Optical calculation error adjusting method and system, medium and program product

Abstract

The invention discloses an optical calculation error adjustment method and system, a medium and a program product, and relates to the technical field of optical calculation, and the method comprises the steps: determining a unified correction deviation value corresponding to the phase noise of an interferometer array; a target function based on a correction deviation direction variable is generated with the minimum difference sum of squares between a measured light intensity value of the photoelectric detector array and an ideal light intensity value as a target; solving the objective function to obtain the optimal value of the correction deviation direction variable of each phase shifter; and sending the correction deviation direction variable optimal value and the unified correction deviation value of each phase shifter to control equipment, so that the control equipment adjusts each phase shifter at the same time. In this way, collective adjustment of all the phase shifters can be realized through the determined unified correction deviation value and in combination with the correction deviation direction variable optimal value of each phase shifter, extra overhead caused by independent calibration is avoided, complexity of light calculation noise mitigation is reduced, a noise mitigation process is accelerated, and output precision of light calculation equipment is improved.

Description

Technical Field

[0001] This invention relates to the field of optical computing technology, and in particular to an optical computing error adjustment method, system, medium, and program product. Background Technology

[0002] As semiconductor technology approaches its physical limits and Moore's Law gradually becomes ineffective, traditional electrical computers face numerous limitations: power consumption increases exponentially with frequency, copper interconnect bandwidth bottlenecks lead to data congestion, and actual computing power is difficult to achieve. Optical computing has become an important direction supporting the continuous evolution of computing power, with interferometer arrays being the core component for achieving linear operations in optical computing devices. However, during operation, optical computing devices typically require frequent individual calibration of each phase shifter in the interferometer array to lock the operating point, increasing system latency, and the output accuracy of optical computing devices is affected by the increase in array depth. Summary of the Invention

[0003] This invention provides a method, system, medium, and program product for adjusting optical computing errors, which can achieve collective adjustment of all phase shifters instead of calibrating each phase shifter individually, thereby accelerating the noise mitigation process and improving the output accuracy of optical computing devices.

[0004] This invention provides a method for adjusting optical computing errors, comprising: Determine the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device; the interferometer array includes at least two phase shifters; With the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device, an objective function based on the correction deviation direction variable is generated. Solving the objective function yields the optimal values ​​for the correction deviation direction variables of each phase shifter; The optimal value of the correction deviation direction variable of each phase shifter and the unified correction deviation value are sent to the control device so that the control device can adjust each phase shifter simultaneously according to the optimal value of the correction deviation direction variable of each phase shifter and the unified correction deviation value.

[0005] The present invention also provides an optical computing error adjustment system, comprising: an optical computing device, a control device, and an external computing device; The optical computing device includes a light source, an interferometer array, and a photodetector array; the interferometer array includes at least two phase shifters. The external computing device is used to determine the unified correction deviation value corresponding to the phase noise of the interferometer array; with the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device, an objective function based on the correction deviation direction variable is generated; the objective function is solved to obtain the optimal value of the correction deviation direction variable of each phase shifter; and the optimal value of the correction deviation direction variable of each phase shifter is sent to the control device. The control device is used to simultaneously adjust each phase shifter based on the optimal value of the correction deviation direction variable of each phase shifter and the unified correction deviation value.

[0006] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of any of the above-described optical computing error adjustment methods.

[0007] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any of the above-described optical computing error adjustment methods.

[0008] This invention first determines the unified correction deviation value corresponding to the phase noise of the interferometer array in an optical computing device; the interferometer array includes at least two phase shifters; then, with the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device, an objective function based on the correction deviation direction variable is generated; subsequently, the objective function is solved to obtain the optimal value of the correction deviation direction variable for each phase shifter; finally, the optimal value of the correction deviation direction variable for each phase shifter and the unified correction deviation value are sent to the control device so that the control device can simultaneously adjust each phase shifter according to the optimal value of the correction deviation direction variable for each phase shifter and the unified correction deviation value. By determining a unified correction deviation value and combining it with the optimal value of the correction deviation direction variable for each phase shifter, collective adjustment of all phase shifters can be achieved. This makes the measured light intensity value of the photodetector array close to the ideal light intensity value, effectively improving the output accuracy of the optical computing device and avoiding the additional overhead of individually calibrating the phase shifters. This significantly reduces the overall overhead of the entire interferometer array during operation. At the same time, the collective adjustment mode eliminates the need to debug each phase shifter individually, greatly simplifying the operation process of optical computing noise mitigation, reducing the complexity of optical computing noise mitigation, accelerating the noise mitigation process, and reducing the time required for calibration. It can be widely used in integrated optical networks, providing strong support for the efficient and low-power operation of integrated optical networks.

[0009] In addition, the present invention also provides a corresponding optical computing error adjustment system, computer-readable storage medium and program product for the optical computing error adjustment method, which has the same or corresponding technical features as the optical computing error adjustment method mentioned above, and has the same effect. Attached Figure Description

[0010] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A flowchart of the optical computing error adjustment method provided in an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a single Mach-Zehnder interferometer provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a Mach-Zehnder interferometer array provided in an embodiment of the present invention. Detailed Implementation

[0012] As semiconductor technology approaches its physical limits, the tunneling effect is bringing transistor miniaturization to a standstill, and Moore's Law is destined to fail. More seriously, the energy consumption crisis of traditional electrical computers is becoming increasingly prominent. Joule heating caused by resistance increases exponentially with frequency, and heat dissipation has become a constraint on computing power development. This high-energy-consumption model is unsustainable for sustainable development. Simultaneously, the bandwidth bottleneck of copper interconnects severely restricts data flow, and the "data congestion" of traditional computers means that actual computing power is far below theoretical peaks. Meanwhile, the demand for computing power (especially parallel matrix operations) in fields such as large-scale artificial intelligence model training, climate simulation, and real-time big data processing is exploding at a rate far exceeding the supply capacity of traditional computers. Traditional architectures can no longer support the computing power demands of the future digital society. Optical computing is considered to be able to solve the problems faced by electronic computing systems to some extent. Optical computing refers to using the physical properties of light to perform a series of computational tasks. Its core breakthrough lies in using light to replace electrons for information transmission and processing: light travels at a speed of 300,000 kilometers per second in a vacuum, breaking free from the constraints of resistance and capacitance. Theoretically, it can achieve terahertz-level bandwidth and femtosecond-level latency, completely overcoming the data transmission bottleneck that restricts computing power. More importantly, photonic transmission generates almost no Joule heat, and its energy efficiency is several orders of magnitude higher than that of electrical computing in certain computing tasks. This promises to reduce data center energy consumption by more than tenfold, fundamentally solving the energy consumption dilemma. Light also possesses a natural advantage in parallelism: light of different wavelengths can be transmitted independently in the same channel (wavelength division multiplexing), making the transmission capacity of a single optical fiber far exceed that of copper cables by tens of thousands of times. This characteristic has disruptive value in the field of artificial intelligence: light can directly perform matrix multiplication in the optical domain through physical processes such as interference and diffraction, and matrix operations are the core of neural networks.

