A method and system for optical power equalization control

Abstract

The application provides a kind of optical power equalization control method and system, it is related to optical power equalization technical field.First, the output optical power of each core in multicore fiber is acquired, and distribution matrix is acquired;Then according to the deviation of the output optical power of each core and the preset target power, the compensation coefficient of each core is generated;Again, the input optical power of each core is adjusted according to the compensation coefficient;Finally, the compensation coefficient is dynamically corrected until the power error of each core is less than the threshold value.Because the optical power equalization control method provided in the application, by generating the compensation coefficient of each core, the output optical power of each core is more uniform.

Description

Technical Field

[0001] This application relates to the field of optical power equalization technology, and more specifically, to an optical power equalization control method and system. Background Technology

[0002] Multicore fiber (MCF) is a novel type of optical fiber that contains multiple independent cores within a common cladding region, enabling independent transmission of multiple signals through space division multiplexing (SDM). Its typical structure employs a fluorine-doped cladding refractive index profile or microstructure design, significantly improving transmission capacity and reducing inter-core crosstalk.

[0003] However, crosstalk and power imbalance exist between the fiber cores, leading to signal degradation. Summary of the Invention

[0004] The purpose of this application is to provide an optical power equalization control method and system to solve the problems of crosstalk and power unevenness between fiber cores.

[0005] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows: On one hand, embodiments of this application provide an optical power equalization control method, the method comprising: Obtain the output optical power of each core in a multi-core optical fiber and obtain the distribution matrix; Based on the deviation between the output optical power of each fiber core and the preset target power, a compensation coefficient for each fiber core is generated. The input optical power of each fiber core is adjusted according to the compensation coefficient. The compensation coefficient is dynamically adjusted until the power error of each fiber core is less than the threshold.

[0006] Optionally, the steps of obtaining the output optical power of each core in a multi-core optical fiber and obtaining the distribution matrix include: Acquire the output optical power of a portion of the fiber core transmitted by the optical power monitoring module; The output optical power of each fiber core is determined based on the output optical power of some fiber cores and the coupling matrix formula; whereby the coupling matrix formula is:

[0007] Let J represent the output optical power of the j-th fiber core, and K represent the total number of fiber cores. This represents the coupling amplitude coefficient from fiber core i to fiber core j. This represents the time-varying phase difference from the i-th fiber core to the j-th fiber core. This represents the input optical power of the i-th fiber core. This represents the random measurement noise or model error of fiber core j.

[0008] Optionally, after determining the output optical power of each fiber core based on the output optical power of a portion of the fiber cores and the coupling matrix formula, the method further includes: When there is a correlation error between the output optical power determined by the coupling matrix formula and the actual measured output optical power, environmental compensation is performed on the output optical power determined by the coupling matrix formula.

[0009] Optionally, the output optical power after environmental compensation is ,

[0010] in, This represents the systematic error caused by environmental factors, where a0, a1, a2, and a3 represent setpoint coefficients, and T represents temperature. Indicates strain, Indicates phase change, Indicates the wavelength of light. Represents the temperature coefficient of refractive index. Indicates the amount of temperature change. Indicates the refractive index of a material. Indicates the coefficient of thermal expansion. Indicates the change in length. Represents the dynamic coupling factor. This represents the nominal coupling coefficient.

[0011] Optionally, after the step of generating compensation coefficients for each fiber core based on the deviation between the output optical power of each fiber core and the preset target power, the method further includes: The correlation matrix is ​​determined based on the fiber core spacing and refractive index distribution; The compensation coefficient is adjusted based on the correlation matrix.

[0012] Optionally, the correlation matrix satisfies the formula:

[0013] in, Represents the correlation matrix. Let represent the covariance of the output optical power between fiber core i and fiber core j. This represents the standard deviation of the i-th fiber core. This represents the standard deviation of the j-th fiber core.

[0014] Optionally, the output optical power of each fiber core includes the output optical power transmitted by the optical power monitoring module and the predicted optical power. After adjusting the input optical power of each fiber core according to the compensation coefficient, the method further includes: Obtain the correlation coefficient between the predicted optical power and the output optical power transmitted by the optical power monitoring module; Determine the difference between the correlation coefficient at the current time and the correlation coefficient at the previous time. When the difference value is greater than the threshold, the optical power monitoring module is controlled to monitor the output optical power of the fiber core with the largest change.

