0037] In this embodiment, demodulation refers to canceling the spiral wavefront of the orbital angular momentum mode using a conjugate spiral phase, restoring it to a plane wavefront, thereby forming a Gaussian spot after focusing. If the transmitted light field matches the target mode, the conjugate phase cancels the spiral phase, and the demodulated light field is concentrated into a fundamental or near-fundamental Gaussian spot, i.e., a spot with concentrated central energy and a circularly symmetrical distribution. If the transmitted light field contains non-target orbital angular momentum orders, tilt, defocus, astigmatism, coma, or coupling distortion, the intensity of the target region decreases, manifested as a darkening of the spot center, energy diffusion, or the appearance of side lobes. This embodiment does not limit the specific implementation of the conjugate topological charge phase element. For example, the conjugate topological charge phase element can be a vortex phase plate; or the conjugate topological charge phase element can be a spatial light modulator loaded with a conjugate spiral phase. This embodiment does not limit the specific method of acquiring the intensity data of each Gaussian spot. The feedback detector can acquire a complete two-dimensional light intensity image or only acquire the power value of a single photodetector.

[0038] It should be noted that this embodiment can use a CCD camera as a feedback detector, or it can use a CMOS camera, a four-quadrant detector, a single-photon detector array, a photodiode array, or an optical power meter connected to a single-mode fiber. As long as the detection measurement can reflect the degree of matching of the target mode after conjugate demodulation, it can be used as a feedback detector.

[0039] S105. Based on the intensity data of each Gaussian spot, calculate the corresponding comprehensive evaluation value. If the comprehensive evaluation value is less than the preset evaluation value threshold, construct an optimization problem with the goal of maximizing the comprehensive evaluation value and perform optimization iteration until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, determine the current composite phase mask as the target composite phase mask.

[0040] In this embodiment, the target region refers to the region of interest set at the position of the demodulated light spot corresponding to each fiber core on the image acquired by the feedback detector. After acquiring the image, the controller sets the target region at the corresponding position of each fiber core and calculates the intensity data in each region. If the intensity maximization form is adopted, the cost function can be defined as the negative of the evaluation function, transforming the problem into cost minimization, and then performing optimization iterations until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, the current composite phase mask is determined to be the target composite phase mask.

[0041] In one embodiment, based on the intensity data of each Gaussian spot, the corresponding comprehensive evaluation value is calculated, including: Set the target area at the corresponding position of each fiber core; Based on the intensity data within the target area, calculate the set of light spot quality evaluation indicators within the target area. The set of light spot quality evaluation indicators includes at least the integral intensity of the light spot quality evaluation value, the peak intensity, the weighted integral intensity, the ratio of the intensity of the target area to the intensity of the background area, and the normalized correlation coefficient between the target Gaussian template and the measured light spot. The comprehensive evaluation value is calculated using one or more of the spot quality evaluation indicators from the set of spot quality evaluation indicators.

[0042] In this embodiment, when using a camera, the controller can set a target region at the corresponding position of each fiber core. The evaluation function can take any of the following forms or a combination thereof: integral intensity within the target region, peak intensity within the target region, weighted integral intensity of the target region, ratio of target region intensity to background region intensity, normalized correlation coefficient between the target Gaussian template and the measured spot, or a comprehensive evaluation value considering penalties for adjacent fiber cores / adjacent modes. The formula for calculating the comprehensive evaluation value is: in, The weighting function can be Gaussian weight, circular uniform weight, or weight determined based on the mode field distribution of a single-mode fiber. This is the detection window corresponding to the c-th fiber core; To prevent division by zero of constants; Crosstalk estimation for adjacent ROIs or non-target patterns; This is an insertion loss penalty term; These are the weighting coefficients.

[0043] It should be noted that the aforementioned indicators such as integrated intensity, peak intensity, weighted integrated intensity, and the ratio of target area intensity to background area intensity can all be calculated using existing technologies that can achieve equivalent results. The following explanation only uses weighted integrated intensity as an example; the other indicators can be deduced by analogy from existing technologies: [Settings and parameters related to two-dimensional light intensity images are omitted as they are not relevant to the main text.] Weight functions of the same size The weighted integral intensity is obtained by multiplying the weight coefficients of each pixel within the template with the measured light intensity values ​​at the corresponding positions and then summing the results. The peak intensity is taken as the maximum gray value within the ROI, the integral intensity is taken as the sum of the gray values ​​within the ROI, and the background ratio is taken as the ratio of the average gray value within the ROI to the average gray value of the outer ring area of ​​the ROI. These details will not be elaborated further in this embodiment.