[0013] In optical computing architectures, the Mach-Zehnder interferometer (MZI) array is a core component for achieving linear computation, and its performance is extremely sensitive to noise. The presence of noise fundamentally undermines the accuracy, reliability, and potential advantages of optical computing. Noise directly disrupts the interference process upon which the Mach-Zehnder interferometer relies. For example, fluctuations in ambient temperature can cause slight changes in the refractive index of waveguide materials, leading to phase drift that is difficult to track and compensate for in real time; minute mechanical stresses or vibrations can cause random fluctuations in the optical path difference. These phase noises, superimposed on the phase noise of the light source itself (the finite linewidth of the laser), blur the optical field state that originally required precise constructive or destructive interference. Inherent non-uniformities in the manufacturing process (such as etching depth deviations) also introduce static beam splitting ratio errors. The direct consequence is a catastrophic decrease in computational accuracy. When optical neural networks perform large-scale matrix multiplications, each tiny phase or amplitude error is amplified by the cascaded multilayer structure, resulting in a huge deviation between the final output and the theoretical value, severely affecting the accuracy of applications such as image recognition and signal processing. Furthermore, noise forces the system to bear a huge additional burden. To combat dynamic drift, frequent and complex online calibrations are necessary to relock the operating point, significantly increasing system latency and power consumption. Maintaining an acceptable signal-to-noise ratio even requires increasing input optical power or error correction overhead, undermining the inherent low-power advantage of optical computing. Noise also limits the size of crosstalk-free cells that can be used simultaneously, hindering the realization of large-scale parallel computing. Essentially, noise erodes the precision interferometric foundation upon which optical computing relies; if not effectively suppressed, its theoretical speed and energy efficiency advantages will be difficult to translate into reliable performance in practical applications. Although the impact of noise is small for a single phase shifter, its influence gradually increases with the depth of the Mach-Zehnder interferometer array, affecting the accuracy of the final output. To address these issues, this invention provides an optical computing error adjustment method.

[0014] 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. 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 protection scope of the present invention.

[0015] It should be noted that, in the description of this invention, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., used in this invention are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0016] To enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0017] The specific application environment architecture or specific hardware architecture on which the optical computing error adjustment method depends is described here.

[0018] The embodiments of the present invention provide a method for adjusting optical computing errors, and the method is described in detail below in conjunction with the execution flow of the optical computing error adjustment method. Figure 1 A flowchart of the optical computing error adjustment method provided in the embodiments of the present invention is shown below. Figure 1 As shown, the method includes: S101. Determine the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device; the interferometer array includes at least two phase shifters.

[0019] Before executing step S101, initialization settings can be performed first. Start the optical computing device, input a test light signal through the control device, and load a simple vector (such as an all-1 vector) through the spatial light modulator to ensure that all phase shifters are in the preset state (phase) of the control device. , (Set to known default values); connect to the photodetector array, measure the light intensity at the output port, and return the result to the control device. The control device can control the optical computing device to emit a preset light signal, control the selection of the phase shifter, and control the corresponding phase setting. The execution entity of the optical computing error adjustment method provided by this invention can be an external computing device, which is responsible for the calculation task. The aforementioned interferometer array can be a Mach-Zehnder interferometer array or other interferometer arrays, which are not limited here.

[0020] It should be noted that the core idea of ​​this invention is not to perfectly correct the error (i.e., noise) of each phase shifter, but to reduce the error range of the phase shifters, thereby reducing the overall error level of the interferometer array. Since training tasks such as optical neural networks can tolerate a certain degree of error, the optical computation error adjustment method provided by this invention can increase the noise robustness of optical neural networks based on interferometer arrays.

[0021] This invention primarily considers a type of phase drift noise that originates from length deviations during waveguide manufacturing. This causes additional phase to accumulate as light propagates within the waveguide, and this phase can be reflected in the phase of the phase shifter. and The minute changes, coupled with imperfections in the manufacturing process of the phase shifter, prevent it from precisely adjusting the phase. or There will be a certain background error.

[0022] During step S101, the phase noise of the interferometer array is determined by an external computing device, which identifies the corresponding uniform correction deviation value. Unified correction deviation value This refers to the global compensation phase value set to mitigate the phase noise faced by all phase shifters in an interferometer array; it can also be understood as a variable for uniformly correcting the magnitude of the deviation. A single global adjustment can replace the correction of each phase shifter individually.

[0023] S102. With the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device, generate an objective function based on the correction deviation direction variable.

[0024] In implementation, this invention constructs an objective function for phase noise compensation of phase shifters in optical computing devices: minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array is taken as the core objective, and the correction deviation direction variable is taken as the core optimization variable. The correction deviation direction variable can be... This relates to the adjustment direction of phase noise compensation; +1 indicates that the phase noise of the current phase shifter causes the actual phase to deviate from the ideal value, which needs to be adjusted by positive phase compensation; -1 indicates the opposite situation, which needs to be adjusted by reverse phase adjustment. This invention transforms the noise mitigation problem of interferometer arrays into a global optimization task. By establishing a mapping between variables and light intensity deviation, the optimal phase correction scheme that allows the measured light intensity to accurately match the ideal value can be found by solving for the minimum value of the objective function.