[0015] Optionally, the correlation coefficient satisfies the formula:

[0016] in, This represents the predicted estimate of the j-th fiber core. This represents the historical average value of fiber core j. This represents the correlation coefficient between the optical power of fiber core i and fiber core j. This represents the input optical power of the i-th fiber core. This represents the historical average of the i-th fiber core.

[0017] Optionally, the step of dynamically correcting the compensation coefficient until the power error of each fiber core is less than a threshold includes:

[0018]

[0019]

[0020] in, Let Hδ(r) denote the objective function to minimize variable P, where P represents the total core power vector to be solved, K represents the total number of constraints or residual terms considered in the optimization problem, and Hδ(r) denote the Huber loss function, where δ represents the threshold parameter of the Huber loss function. This represents the residual of the k-th measurement or prediction. This represents the regularization coefficient of the gradient sparse constraint term. The L1 norm of the gradient term of the power distribution P is represented. Represents the coefficient vector / matrix of the k-th measurement equation or constraint. This represents the actual measured value of the k-th power.

[0021] On the other hand, this application also provides an optical power equalization control system, the system including a controller, the controller being used for the above-described optical power equalization control method.

[0022] Compared with the prior art, this application has the following advantages: This application provides an optical power equalization control method and system. First, the output optical power of each core in a multi-core optical fiber is obtained, and a distribution matrix is ​​acquired. Then, based on the deviation between the output optical power of each core and a preset target power, a compensation coefficient is generated for each core. Next, the input optical power of each core is adjusted according to the compensation coefficients. Finally, the compensation coefficients are dynamically corrected until the power error of each core is less than a threshold. Because the optical power equalization control method provided in this application generates compensation coefficients for each core, the output optical power of each core becomes more uniform.

[0023] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is an exemplary flowchart of the optical power equalization control method provided in an embodiment of this application.

[0026] Figure 2 This is a schematic diagram of the optical power equalization control system provided in an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0028] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0029] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0030] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0031] As described in the background section, currently, in multi-core optical fibers, the individual fibers may experience uneven output optical power and signal degradation due to crosstalk and other reasons.

[0032] In view of this, to solve the above problems, this application provides an optical power equalization control method, which improves the uniformity of the output optical power of the fiber core by combining a compensation coefficient. The optical power equalization control method provided in this application is illustrated below: As one implementation method, please refer to Figure 1 The method includes: S102, obtain the output optical power of each core in the multi-core optical fiber, and obtain the distribution matrix.

[0033] S104 generates compensation coefficients for each fiber core based on the deviation between the output optical power of each fiber core and the preset target power.

[0034] S106, adjust the input optical power of each fiber core according to the compensation coefficient.

[0035] S108, dynamically correct the compensation coefficient until the power error of each fiber core is less than the threshold.

[0036] By adjusting the output optical power of each fiber core using a compensation coefficient, PID control of each fiber core is achieved, resulting in more uniform output optical power across all fiber cores.

[0037] As one approach, when acquiring the output optical power of each fiber core, a real-time monitoring module is required. For example, a power meter can be used as the optical power monitoring module. Furthermore, in existing technologies, to monitor the output optical power of each fiber core, an equal number of power meters as the number of fiber cores needs to be set up. For example, when there are 7 fiber cores, 7 power meters are required.

[0038] However, setting up power meters equal to the number of fiber cores would increase monitoring costs and overall space requirements. Therefore, this application uses only a subset of power meters to acquire the output optical power of each fiber core. Then, based on fiber core correlation modeling, the output optical power of all fiber cores is predicted. This achieves the goal of obtaining the output optical power of all fiber cores by monitoring as few fiber cores as possible, reducing the number of power meters required. For example, when there are 7 fiber cores, only 3 power meters are needed. These 3 power meters can acquire the output optical power of 3 of the fiber cores. Then, combined with the fiber core correlation model, the output optical power of the remaining 4 fiber cores is predicted, thus obtaining the distribution matrix of the output optical power of all fiber cores.