[0044] The comprehensive evaluation value calculation process proposed in this embodiment does not require reference light, interference fringe stabilization, complete phase recovery, or a high-speed coherent receiver for calibration. As long as the intensity of the demodulated light spot can reflect the degree of target mode matching, it can be used as feedback for the optimization algorithm. Therefore, this scheme can be executed automatically during the setup phase, system operation, or after environmental disturbances.

[0045] In one embodiment, the optimization iteration process is as follows: An optimization problem is constructed with the goal of maximizing the evaluation value, and the simulated annealing algorithm is used to solve it to obtain candidate parameters; A new composite phase mask is calculated based on the candidate parameters. The Gaussian beam after being shaped by the corresponding fiber core is re-phase-modulated, coupled, and demodulated by the conjugate topological charge phase element to obtain the updated Gaussian spot. A new comprehensive evaluation value is calculated based on the intensity data of the updated Gaussian spot. If the new comprehensive evaluation value is greater than the preset evaluation threshold, the new composite phase mask is determined as the target composite phase mask. If the new comprehensive evaluation value is less than the preset evaluation threshold, the optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is carried out until the target composite phase mask is obtained.

[0046] When this embodiment is being optimized and iterated, such as Figure 4 As shown, the controller initialization method has a set of parameters to be optimized. For the first... The parameter set can include: one fiber core and the first OAM order. (Global phase coefficient) (Global phase offset) (No. (Each fiber core's related phase coefficient) (Horizontal amplitude coefficient on the cross-section perpendicular to the fiber propagation direction (z-axis)) (Vertical amplitude coefficient on the cross-section perpendicular to the fiber propagation direction (z-axis)) (No. (horizontal displacement of each fiber core) (No. (vertical displacement of each fiber core) (No. (phase propagation constant of the m-th mode in each fiber core) (No. The fiber core along the light transmission direction The initial phase mask is generated by taking the position of each modulation point and other control variables related to the phase modulator hardware. This embodiment does not limit the specific method of generating the initial phase mask. For example, the initial phase mask can be generated by geometric initialization: calculating the initial phase mask based on the fiber core spacing, imaging magnification, lens focal length, and spatial light modulator pixel size; or the initial phase mask can be generated by historical initialization: reading the parameters saved during the last stable run as the initial value; or the initial phase mask can be generated by coarse scan initialization: scanning the focal length or linear phase over a large range, finding the detectable intensity peak, and then proceeding to fine optimization.

[0047] To ensure that the candidate phase mask does not exceed the hardware capabilities, the controller can set upper and lower limits and step sizes for each parameter. For example, the linear phase angle is limited to the coupling angle range, the local focal length is limited to the range allowed by the imaging system, and the partitioned phase is limited to the range of 0 to 2π. If an integrated phased array is used, constraints on the phase modulator voltage range, thermal crosstalk, response speed, and power consumption can also be added.

[0048] The optimization algorithm applies random or regular perturbations to the current parameter set to obtain candidate parameters θ', and generates candidate phase masks accordingly. Perturbations can be applied to global focal length, local focal length, linear phase, partitioned phase, or higher-order aberration parameters. The candidate phase masks are then loaded into the phase modulator. After hardware refresh and optical path stabilization, the feedback intensity is acquired again, and the candidate evaluation function or candidate cost is calculated. This process is repeated, performing several Markov chain searches at each temperature level, gradually reducing the acceptable range of random perturbations to transition from global search to local fine-grained search. When the termination condition is met, the optimal parameters are saved and the optimal phase mask is output. The termination condition can be: temperature below a threshold, no improvement after several consecutive rounds, evaluation function reaching a threshold, iteration count reaching an upper limit, or the system entering the service transmission phase.