[0025] S103. Solve the objective function to obtain the optimal values ​​of the correction deviation direction variables for each phase shifter.

[0026] In practice, this invention determines the deviation adjustment direction of each phase shifter, such as phase addition or subtraction, by solving the objective function.

[0027] S104. Send the optimal value of the correction deviation direction variable and the unified correction deviation value of each phase shifter to the control equipment so that the control equipment can adjust each phase shifter simultaneously according to the optimal value of the correction deviation direction variable and the unified correction deviation value of each phase shifter.

[0028] It should be noted that the present invention synchronously transmits the optimal values ​​of the correction deviation direction variables of each phase shifter obtained in step S103 and the unified correction deviation value determined in step S101 to the control device. In other words, the present invention provides two parameters for each phase shifter: the optimal value of the correction deviation direction variable and the unified correction deviation value. The control device then combines these two parameters for coordinated adjustment, simultaneously adjusting each phase shifter. This collective adjustment adds the product of the optimal value of the correction deviation direction variable and the unified correction deviation value to the phase of all phase shifters, ultimately restoring the phase state of the interferometer array from a noisy state to a near-ideal state. This ensures that the measured light intensity value of the photodetector array approaches the ideal light intensity value, guaranteeing the accuracy of subsequent optical calculations.

[0029] In the optical computing error adjustment method provided by the embodiments of the present invention, by determining a unified correction deviation value and combining it with the optimal value of the correction deviation direction variable of each phase shifter, collective adjustment of all phase shifters can be achieved, making the measured light intensity value of the photodetector array close to the ideal light intensity value, effectively improving the output accuracy of the optical computing device, and avoiding the additional overhead of individually calibrating the phase shifters, thereby significantly reducing the overall overhead of the entire interferometer array during operation. At the same time, the collective adjustment mode does not require individual debugging of each phase shifter, greatly simplifying the operation process of optical computing noise mitigation, reducing the complexity of optical computing noise mitigation, accelerating the noise mitigation process, and reducing the time required for calibration. It can be widely used in integrated optical networks, providing strong support for the efficient and low-power operation of integrated optical networks.

[0030] In practical applications, this invention focuses on interferometric optical computation, with an interferometer array as its core component. The computation utilizes the principle of light interference to directly perform linear algebraic operations at the physical level. Its core lies in constructing a unitary matrix transformation network through programmable phase modulation. This architecture consists of multiple interferometer units integrated into a photonic chip in a triangular z-grid or rectangular grid topology, with the input light field... Output after array transformation ,in The target operation matrix. Figure 2 This is a schematic diagram of the structure of a single Mach-Zehnder interferometer provided in an embodiment of the present invention. Figure 2As shown, a single Mach-Zehnder interferometer consists of two 50:50 beam splitters and two parallel waveguide arms, with one arm integrating two phase shifters (introducing a controllable phase difference). and Its transmission characteristics are described by a unitary matrix: The first phase modulation matrix is: ; The transmission matrix of the first beam splitter is: ; The second phase modulation matrix is: ; The transmission matrix of the second beam splitter is: ; The unitary transmission matrix of a single Mach-Zehnder interferometer is: ; By adjusting It can achieve any Unitary transformation.

[0031] To construct large-scale matrix operations, Mach-Zehnder interferometer elements are cascaded in a specific topology (such as a triangular mesh) to form a Mach-Zehnder interferometer array. Figure 3 This is a schematic diagram of the structure of a Mach-Zehnder interferometer array provided in an embodiment of the present invention. Figure 3 As shown, a single Mach-Zehnder interferometer is simplified to the portion shown in the dashed line.

[0032] for 3D input vector Its light field representation injection After entering the port, the signal undergoes a series of interferometric transformations within the array. The transmission matrix of the entire network is the product of the unitary matrices of each Mach-Zehnder interferometer element: ; in It is important to note that this network can implement arbitrary unitary transformations, because according to the Clems decomposition theorem, any... unitary matrix It can be decomposed into: ; here This is the diagonal phase compensation matrix. It is a specific Mach-Zehnder interferometer sequence. This can be understood as the coordinates of a Mach-Zehnder interferometer. Through precise configuration of all... This allows for the programming implementation of the target unitary matrix. Output light field This is the result of matrix multiplication.

[0033] In optical computation based on Mach-Zehnder interferometer arrays, the input vector The data loading process can be completed by injecting the optical signal into the input port via a spatial light modulator. speed( The light travels through the array (at the effective refractive index of the waveguide), undergoing continuous interference transformations, with a transmission time on the order of picoseconds (ps). Finally, the light intensity is measured at the output end via a photodetector array to obtain the result.

[0034] Furthermore, in a specific implementation, in the optical computing error adjustment method provided in the embodiments of the present invention, step S101, which determines the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device, may specifically include: after the control device randomly selects a set number of phase shifters and adjusts the phase of the phase shifters, obtaining the light intensity change output by the photodetector array; obtaining the average noise amplitude based on the fitting curve corresponding to the light intensity change output by the photodetector array, and using the average noise amplitude as the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device.

[0035] In practice, this invention can use a control device to randomly select a small number (e.g., at least three) of phase shifters for phase offset testing to calibrate their phase error. The specific method is as follows: Use the control device to fine-tune the phase of one phase shifter (e.g., increase the phase by a small amount). The method uses a photodetector array to measure changes in output light intensity. The average noise amplitude is estimated by fitting the change curve on a computing device and set as a uniform correction deviation value. This allows for the random selection and adjustment of phase shifters by the control device, combined with the fitted curve of the photodetector array's output light intensity change, to obtain the average noise amplitude. This eliminates the need for complex point-by-point detection across the entire array, making the operation simple and efficient. Using this as a uniform correction deviation value allows for the rapid capture of the overall characteristics of the array's phase noise, laying the foundation for subsequent accurate noise correction and mitigation of noise interference with the output. This helps improve the stability and output accuracy of the interferometer array while reducing resource consumption and time costs in the noise detection process.