[0039] Specifically, the output optical power of each fiber core can be determined based on the output optical power of some fiber cores and the coupling matrix formula. This coupling matrix is ​​a power coupling matrix between fiber cores constructed using a nonlinear model of the Mach-Zehnder interferometer field. As one implementation method, the formula for this coupling matrix is:

[0040] Let J represent the output optical power of the j-th fiber core, and K represent the total number of fiber cores. This represents the coupling amplitude coefficient from fiber core i to fiber core j. This represents the time-varying phase difference from the i-th fiber core to the j-th fiber core. This represents the input optical power of the i-th fiber core. This represents the random measurement noise or model error of fiber core j.

[0041] This coupling matrix differs from traditional static linear coupling methods by introducing phase dynamics. This coupling coefficient matrix constructs a "linear" model based on the Mach-Zehnder interferometry principle, which effectively describes the optical energy coupling between fiber cores caused by phase difference changes. This model is highly effective for linear perturbations such as microbending, stress, and temperature gradients. However, the influence of real-world environments is complex and nonlinear, and some effects may not be fully captured by the "linear phase difference" model. For example, extreme temperature changes may cause nonlinear changes in fiber material properties. Complex stress distributions may induce polarization state changes and mode coupling beyond simple phase models, or other unknown or difficult-to-model physical effects.

[0042] Therefore, this application can also introduce a nonlinear environmental compensation function when necessary. When there is a correlation error between the output optical power determined by the coupling matrix formula and the actual measured output optical power, environmental compensation is performed on the output optical power determined by the coupling matrix formula.

[0043] In this application, for the same fiber core, when there is a systematic error related to environmental parameters between the predicted output optical power and the actual measured output optical power, environmental compensation is performed on the predicted output optical power of the fiber core after the output optical power of the fiber core is predicted using the above coupling matrix formula. For example, when there are 7 fiber cores, defined as No. 1 to No. 7, by setting 3 power meters, the output optical power of 3 fiber cores can be measured, such as the output optical power of fiber cores 1 to 3. Based on this, the real-time output optical power of fiber cores 1 and 2 can be used, combined with the above coupling matrix formula, to predict the output optical power of fiber core 3. Then, the real-time output optical power of fiber core 3 measured by the power meter is compared with the predicted output optical power. If there is a systematic error related to environmental parameters between the two, environmental compensation needs to be added for the predicted output optical power. At the same time, when predicting the output optical power of subsequent fiber cores 4 to 7, environmental compensation also needs to be added. If the error between the two is relatively small, environmental compensation is not required when predicting the output optical power of the remaining fiber cores using the formula.

[0044] The output optical power after environmental compensation is: ,

[0045] in, This represents the systematic error caused by environmental factors, where a0, a1, a2, and a3 represent setpoint coefficients, and T represents temperature. Indicates strain, Indicates phase change, Indicates the wavelength of light. Represents the temperature coefficient of refractive index. Indicates the amount of temperature change. Indicates the refractive index of a material. Indicates the coefficient of thermal expansion. Indicates the change in length. Represents the dynamic coupling factor. This represents the nominal coupling coefficient.

[0046] By introducing systematic errors caused by environmental factors, the final predicted value can be made closer to the true value (i.e., the output optical power directly measured by the power meter). Furthermore, after obtaining the output optical power of all fiber cores (partially obtained directly from the power meter, and partially predicted), a distribution matrix of the output optical power, P=[P1, P2, ..., P...], is generated. N ], where N represents the number of fiber cores.

[0047] After obtaining the output optical power of all fiber cores, a compensation coefficient for each fiber core can be generated based on the deviation between the output optical power of each fiber core and the preset target power. The compensation coefficient for each fiber core is C=[C1, C2, ..., C...]. N The compensation coefficient can be obtained by looking up a table or other means.

[0048] To make the compensation coefficient more accurate, the correlation matrix X can be calculated based on the fiber core spacing and refractive index distribution. Then, the compensation coefficient generation module can be used to correct the compensation coefficient. The corrected compensation coefficient can be expressed as: C =C×X -1 .

[0049] The correlation matrix X can be constructed based on historical data, and can be specifically represented as follows:

[0050] in, Represents the correlation matrix. Let represent the covariance of the output optical power between fiber core i and fiber core j. This represents the standard deviation of the i-th fiber core. This represents the standard deviation of the j-th fiber core.