[0049] In one specific embodiment, the system uses a 19-core single-mode fiber fan-out as the input beam array module. Nineteen Gaussian beams are collimated by lenses and then incident onto the SLM. The SLM plane is divided into 19 sub-apertures, each corresponding to a core position of the target 19-core ring-core fiber. For the target OAM order |l|=1, 2, or 3, the controller loads a spiral phase term on the corresponding sub-aperture and superimposes a global Fresnel lens phase, a local Fresnel lens phase, and a linear phase term. After SLM modulation, the beam is coupled to the input end of the 19-core ring-core fiber via a 4-f imaging system. The 4-f system maintains the relative position and optical axis relationship of the multi-beam array and projects the modulated OAM mode onto each ring-core fiber core. To achieve circular polarization or dual polarization operation, a quarter-wave plate, a half-wave plate, a polarization beam splitter, or a polarization beam combiner can be used before coupling. The OAM mode output from the target fiber core at the receiver is collimated and incident onto a vortex phase mask or SLM with conjugate topological charges, converting the target mode into an approximately Gaussian light spot. A CCD camera acquires this light spot, and the controller uses the intensity of the center region of interest (ROI) as the evaluation function to perform simulated annealing optimization on the phase parameters of the transmitter's SLM. After optimization, the optimal phase mask is used for data transmission of the corresponding fiber core and OAM order, or for subsequent cross-order migration.

[0050] In one embodiment, after calculating the new comprehensive evaluation value based on the updated intensity data of the Gaussian spot, the method further includes: Calculate the difference between the new comprehensive evaluation value and the comprehensive evaluation value. If the difference is greater than the preset difference threshold, then accept the candidate parameter. If the difference is less than a preset difference threshold, then whether to accept the decision is determined based on a preset probability, whereby the preset probability is calculated using the following formula: In the formula, As a preset probability, For comprehensive evaluation, This is the new comprehensive evaluation value. This represents the current iteration number.

[0051] In this embodiment, if a candidate parameter increases the target strength or reduces the cost, the candidate parameter is accepted; if the target strength decreases, a preset probability criterion is used to decide whether to accept it, in order to avoid getting trapped in local optima. The controller uses simulated annealing to solve the problem. Since the coupling efficiency of the OAM mode and the phase parameter are usually nonlinear, there are multiple local peaks; simulated annealing, by accepting poor candidate solutions with probability, can escape local optima in the early search stage and gradually converge after the temperature decreases. During the solution process, the controller first sets the initial temperature T0, the termination temperature Tmin, the temperature decay coefficient η, the Markov chain length L of each temperature level, the parameter perturbation range Δθ, and the maximum number of iterations Nmax. For the current parameters Calculate the current cost Ck. Generate candidate parameters θ' = θk + δ, where δ is a perturbation that is uniformly distributed, normally distributed, or adaptively adjusted based on historical sensitivity. Calculate C' after loading the candidate phase. If C' ≤ Ck, accept the candidate parameter θ'; if C' > Ck, then... The probability of accepting candidate parameters θ' is determined. After completing L candidate evaluations for a temperature level, let Tk+1 = ηTk, and use the best parameter within that level as the initial value for the next level. To reduce invalid calculations, the process can be terminated early if no new solution is accepted for several consecutive temperature levels, or if the fluctuation of the evaluation function is less than a preset threshold.

[0052] As an example of this embodiment, the initial temperature can be set to 100, the termination temperature can be set to 0.01, the Markov chain length can be set to 5, the initial values ​​of the global and local secondary phase focal lengths can be set according to the optical path design, for example, 206 mm, the focal length perturbation range can be ±5 mm, the initial value of the linear phase angle can be 0, and the perturbation range can be ±0.01 rad. Termination occurs early when no new solution is received for five consecutive temperature levels. The above values ​​are only examples and can be adjusted according to the optical path scale, SLM pixel size, camera noise, and feedback refresh rate.

[0053] In other embodiments, simulated annealing can also be combined with coarse scanning, hill climbing, stochastic parallel gradient descent, particle swarm optimization, genetic algorithms, Bayesian optimization, or reinforcement learning. For example, coarse scanning can be used to determine the approximate range of focal length and linear phase, and then simulated annealing can be used to optimize the local partition phase; or each fiber core can be annealed independently first, and then multi-core collaborative fine-tuning can be performed using a global uniformity index.

[0054] In one embodiment, after determining that the current composite phase mask is the target composite phase mask, the method further includes: Select a target composite phase mask corresponding to a reference topology order in the same fiber core as the calibration mode, and continue to optimize and iterate to obtain the compensation parameters of the fiber core; When it is necessary to generate the target OAM mode of the target topology order of the fiber core, the OAM mode transformation phase in the target composite phase mask corresponding to the reference topology order is adjusted according to the compensation parameters to obtain the adjusted spiral phase term. The adjusted spiral phase term and other wavefront compensation phases in the target composite phase mask are calculated to obtain the target composite phase mask corresponding to the target topological order.