[0036] Furthermore, in a specific implementation, in the above-mentioned optical calculation error adjustment method provided in the embodiments of the present invention, step S102, which generates an objective function based on the correction deviation direction variable, may specifically include: after obtaining the measured light intensity value and the ideal light intensity value of the photodetector array, generating the objective function by calculating the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array.

[0037] In practice, a single interferometer contains two phase shifters, with the first phase shifter having a phase of... And the noise is The phase of the second phase shifter is And the noise is The transmission matrix of the noisy interferometer is as shown above.

[0038] Assume the input of the interferometer is Therefore, its output under the first-order approximation should be: ; Note that the matrix above contains... The first term enclosed in parentheses represents the output of the interferometer under ideal conditions. If... The coefficient matrix above is denoted as The output of a noisy interferometer under ideal conditions is denoted as Therefore, under the first-order approximation, the output of the noisy interferometer should be: ; The above equation expresses the output error as a linear function of noise under a first-order approximation, making the correction process independent of specific phase values. This means that the output under noisy conditions is the output under ideal conditions plus some linear noise terms. Furthermore, this linear noise can be mitigated by applying a small phase shift to the phase shifter. These linear noise terms are independent of the specific phase value at this point, meaning that after one calibration, regardless of the phase setting, the calibration result can be reused, reducing the computational overhead of subsequent calculations.

[0039] The above analysis only considers the transformation that occurs when an ideal beam of light is input into a single interferometer. However, in real-world conditions, the noisy light often passes through one interferometer before entering another. In this case, it is necessary to consider the transformation that occurs when the noisy input is input into a noisy interferometer. Let's assume the noisy input of a certain noisy interferometer is... ,in and This represents the shift of the beam relative to the noise-free state. In a first-order approximation, it can be expressed as a linear combination of the noise levels from the phase shifters through which the beam passes. After passing through the noisy interferometer, its output in the first-order approximation can be expressed as: ; Similar to the analysis above, the first term in the above expression represents the output under ideal conditions. Also, note that... and This represents the noise term of the noisy phase shifter through which the light path passes. The output at this point can be rewritten as: ; in, This indicates the noise level of the phase shifter. This is the corresponding noise coefficient vector. Indicates the first A phase shifter.

[0040] In this case, this noise can be mitigated by introducing a small offset in each phase shifter. To alleviate this, in this case, following the same analysis, the corrected output can be obtained as follows: ; in, To correct the deviation direction variable, only one can be selected. This represents the direction of correction, and To standardize the correction deviation value, it represents the magnitude of the correction error. This is because it assumes noise in the interferometer array... Since the sizes are similar and the directions are random, this method can be used to approximately correct most errors. The value can be roughly determined by pre-calibrating several phase shifters, while The specific value needs to be determined by judging the noisy output. This reduces parameter complexity, accurately matches the deviation direction of each phase shifter, and eliminates the need for complex calculations of independent compensation for each phase shifter; it only requires judging the noisy output. Sure This can efficiently and approximately cancel out most of the noise, allowing the output to... Close to the ideal value While ensuring noise reduction, it significantly simplifies the phase calibration process of interferometer arrays in optical computing devices.

[0041] If we assume there are a total of If there is an interferometer, then there exists... By comparing the noisy output after correction with the ideal output using a phase shifter, we can obtain the following: ; Here, the constant terms are ignored. In this case, the problem of mitigating phase shifter noise in the interferometer array becomes finding a set of... The distribution of makes and The difference between them is minimal. This difference is determined by comparing the sum of the squared magnitudes of the differences between the output complex amplitudes of all ports and the ideal complex amplitude. The ideal output complex amplitude can be obtained by matrix multiplication between the ideal interferometer array transmission matrix and the ideal input. However, the problem lies in the need to measure the output complex amplitude of the interferometer under noisy conditions. This measurement of complex amplitude is difficult to perform in practical applications. In experiments, it is common to measure the output light intensity, thereby summing the squares of the light intensity differences between each output port and the ideal value.

[0042] First, consider the first If there are one output port, then its noisy output complex amplitude can be expressed as: ; in, Noise figure vector The The first component. Then the second... The light intensity output from each port can be expressed as: ; in, Let represent the real part. Then the sum of squares of the light intensity differences between all ports and the ideal output can be written as: ; in, Represent the objective function; This represents the measured light intensity value of the photodetector array; Indicates the ideal light intensity value; , They represent the first Phase shifter, the first A phase shifter.

[0043] If the two bodies The preceding coefficient is denoted as , to monomer The preceding coefficient is denoted as Then the above formula can be written as: ; This means that it can be adjusted This is done to minimize the difference between the noisy output and the ideal output, thereby mitigating the noise.

[0044] As can be seen from the above, the present invention can be adjusted That is, the noise mitigation direction of each phase shifter, measuring the output light intensity of each port, and calculating the sum of squares of the difference from the ideal output light intensity.

[0045] Based on this, in the specific implementation of the above-mentioned optical computing error adjustment method provided in the embodiments of the present invention, step S103 solves the objective function to obtain the optimal value of the correction deviation direction variable of each phase shifter. Specifically, it may include: mapping the objective function to the Ising model; solving the ground state of the Ising model to obtain the optimal value of the correction deviation direction variable of each phase shifter.

[0046] In implementation, this invention maps the objective function (i.e., the problem of minimizing output error) to an Ising model based on the correction bias direction variable, and then solves the ground state of the Ising model to obtain the optimal values ​​of the correction bias direction variables for each phase shifter. By leveraging the adaptability of the Ising model to discrete optimization problems, it accurately identifies the combination of variables that minimizes the sum of squares of the difference between the measured light intensity and the ideal light intensity. This approach directly generates the optimization objective function using the physical response of the optical system, eliminating the need to separately measure the noise offset of each phase shifter. This significantly reduces calibration complexity, provides a reliable basis for subsequent precise adjustment of the phase shifters, further mitigates phase noise interference, and improves the output accuracy of optical computing equipment while ensuring the efficiency and stability of the noise mitigation process.