[0051] Then, the input optical power of each fiber core is adjusted according to the compensation coefficient. Furthermore, to achieve more precise and robust adjustment of the input optical power of each fiber core, after the step of adjusting the input optical power of each fiber core according to the compensation coefficient, the method further includes: Obtain the correlation coefficient between the predicted optical power and the output optical power transmitted by the optical power monitoring module.

[0052] Determine the difference between the correlation coefficient at the current time and the correlation coefficient at the previous time.

[0053] When the difference value is greater than the threshold, the optical power monitoring module is controlled to monitor the output optical power of the fiber core with the largest change.

[0054] The correlation coefficient satisfies the following formula:

[0055] in, This represents the predicted estimate of the j-th fiber core. This represents the historical average value of fiber core j. This represents the correlation coefficient between the optical power of fiber core i and fiber core j. This represents the input optical power of the i-th fiber core. This represents the historical average of the i-th fiber core.

[0056] The difference value satisfies the formula:

[0057] The correlation coefficient can be used to determine the correlation coefficient between the optical power of fiber core i and fiber core j.

[0058] It should be noted that, This measures the overall difference between the current inter-core power correlation matrix and the matrix from the previous time step. A large The value indicates that the environment (temperature, stress) or system state of the optical fiber has undergone drastic changes in a short period of time (e.g., the fiber is severely bent, or a cooling system failure causes localized overheating). Therefore, when Δρ(t) is small, the environment can be considered stable, and the coupling relationship between fiber cores is predictable. The system relies on the previously established correlation matrix and adopts a "sparse monitoring" strategy. That is, only a few fiber cores (set S) are actually monitored, and then the output optical power of all other fiber cores is estimated using a formula. Based on these estimates, the compensation coefficient C is calculated to drive the VOA adjustment module for adjustment, achieving the effects of saving hardware (number of power meters), low computational load, and fast response.

[0059] And when When a certain preset threshold is exceeded, a system alarm is triggered. Drastic changes in the correlation matrix mean that the previously established ρ(ij) based on historical data has temporarily become "invalid" or inaccurate, and continuing to use it for estimation will lead to significant errors. At this point, the system's control strategy will dynamically change, reselecting the fiber cores measured by the power meter and dynamically adding volatile fiber cores for additional compensation. Here, "volatile fiber cores" refer to the fiber core pairs (i,j) with the largest values; the coupling relationship between these fiber cores changes most drastically and is the primary source of uncertainty. "Additional selection" means that the system will temporarily increase actual monitoring of these "volatile fiber cores." That is, it will instruct the hardware system (through optical switches or multiplexers) to prioritize the allocation of power meter resources to these most volatile fiber cores, adding them to set S.

[0060] The aim is to obtain accurate measurements of these critical fiber cores, rather than relying on unreliable estimates. Therefore, Δρ(t) dynamically optimizes the allocation of monitoring resources by monitoring the health of the system model, ensuring the accuracy of power estimation and ultimately achieving more precise and robust regulation of the input optical power.

[0061] By dynamically adjusting the compensation coefficients until the power error of each fiber core is less than the threshold, this application constructs a robust estimation algorithm to suppress the interference of outliers. Unlike the traditional least squares method, this application adopts the Huber loss function + gradient sparsity constraint.

[0062] Specifically, it satisfies:

[0063]

[0064]

[0065] in, Let Hδ(r) denote the objective function to minimize variable P, where P represents the total core power vector to be solved, K represents the total number of constraints or residual terms considered in the optimization problem, and Hδ(r) denote the Huber loss function, where δ represents the threshold parameter of the Huber loss function. This represents the residual of the k-th measurement or prediction. This represents the regularization coefficient of the gradient sparse constraint term. The L1 norm of the gradient term of the power distribution P is represented. Represents the coefficient vector / matrix of the k-th measurement equation or constraint. This represents the actual measured value of the k-th power.

[0066] Simultaneously, a dynamic sparsity monitoring and compensation algorithm is introduced. First, the fiber cores are clustered, dividing N fiber cores into M local clusters (e.g., 7-core fiber divided into 3 clusters), and each cluster independently runs Kalman filtering.

[0067] Here is the state transition matrix. The Kalman gain is then applied. The cluster results are then weighted and fused using a consensus protocol.

[0068] The above formula characterizes which fiber cores exhibit convergent power behavior under most normal conditions, revealing the steady-state coupling characteristics determined by the inherent physical structure of the optical fiber (such as core spacing). Properties: Relatively stable, changing slowly. Unless the physical structure of the optical fiber undergoes a permanent change, its macroscopic correlation pattern remains unchanged.