[0055] In this embodiment, since the OAM modes have similar ring intensity distributions in the ring-core fiber, the main difference between different topological orders lies in the helical phase in the azimuth direction. Therefore, many mismatches caused by the optical system, such as lens aberrations, end-face tilt, local defocus, core position errors, and non-ideal wavefront curvature, have similar effects on different OAM orders within the same core. Based on this characteristic, a cross-order parameter transfer method is proposed. Figure 5 As shown, the specific implementation process of the cross-order parameter transfer method is as follows: select a reference topology order in the same fiber core. As the calibration mode, the above closed-loop optimization is performed to obtain the compensation parameters for the fiber core. When it is necessary to generate other topological orders When using a phase mask, only the spiral phase term is changed from... Replace with At the same time, retain The system includes global secondary phase, local secondary phase, linear phase, and local aberration correction parameters. After migration, the system can be directly used for service transmission or undergo minor fine-tuning. Fine-tuning can be limited to a small perturbation range, optimizing only the linear phase or local phase to compensate for residual mode field differences between different topological orders. Using this method, each fiber core no longer needs to be fully optimized for all OAM orders individually; instead, it can be replaced by "full optimization of a reference order + migration / small-range fine-tuning of other order parameters," significantly reducing calibration time and control resources. A similar approach can be used for l=0 mode, ±l mode, or dual-polarization mode. For OAM modes with opposite chirality (±l), if the optical path non-reciprocity is weak, compensation parameters other than the spiral phase sign can be shared; if there is a significant polarization-dependent error, a small number of independent correction parameters can be reserved for different polarizations or different chirities.

[0056] Furthermore, in a multi-core ring-core fiber optic system, different fiber cores form demodulated light spots at different positions on the feedback detector. The controller can set multiple ROIs (1 to 19) in the same camera image and calculate the evaluation functions J1 to J19 for each fiber core separately. Since the optical path sub-aperture and phase parameters corresponding to each fiber core are relatively independent, the system can perform optimization on multiple fiber cores in parallel. Parallelism methods include, but are not limited to: first, multi-threaded parallelism, where each thread corresponds to the parameter update of one fiber core; second, grouped parallelism, where fiber cores that are spatially far apart or have low crosstalk are grouped together for simultaneous optimization; third, time-division multiplexing, where the system quickly polls the fiber cores in sequence when the camera frame rate is limited; and fourth, master-slave control, where the central fiber core is optimized first, and then some global parameters are synchronized to the peripheral fiber cores. To avoid feedback confusion caused by simultaneous perturbation of multiple fiber cores, pilot coding, time-division perturbation, or orthogonal perturbation can be used. Pilot coding involves applying minute phase perturbations of different frequencies or time windows to different fiber cores and decoding the corresponding intensity changes in the feedback signal. Time-division perturbation updates only one fiber core or a group of fiber cores at a time. Orthogonal perturbation makes the perturbation vectors of different fiber cores independent, thus separating their respective contributions through correlation calculations. When the system requires uniform coupling efficiency across all channels, uniformity constraints can be added to the evaluation function, such as maximizing the average ROI intensity of all fiber cores while penalizing excessively low-intensity fiber cores or excessively large intensity variances between fiber cores. This avoids over-optimization of a few fiber cores while underperforming others.

[0057] In addition, this method can be used not only for initial system setup but also for online stabilization during operation. Since temperature drift, mechanical vibration, air disturbance, device aging, or micro-displacement of the fiber endface may cause mode mismatch, the controller can periodically or event-triggeredly acquire feedback intensity and initiate small-scale re-optimization when the intensity decreases. One online control process is as follows: a low duty cycle pilot light or monitoring frame is inserted into the service signal; the receiver acquires the target ROI intensity within a time window that does not affect the main service signal; the controller compares the current intensity with the historical best intensity or a threshold; if the decrease exceeds the threshold, local simulated annealing or random parallel gradient descent is performed near the current optimal parameters; after optimization, the new parameters are saved and the monitoring state is returned. Another online control process is bypass monitoring: a small portion of the output light is split to a feedback detector for continuous monitoring of the demodulated spot intensity; the main signal still enters the communication receiver or single-mode coupling port. This method is suitable for high-capacity transmission systems that require continuously maintaining low crosstalk. To reduce the impact of control disturbances on service transmission, online re-optimization can limit the amplitude and frequency of parameter disturbances and prioritize adjusting phase terms that have less impact on the communication signal. For example, the linear phase and local focal length are adjusted first, and the higher-dimensional local partition phase is only re-optimized when the intensity is below the threshold for a long period of time.