[0047] Furthermore, in specific implementations, in the optical computing error adjustment method provided in the embodiments of the present invention, the constructed objective function can be mapped to the Ising model using the following formula: ; in, The objective function is... Let be the direction variable for correcting the noise of the i-th phase shifter; Let be the direction variable for correcting the noise of the j-th phase shifter; These are the elements of the quadratic coefficient matrix, reflecting... and The coupling relationship between them is determined by both the noise figure and the ideal output. The coefficient of the linear term reflects The linear effect is determined by the inherent noise, noise figure, and ideal output.

[0048] In implementation, this invention maps the objective function to the Ising model using the aforementioned formula. This clearly quantifies the coupling relationship between the phase shifter correction deviation direction variables and the linear influence of individual variables, transforming the complex noise mitigation optimization problem into a structured physical model solution problem. This mapping retains the core optimization logic of the objective function while providing an adaptive framework for efficient solution through the characteristics of the Ising model. It helps to accurately find the variable combination that minimizes the sum of squares of the light intensity difference, providing guidance for subsequent phase shifter adjustment, thereby improving the accuracy and efficiency of noise correction.

[0049] Furthermore, in specific implementation, in the above steps, solving the ground state of the Ising model to obtain the optimal values ​​of the correction bias direction variables for each phase shifter can specifically include: using a heuristic algorithm to iteratively update and solve the ground state of the Ising model; wherein, starting from the initial state, the correction bias direction variables and auxiliary variables are updated step by step; each time it is updated, the next correction bias direction variable is calculated based on the current correction bias direction variables and auxiliary variables, and then the auxiliary variables are updated in combination with the time evolution function that changes with the number of steps; during the update process, when the correction bias direction variables reach the positive and negative limit values, they remain unchanged, and the corresponding auxiliary variables are set to zero; when all correction bias direction variables reach the positive and negative limit values, the combination of the values ​​of all correction bias direction variables is used as the ground state solution of the Ising model.

[0050] In implementation, a heuristic algorithm is used to iteratively solve the ground state of the Ising model. By progressively updating the correction bias direction variables and auxiliary variables, and combining this with a time evolution function that changes with the number of steps to optimize the update logic, variables that reach positive or negative limits are locked and their corresponding auxiliary variables are reset to zero. This approach can efficiently converge to a stable solution, simultaneously optimizing the correction direction of all phase shifters. This method avoids complex computational redundancy and accurately identifies the variable combination that minimizes the objective function, quickly obtaining the optimal values ​​of the correction bias direction variables for each phase shifter. This provides a reliable basis for subsequent phase shifter adjustments, helps improve the efficiency and accuracy of interferometer array noise mitigation, and further ensures output accuracy.

[0051] Furthermore, in specific implementation, in the optical computing error adjustment method provided in the embodiments of the present invention, the iterative update solution method of the heuristic algorithm can be: ; ; in, For time interval variables, Let be the time evolution function at step t+1. For the corresponding time evolution coefficient, Let be the direction variable for correcting the noise of the i-th phase shifter at step t. For step t and Related auxiliary variables, Let be the direction variable of the correction deviation of the noise of the i-th phase shifter at step t+1. When it is the (t+1)th step and Related auxiliary variables, The constant coefficients, Let $\frac{t+1}{t}$ be the direction variable for correcting the noise of the $j$-th phase shifter.

[0052] In implementation, the aforementioned heuristic algorithm can employ a simulated bifurcation algorithm. When solving using the iterative formula of this heuristic algorithm, the correction deviation direction variable and auxiliary variables can be updated in an orderly manner through the time interval variable, the time evolution function that dynamically changes with the number of steps, and constant coefficients: first, the correction deviation direction variable is updated based on the current auxiliary variable, and then the auxiliary variable is optimized by combining variable coupling relationships and linear effects. This structured iterative logic ensures the accuracy of variable updates while efficiently advancing the solution process, quickly converging to stable values ​​for each correction deviation direction variable. This provides reliable support for obtaining the Ising model ground state and determining the optimal adjustment direction of the phase shifter, thereby improving the efficiency of interferometer array noise mitigation.

[0053] Furthermore, in specific implementation, in the above steps, the next correction deviation direction variable is calculated based on the current correction deviation direction variable and auxiliary variables, and then the auxiliary variables are updated in conjunction with the time evolution function that changes with the number of steps. Specifically, this may include: at step t+1, based on the current... and auxiliary variables ,calculate Combined with the time evolution function To obtain new auxiliary variables .

[0054] Accordingly, when the direction variable for correcting deviation reaches its positive or negative limit, it remains unchanged, and the corresponding auxiliary variable is set to zero. Specifically, this may include: when When it reaches +1 or -1, stay at Keep it still and set the corresponding Set it to 0.

[0055] In implementation, updating the variables according to the above rules at step t+1 not only relies on the current state of the correction deviation direction variable and auxiliary variables, but also combines the time evolution function to achieve ordered iteration of the variables, ensuring the consistency and accuracy of the update logic; it also achieves [something] through locking. By identifying the correction direction variable and the corresponding auxiliary variable for zeroing out, the already stable variables are prevented from being disturbed by subsequent iterations, thus accelerating convergence to a stable state. This efficiently advances the solution of the Ising model's ground state, quickly determines the optimal correction direction of the phase shifter, lays the foundation for accurately mitigating phase noise and improving output accuracy of the interferometer array, and simultaneously reduces redundant calculations, improving noise mitigation efficiency.

[0056] Furthermore, in specific implementation, in the above steps, when all correction bias direction variables reach their positive and negative limit values, the combination of all correction bias direction variable values ​​is used as the ground state solution of the Ising model. Specifically, this may include: when all correction bias direction variables reach their positive and negative limit values ​​and there is no numerical change in multiple consecutive iterations, the correction bias direction variables are binarized to obtain a sequence containing +1 and -1, and the sequence is used as the ground state solution of the Ising model; the ground state solution minimizes the objective function.