[0069] The Huber loss function, combined with gradient sparsity constraints, is calculated and filtered based on instantaneous real-time measurements. It can be dynamically updated in real time and is specifically designed to handle transient disturbances, noise, and outliers.

[0070] Specifically, in the formula In the equation, the left side represents the residual, which directly characterizes the difference between the "function prediction" and the "actual measurement value". The first term on the right side is the theoretical, predicted power value, which comes from the coupling matrix formula or the corrected model after environmental compensation; the second term is the actual measurement value directly from the optical power monitoring module.

[0071] The Huber loss function operates directly on the residuals mentioned above, and its function is to adjudicate the error. When the measurement is reliable (small residuals): if the measurement of a fiber core contains only small, random noise, the Huber function uses squared loss. Its function is that the algorithm behaves similarly to the classic least squares method, finely adjusting the power vector P to make the model's predicted value as close as possible to this reliable measurement. This ensures the high accuracy of the control system under normal conditions. When the measurement value is questionable or outlier (large residual): If a power meter temporarily malfunctions, or due to strong, transient external disturbances (such as micro-bending), a measurement value deviates significantly from the model's prediction. Its function is to switch the Huber function to linear loss, which greatly reduces the impact of outlier measurements on the overall optimization result. Without this mechanism, the outlier, due to its large squared term, would affect the entire optimization process, leading to completely incorrect compensation coefficients and potentially causing system oscillations. The Huber function ensures system stability. Therefore, by using the Huber loss function, instead of directly changing the measured value, different weights are assigned to residuals with varying reliability in the optimization objective, ensuring that the final power vector is more aligned with normal measurements while ignoring obvious outliers.

[0072] Gradient sparsity constraint Unlike other optical fibers, it acts as a "spatial smoothness enforcer." On one hand, it utilizes the spatial relationships between measurements. In multi-core optical fibers, adjacent cores (e.g., cores 1 and 2) are closely spaced, and their physical properties (crosstalk, temperature) are highly correlated. Therefore, a correlation matrix can be established using core spacing and refractive index distribution. Gradient sparsity constraint. The first constraint directly penalizes drastic power changes between adjacent fiber cores. It requires the optimizer to consider the spatial correlation between cores when searching for the optimal solution, aiming to find a spatially "smooth" power distribution. On the other hand, gradient sparsity constraints can combat the propagation of crosstalk and local disturbances. For example, suppose the measured value of fiber core 1 suddenly drops abnormally due to some local disturbance. Relying solely on the Huber function, while mitigating the impact of this point, might still allow unreasonable fluctuations in the power estimates of its neighboring cores. Gradient sparsity constraints intervene here, determining that since cores 1 and 2 are adjacent, their powers should not differ too much. Therefore, it forces the estimated power of core 2 to maintain a smooth transition, thus suppressing the spatial propagation of local disturbances through crosstalk and obtaining a more physically accurate overall power distribution.

[0073] Gradient sparsity constraints inject experience regarding the physical structure of optical fibers (core spacing) into the optimization process. It utilizes the spatial correlation revealed by measurements from all cores to correct for spatially unreasonable power estimates that may result from a single or a few unreliable measurements.

[0074] As can be seen, this application utilizes the Huber loss function and gradient sparsity constraints to dynamically adjust the compensation coefficients; the combined effect of these two methods forms a powerful and robust estimation algorithm. This estimation algorithm possesses the following characteristics: Data input: Actual measurements from part or all of the fiber core are input into the system to calculate the residuals.

[0075] Huber filtering: The Huber loss function first filters at the numerical level, judging the reliability of each measurement and determining its weight in the optimization.

[0076] Spatial constraints: Gradient sparsity constraints are then smoothed at the spatial level to ensure that even after Huber filtering, the power distribution obtained conforms to the physical laws of inter-core correlation.

[0077] Output: The optimization algorithm solves this constrained objective function and obtains a new, more robust, and more physically accurate full-core power estimation vector P.

[0078] Closed-loop control: The P obtained from this optimization is used to calculate the new compensation coefficient, adjust the input optical power, and then the system measures again to start a new round of iteration until power balance is achieved.