[0058] This invention also provides an adaptive control system for orbital angular momentum modes, which is implemented using the adaptive control method for orbital angular momentum modes described in this embodiment. Figure 6 A flowchart of an adaptive control system 200 for orbital angular momentum modes according to an embodiment of the present invention is shown, including: The acquisition module 201 is used to acquire multiple optical signals and process each optical signal to obtain multiple Gaussian beams. The multiple Gaussian beams are arranged in a certain geometric arrangement in space, and each Gaussian beam corresponds to a core position of the target multi-core ring fiber. The collimation and polarization shaping module 202 is used to collimate and polarize each Gaussian beam to obtain the corresponding shaped Gaussian beam. The control module 203 is used to apply a composite phase mask corresponding to the target orbital angular momentum mode to each shaped Gaussian beam to obtain the corresponding target OAM mode, and to couple the target OAM mode to the corresponding core of the target multi-core ring fiber to obtain the OAM mode optical field corresponding to each core. The composite phase mask includes an OAM mode transformation phase and at least one wavefront compensation phase. The Gaussian spot conversion module 204 is used to demodulate the optical fields of each OAM mode through a conjugate topological charge phase element, convert them into Gaussian spots, and acquire the intensity data of each Gaussian spot. The optimization module 205 is used to calculate the corresponding comprehensive evaluation value based on the intensity data of each Gaussian spot. If the comprehensive evaluation value is less than the preset evaluation value threshold, the optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is performed until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, the current composite phase mask is determined to be the target composite phase mask.

[0059] The specific implementation of the adaptive control system for orbital angular momentum mode is basically the same as the specific implementation of the adaptive control method for orbital angular momentum mode described above, and will not be repeated here.

[0060] In one embodiment, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0061] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., SSD), etc.

[0062] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed, it can include the processes of the embodiments of the above methods.

[0063] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this invention, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. An adaptive control method for orbital angular momentum modes, characterized in that, include: Multiple optical signals are acquired and processed to obtain multiple Gaussian beams, wherein the multiple Gaussian beams are arranged in a certain geometric arrangement in space, and each Gaussian beam corresponds to a core position of the target multi-core ring fiber. Each of the Gaussian beams is collimated and polarized to obtain the corresponding shaped Gaussian beam; A composite phase mask corresponding to the target orbital angular momentum mode is applied to each of the shaped Gaussian beams to obtain the corresponding target OAM mode, and the target OAM mode is coupled to the corresponding core of the target multi-core ring fiber to obtain the OAM mode optical field corresponding to each core. The composite phase mask includes an OAM mode transformation phase and at least one wavefront compensation phase. The optical fields of each OAM mode are demodulated by a conjugate topological charge-phase element and converted into Gaussian light spots, and the intensity data of each Gaussian light spot are obtained. Based on the intensity data of each Gaussian spot, the corresponding comprehensive evaluation value is calculated. If the comprehensive evaluation value is less than the preset evaluation value threshold, an optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is performed until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, the current composite phase mask is determined to be the target composite phase mask.

2. The adaptive control method for orbital angular momentum modes as described in claim 1, characterized in that, The wavefront compensation phase can be any one of global quadratic phase, local quadratic phase, linear phase, aberration correction, etc.

3. The adaptive control method for orbital angular momentum modes as described in claim 1, characterized in that, The calculation of the corresponding comprehensive evaluation value based on the intensity data of each Gaussian spot includes: Target areas are set at corresponding positions of each of the fiber cores; Based on the intensity data within the target area, a set of light spot quality evaluation indicators is calculated within the target area. The set of light spot quality evaluation indicators includes at least the integral intensity of the light spot quality evaluation value, the peak intensity, the weighted integral intensity, the ratio of the intensity of the target area to the intensity of the background area, and the normalized correlation coefficient between the target Gaussian template and the measured light spot. A comprehensive evaluation value is calculated using one or more of the light spot quality evaluation indicators from the set of light spot quality evaluation indicators.