[0057] In implementation, the solution method of this invention is a heuristic algorithm that utilizes the current... At that time, the stable point of the system is the point that minimizes the objective function. Then, when each... After evolution ceases, by keeping it in Binarize it to obtain a set of values. of The sequence is used to adjust each phase shifter to mitigate noise in the interferometer array. By waiting for all correction deviation direction variables to reach their positive and negative limits and remain unchanged for several consecutive steps, and then binarizing them to obtain the ground state solution containing +1 and -1, the stability and reliability of the solution are ensured, avoiding optimization deviations caused by incomplete convergence. Furthermore, binarization precisely identifies the variable combination that minimizes the objective function, providing a basis for determining the optimal correction direction of the phase shifter. This improves output accuracy while ensuring the effectiveness and efficiency of the noise mitigation process.

[0058] Furthermore, in a specific implementation, in the optical computing error adjustment method provided in the embodiments of the present invention, the quadratic coefficient matrix elements of the Ising model are calculated using the following formula. and coefficient of the first term : ; ; in, Noise figure vector The kth component, Noise figure vector The kth component, for The complex conjugate, To achieve the ideal output complex amplitude, To standardize the correction deviation values, Let be the noise of the j-th phase shifter.

[0059] In practice, the quadratic coefficient matrix elements of the Ising model are calculated using the above formula. and coefficient of the first term This method accurately integrates key parameters such as ideal output complex amplitude, noise figure vector, unified correction bias value, and phase shifter noise into the model, clearly quantifying the coupling relationship between variables in each correction bias direction and the linear influence of individual variables. This precise parameter calculation method provides a foundation for the Ising model construction that closely matches the actual noise scenario, ensuring that the model can accurately map the objective function. This, in turn, helps to efficiently solve the ground state, find the optimal correction direction for the phase shifter, and effectively improve the accuracy of interferometer array noise mitigation.

[0060] Furthermore, in a specific implementation, the optical calculation error adjustment method provided in the embodiments of the present invention may further include, before solving the objective function, the following steps: obtaining the correction deviation direction variable randomly assigned by the control device to each phase shifter; the value of the correction deviation direction variable is a positive limit value or a negative limit value; wherein, the positive limit value represents an increase in the uniform correction deviation value, and the negative limit value represents a decrease in the uniform correction deviation value; and using the correction deviation direction variable randomly assigned by the control device to calculate the initial value of the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array.

[0061] In practice, this invention allows the control device to randomly assign a correction deviation direction variable (such as +1 or -1, corresponding to increasing or decreasing the uniform correction deviation value, respectively) to each phase shifter, and calculates the initial value of the sum of squares of the light intensity difference accordingly. This can quickly provide an initial reference benchmark for subsequent optimization solutions, avoid the solution blind zone without an initial state, and also rely on the randomly assigned discrete variables to initially cover the possible correction direction range, laying the foundation for subsequent accurate iterative optimization. At the same time, it simplifies the initial value calculation process, reduces the time spent on preliminary preparation, and helps improve the overall efficiency of noise mitigation of the interferometer array.

[0062] Furthermore, in a specific implementation, the optical calculation error adjustment method provided in the embodiments of the present invention may further include, before generating the objective function based on the correction deviation direction variable: acquiring the test optical signal input by the control device and the transmission matrix corresponding to the interferometer array under ideal conditions; performing calculations on the test optical signal and the transmission matrix under ideal conditions, and obtaining the signal matrix at the output end of the interferometer array under ideal conditions through the correspondence calculation between the matrices; and processing the signal corresponding to each output port in the obtained signal matrix to obtain the ideal light intensity value.

[0063] In practice, the ideal output signal matrix is ​​obtained by acquiring the test optical signal and the ideal transmission matrix and performing calculations. The ideal light intensity value of each port is obtained by further processing the matrix. This allows for the precise construction of a reference for noise mitigation. It ensures that the ideal light intensity value is highly matched with the ideal working state of the interferometer array, providing a reliable basis for subsequent calculation of the sum of squares of the difference between the measured light intensity and the ideal light intensity. Furthermore, through matrix operations and port signal processing, the ideal light intensity value is obtained systematically and accurately, avoiding errors caused by manually setting the reference. This lays a precise foundation for subsequent positioning of phase noise interference and determination of the optimal correction direction of the phase shifter.

[0064] It should be noted that the optical computing error adjustment method provided by this invention can first simplify the noise of the phase shifter. Assuming that the noise on each phase shifter is of similar magnitude and small in amount, the influence of noise on the phase can be reduced by the deviation from the ideal value. To describe it. Under the above assumptions, considering only the first-order approximation, the difference between the output complex amplitude and the ideal complex amplitude is: A linear function. Under this condition, the phase can be adjusted. or Set to or That is, introducing phase The adjustments, among which It represents the direction of calibration. This represents the magnitude of the actual calibration, which can be obtained through individual phase shifters. Under this phase adjustment, by comparing the difference between the output light intensity of the interferometer array under noisy conditions and the ideal light intensity, the problem of whether to perform positive or negative calibration on each phase shifter can be equated to an Ising model. Using heuristic algorithms, such as the simulated bifurcation algorithm, solving for the ground state of this Ising model yields the noise adjustment method for the interferometer array. It is important to emphasize that this invention executes heuristic algorithms such as the simulated bifurcation algorithm to solve for the ground state of the Ising model on an actual interferometer array, and during the execution of the ground state solution algorithm, the adjustment direction of each phase shifter... It is carried out simultaneously and synchronously, hence the name collective regulation method.

[0065] This invention uses a control device to apply the current correction deviation direction variable. Set to all phase shifters (i.e., adjust the phase of each phase shifter to...) or The test light signal is input using a control device, and the output port is measured using a photodetector array. actual light intensity Calculate the error on an external computing device. Updated on an external computing device according to a heuristic algorithm formula. The values ​​of are repeatedly calculated, and the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device is updated. The steps for obtaining values, until all The values ​​all reached External computing devices will be optimal. The sequence is passed to an external controller, which applies it to all phase shifters. The test light signal is rerun, and the output light intensity is measured to ensure the objective function is met. This significantly reduced [the computational load]. Subsequently, this set of methods was also used when performing other computational tasks. sequence.

[0066] Furthermore, it should be noted that this invention transforms noise mitigation into a global optimization problem by collectively and synchronously adjusting the phase directions of all phase shifters. A heuristic algorithm is used to solve for the optimal solution in one go, transforming the calibration operation from point-by-point debugging to batch processing. This avoids the independent operation of 2M phase shifters (M being the number of interferometers) and reduces reliance on high-precision calibration equipment, significantly saving hardware resources and energy consumption. Moreover, this invention minimizes output light intensity error by globally optimizing the phase shifter correction direction, approximating the ideal value under a first-order approximation. This not only reduces the error rate of tasks such as optical neural networks but also improves the robustness and reliability of the system in complex scenarios due to the long-term effectiveness of a single calibration. This invention does not require modification of the interferometer array's physical structure; noise mitigation can be achieved simply by embedding the optimization algorithm into the control system. This characteristic allows the invention to be directly applied to integrated optical computing devices, seamlessly compatible with traditional optical computing architectures. This method provides high-precision, low-energy computing power support for scenarios such as artificial intelligence training and real-time big data processing, accelerating the transformation of optical computing technology from the laboratory to industrial applications and breaking through the physical bottlenecks of traditional electronic computing.

[0067] From the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by software plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware, but in many cases the former is a better implementation method.

[0068] Embodiments of the present invention also provide an optical computing error adjustment device. Based on functional modules, this device includes: The correction deviation value determination module is used to determine the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device; the interferometer array includes at least two phase shifters; The objective function generation module is used to generate an objective function based on the correction deviation direction variable, with the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device. The objective function solution module is used to solve the objective function and obtain the optimal values ​​of the correction deviation direction variables for each phase shifter; The correction parameter sending module is used to send the optimal value of the correction deviation direction variable and the unified correction deviation value of each phase shifter to the control device, so that the control device can adjust each phase shifter according to the optimal value of the correction deviation direction variable and the unified correction deviation value of each phase shifter to obtain the interferometer array after noise reduction.

[0069] In the optical computing error adjustment device provided in the embodiments of the present invention, collective adjustment of all phase shifters can be achieved by determining a unified correction deviation value and combining it with the optimal value of the correction deviation direction variable of each phase shifter. This makes the measured light intensity value of the photodetector array close to the ideal light intensity value, effectively improving the output accuracy of the optical computing device and avoiding the additional overhead of individually calibrating the phase shifters, thereby significantly reducing the overall overhead of the entire interferometer array during operation. At the same time, the collective adjustment mode does not require individual debugging of each phase shifter, which greatly simplifies the operation process of optical computing noise mitigation, reduces the complexity of optical computing noise mitigation, speeds up the noise mitigation process, and reduces the time required for calibration. It can be widely used in integrated optical networks, providing strong support for the efficient and low-power operation of integrated optical networks.

[0070] Since the embodiments of the optical computing error adjustment device and the optical computing error adjustment method correspond to each other, the descriptions of the features in the embodiments corresponding to the optical computing error adjustment device can be found in the relevant descriptions of the embodiments corresponding to the optical computing error adjustment method, and will not be repeated here. Furthermore, it has the same beneficial effects as the optical computing error adjustment method mentioned above.

[0071] Furthermore, in a specific implementation, in the optical computing error adjustment device provided in the embodiments of the present invention, the correction deviation value determination module can be specifically used to obtain the light intensity change output by the photodetector array after the control device randomly selects a set number of phase shifters and adjusts the phase of the phase shifters; according to the fitting curve corresponding to the light intensity change output by the photodetector array, the average noise amplitude is obtained, and the average noise amplitude is used as the unified correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device.

[0072] Furthermore, in a specific implementation, in the optical computing error adjustment device provided in the embodiments of the present invention, the objective function solving module can be used to map the objective function to the Ising model; solve the ground state of the Ising model to obtain the optimal value of the correction deviation direction variable of each phase shifter.

[0073] Embodiments of the present invention also provide an optical computing error adjustment system, including an optical computing device, a control device, and an external computing device. The optical computing device includes a light source, an interferometer array, and a photodetector array; the interferometer array includes at least two phase shifters. The external computing device is used to determine the unified correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device; generate an objective function based on the correction deviation direction variable with the objective of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device; solve the objective function to obtain the optimal value of the correction deviation direction variable for each phase shifter; and send the optimal value of the correction deviation direction variable for each phase shifter to the control device. The control device is used to simultaneously adjust each phase shifter according to the optimal value of the correction deviation direction variable for each phase shifter and the unified correction deviation value.

[0074] The aforementioned optical computing device can specifically consist of a light source, an interferometer array composed of multiple interferometers, waveguides connecting the interferometers, and a photodetector array. Each interferometer includes two noisy phase shifters, two beam splitters, and a corresponding waveguide. The optical computing device includes both calibration and computation modes. The control device controls the optical computing device to emit a preset light signal and controls the selection of the phase shifters and their corresponding phase settings. The external computing device is responsible for the computational tasks, including fitting the relationship between the output light intensity and the phase deviation to obtain a uniform correction value for the deviation. Calculate the magnitude and direction of the correction deviation. The updated value is then output to the control device.

[0075] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program configured to execute the steps in any of the above-described embodiments of the optical computing error adjustment method when running.

[0076] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0077] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the optical computing error adjustment method.

[0078] Embodiments of the present invention also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described embodiments of the optical computing error adjustment method.

[0079] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0080] The foregoing has provided a detailed description of the optical computing error adjustment method, system, medium, and program product provided by this invention. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only intended to help understand the method and core ideas of this invention. It should be noted that those skilled in the art can make various improvements and modifications to this invention without departing from its principles, and these improvements and modifications also fall within the protection scope of this invention.

Claims

1. A method for adjusting optical computing errors, characterized in that, include: Determine the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device; The interferometer array includes at least two phase shifters; With the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device, an objective function based on the correction deviation direction variable is generated. Solving the objective function yields the optimal values ​​for the correction deviation direction variables of each phase shifter; The optimal value of the correction deviation direction variable of each phase shifter and the unified correction deviation value are sent to the control device so that the control device can adjust each phase shifter simultaneously according to the optimal value of the correction deviation direction variable of each phase shifter and the unified correction deviation value.