4. The adaptive control method for orbital angular momentum modes as described in claim 1, characterized in that, The optimization iteration process described therein is as follows: An optimization problem is constructed with the goal of maximizing the evaluation value, and the simulated annealing algorithm is used to solve it to obtain candidate parameters; A new composite phase mask is calculated based on the candidate parameters. The new composite phase mask is then used to re-modulate, couple, and demodulate the shaped Gaussian beam of the corresponding fiber core using the phase mask, resulting in an updated Gaussian spot. A new comprehensive evaluation value is calculated based on the intensity data of the updated Gaussian spot. If the new comprehensive evaluation value is greater than the preset evaluation threshold, the new composite phase mask is determined to be the target composite phase mask. If the new comprehensive evaluation value is less than the preset evaluation threshold, then continue to construct an optimization problem with the goal of maximizing the comprehensive evaluation value and perform optimization iterations until the target composite phase mask is obtained.

5. The adaptive control method for orbital angular momentum modes as described in claim 4, characterized in that, After calculating the new comprehensive evaluation value based on the intensity data of the updated Gaussian spot, the method further includes: Calculate the difference between the new comprehensive evaluation value and the comprehensive evaluation value. If the difference is greater than a preset difference threshold, then accept the candidate parameter. If the difference is less than a preset difference threshold, then whether to accept the decision is determined based on a preset probability, wherein the preset probability is calculated using the following formula: In the formula, As a preset probability, The comprehensive evaluation value is... This is the new comprehensive evaluation value. This represents the current iteration number.

6. The adaptive control method for orbital angular momentum modes as described in claim 1, characterized in that, Once the current composite phase mask is determined to be the target composite phase mask, the following steps are also included: Select a target composite phase mask corresponding to a reference topology order in the same fiber core as the calibration mode, and continue to perform optimization iterations to obtain the compensation parameters of the fiber core; When it is necessary to generate the target OAM mode of the target topology order of the fiber core, the OAM mode transformation phase in the target composite phase mask corresponding to the reference topology order is adjusted according to the compensation parameter to obtain the adjusted spiral phase term. The adjusted spiral phase term and other wavefront compensation phases in the target composite phase mask are calculated to obtain the target composite phase mask corresponding to the target topological order.

7. An adaptive control system for orbital angular momentum modes, characterized in that, include: The acquisition module is used to acquire multiple optical signals and process each optical signal to obtain multiple Gaussian beams, wherein the multiple Gaussian beams are arranged in a certain geometric arrangement in space, and each Gaussian beam corresponds to a core position of the target multi-core ring fiber. The collimation and polarization shaping module is used to collimate and polarize each of the Gaussian beams to obtain the corresponding shaped Gaussian beams. The control module is used to apply a composite phase mask corresponding to the target orbital angular momentum mode to each of the shaped Gaussian beams to obtain the corresponding target OAM mode, and to couple the target OAM mode to the corresponding core of the target multi-core ring fiber to obtain the OAM mode optical field corresponding to each core. The composite phase mask includes an OAM mode transformation phase and at least one wavefront compensation phase. The Gaussian spot conversion module is used to demodulate the optical fields of each OAM mode through a conjugate topological charge phase element, convert them into Gaussian spots, and acquire the intensity data of each Gaussian spot. The optimization module is used to calculate the corresponding comprehensive evaluation value based on the intensity data of each Gaussian spot. If the comprehensive evaluation value is less than the preset evaluation value threshold, an optimization problem is constructed with the goal of maximizing the comprehensive evaluation value and optimization iteration is performed until the target composite phase mask is obtained. If the comprehensive evaluation value is greater than the preset evaluation value threshold, the current composite phase mask is determined to be the target composite phase mask.

8. The adaptive control method for orbital angular momentum modes as described in claim 7, characterized in that, The wavefront compensation phase can be any one of global quadratic phase, local quadratic phase, linear phase, aberration correction, etc.

9. A computer device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the adaptive control method for orbital angular momentum modes as described in any one of claims 1 to 7 when executing the computer program.

10. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the adaptive control method for orbital angular momentum modes as described in any one of claims 1 to 7.

Images