2. The optical computing error adjustment method according to claim 1, characterized in that, Determine the uniform correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device, including: After the control device randomly selects a set number of phase shifters and adjusts the phase of the phase shifters, it obtains the light intensity change output by the photodetector array; The average noise amplitude is obtained based on the fitting curve corresponding to the change in light intensity output by the photodetector array. The average noise amplitude is then used as the unified correction deviation value corresponding to the phase noise of the interferometer array in the optical computing device.

3. The optical computing error adjustment method according to claim 1, characterized in that, Solving the objective function yields the optimal values ​​for the correction deviation direction variables of each phase shifter, including: The objective function is mapped to the Ising model; Solving for the ground state of the Ising model yields the optimal values ​​for the correction deviation direction variables of each phase shifter.

4. The optical computing error adjustment method according to claim 3, characterized in that, The constructed objective function is mapped to the Ising model using the following formula: ； in, The objective function is... Let be the direction variable for correcting the noise of the i-th phase shifter; Let be the direction variable for correcting the noise of the j-th phase shifter; These are the elements of the quadratic coefficient matrix, reflecting... and The coupling relationship between them; The coefficient of the linear term reflects The linear effect.

5. The optical computing error adjustment method according to claim 4, characterized in that, Solving for the ground state of the Ising model yields the optimal values ​​of the correction bias direction variables for each phase shifter, including: A heuristic algorithm is used to iteratively update and solve the ground state of the Ising model; The process begins from the initial state and progressively updates the correction bias direction variable and auxiliary variables. During each update, the next correction bias direction variable is calculated based on the current correction bias direction variable and auxiliary variables, and the auxiliary variables are then updated using a time evolution function that changes with the number of steps. During the update process, when the correction bias direction variable reaches its positive or negative limit value, it remains unchanged, and the corresponding auxiliary variable is set to zero. When all correction bias direction variables reach their positive or negative limit values, the combination of all correction bias direction variable values ​​is used as the ground state solution of the Ising model.

6. The optical computing error adjustment method according to claim 5, characterized in that, The iterative update solution method of the heuristic algorithm is as follows: ； ； in, For time interval variables, Let be the time evolution function at step t+1. For the corresponding time evolution coefficient, Let be the direction variable for correcting the noise of the i-th phase shifter at step t. For step t and Related auxiliary variables, Let be the direction variable of the correction deviation of the noise of the i-th phase shifter at step t+1. When it is the (t+1)th step and Related auxiliary variables, The constant coefficients, Let $\frac{t+1}{t}$ be the direction variable for correcting the noise of the $j$-th phase shifter.

7. The optical computing error adjustment method according to claim 6, characterized in that, Calculate the next correction deviation direction variable based on the current correction deviation direction variable and auxiliary variables, and then update the auxiliary variables using a time evolution function that changes with the number of steps, including: At step t+1, based on the current and auxiliary variables ,calculate Combined with the time evolution function To obtain new auxiliary variables ; When the direction variable for correcting deviation reaches its positive or negative limit, it remains unchanged, and the corresponding auxiliary variables are set to zero, including: when When it reaches +1 or -1, it remains unchanged and the corresponding Set it to 0.

8. The optical computing error adjustment method according to claim 5, characterized in that, When all corrective bias direction variables reach their positive and negative limits, the combination of all corrective bias direction variable values ​​is taken as the ground state solution of the Ising model, including: When all correction bias direction variables reach their positive and negative limits and there is no numerical change in multiple consecutive iterations, the correction bias direction variables are binarized to obtain a sequence containing +1 and -1. This sequence is used as the ground state solution of the Ising model; the ground state solution minimizes the objective function.

9. The optical computing error adjustment method according to claim 4, characterized in that, The quadratic coefficient matrix elements of the Ising model are calculated using the following formula. and coefficient of the first term : ； ； in, Noise figure vector The kth component, Noise figure vector The kth component, for The complex conjugate, To achieve the ideal output complex amplitude, The unified correction deviation value, Let be the noise of the j-th phase shifter.

10. The optical computing error adjustment method according to claim 1, characterized in that, Before solving the objective function, the following steps are also included: The control device randomly assigns a correction deviation direction variable to each phase shifter; the value of the correction deviation direction variable is a positive limit value or a negative limit value; wherein, a positive limit value indicates an increase in the uniform correction deviation value, and a negative limit value indicates a decrease in the uniform correction deviation value. The initial value of the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array is calculated using the correction deviation direction variable randomly assigned by the control device.

11. The optical computing error adjustment method according to claim 1, characterized in that, Before generating the objective function based on the correction bias direction variable, the following steps are also included: Acquire the test optical signal input from the control device and, under ideal conditions, the transmission matrix corresponding to the interferometer array; The test optical signal is processed in conjunction with the ideal transmission matrix, and the signal matrix at the output of the interferometer array under ideal conditions is obtained through the correspondence calculation between the matrices. The signals corresponding to each output port in the obtained signal matrix are processed to obtain the ideal light intensity value.

12. The optical computing error adjustment method according to claim 1, characterized in that, Generate an objective function based on the direction variable of the correction deviation, including: After obtaining the measured light intensity value and the ideal light intensity value of the photodetector array, the objective function is generated by calculating the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array.

13. A system for adjusting optical computing errors, characterized in that, include: Optical computing devices, control devices, and external computing devices; The optical computing device includes a light source, an interferometer array, and a photodetector array; The interferometer array includes at least two phase shifters; The external computing device is used to determine the uniform correction deviation value corresponding to the phase noise of the interferometer array; With the goal of minimizing the sum of squares of the differences between the measured light intensity value and the ideal light intensity value of the photodetector array in the optical computing device, an objective function based on the correction deviation direction variable is generated; the objective function is solved to obtain the optimal value of the correction deviation direction variable of each phase shifter; and the optimal value of the correction deviation direction variable of each phase shifter is sent to the control device. The control device is used to simultaneously adjust each phase shifter based on the optimal value of the correction deviation direction variable of each phase shifter and the unified correction deviation value.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it implements the steps of the optical computing error adjustment method as described in any one of claims 1 to 12.

15. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the optical computing error adjustment method as described in any one of claims 1 to 12.

Images