[0079] It is evident that by using the Huber loss function and gradient sparsity constraints to dynamically correct the compensation coefficients, the control system in this application not only relies on real measurement data but also intelligently processes noise and anomalies in these data, achieving stable and reliable optical power equalization in complex and ever-changing real environments.

[0080] The Huber loss function plus gradient sparsity constraint described above mainly achieves the following functions: 1. The most important manifestation: providing a basis for subsequent "clustering".

[0081] Clustering principle: Fiber cores with high ρ(ij) values ​​(i.e., highly correlated power behavior) are grouped into the same cluster. For example, in a standard 7-core fiber (1 core in the center and 6 cores on the periphery), the two adjacent cores on the periphery have the strongest crosstalk because they are closest to each other, and their ρ(ij) values ​​will be very high, so they are very likely to be grouped into the same cluster.

[0082] Since the accuracy of a Kalman filter depends on an accurate system model (state transition matrix Fc and observation matrix Hc), fiber cores within the same cluster, due to their highly correlated behavior, can have their state changes described by a simpler and more accurate model. If unrelated fiber cores are forcibly placed in a cluster, the model becomes extremely complex and inaccurate.

[0083] The correlation is reflected in the fact that the ρ(ij) matrix, as prior knowledge, determines the "organizational structure" of the Kalman filter. It ensures that each local Kalman filter operates on an optimal subset of the most relevant data, thereby improving the accuracy of local estimation.

[0084] 2. This serves as the initialization basis for the Kalman filter described above.

[0085] A Kalman filter requires an initial state estimate before iterative processing can begin. This initial value can be derived from an estimation model based on ρ(ij). Although the system primarily relies on real-time data once operational, a good initial value can accelerate the filter's convergence.

[0086] 3. As an indicator for system health monitoring and fault diagnosis.

[0087] The stability of ρ(ij) is a characteristic of the normal operation of the system.

[0088] If the calculated long-term ρ(ij) suddenly shows a significant and persistent difference from the previously calibrated value on a certain day, it may mean that the fiber optic link has suffered permanent physical damage (such as microcracks or extrusion deformation in a fiber core).

[0089] At this point, the system can issue an alert, indicating that maintenance is needed. This goes beyond real-time control, but it is crucial for ensuring the reliability of data center infrastructure.

[0090] 4. As a guarantee for redundant design.

[0091] This is a supplement to the previous discussion. When Δρ(t) is too large and the system decides to "add" more fiber cores for monitoring, it still needs to know which fiber cores are "important". The ρ(ij) matrix can help identify critical fiber cores: Highly central fiber cores: For example, in multi-core optical fibers, the central core often has strong coupling with all other cores (ρ(ij) value is not low). Monitoring it may yield the most system information. Bridge cores: Cores connecting two highly correlated clusters; changes in their behavior may affect multiple clusters.

[0092] Based on the optical power equalization control method provided in this application, the applicant conducted the following tests. In the seven-core fiber optic system, the monitoring configuration is as follows: a fixed central core (#4) and dynamically added edge cores (#1, #7), with a sampling rate of 50kHz. The control module performs complex calculations via a host computer, and the crosstalk matrix X is pre-calibrated based on the fiber parameters. The VOA array adjustment range is 0-5dB, with a step accuracy of 0.05dB. Test results show that the power fluctuation of each core decreased from ±2dB to ±0.2dB. It is evident that the optical power equalization control method provided in this application can effectively achieve output optical power equalization.

[0093] Based on the above implementation, this application also provides an optical power equalization control system, which includes a controller for executing the aforementioned optical power equalization control method. Please refer to [link to relevant documentation]. Figure 2 The optical power equalization control system also includes a silicon photonics modulator module, an optical power monitoring module, an adjustment and control module, and a multi-channel VOA adjustment module.

[0094] In summary, this application provides an optical power equalization control method and system. First, the output optical power of each core in a multi-core optical fiber is obtained, and a distribution matrix is ​​acquired. Then, based on the deviation between the output optical power of each core and a preset target power, a compensation coefficient for each core is generated. Next, the input optical power of each core is adjusted according to the compensation coefficients. Finally, the compensation coefficients are dynamically corrected until the power error of each core is less than a threshold. Because the optical power equalization control method provided in this application generates compensation coefficients for each core, the output optical power of each core becomes more uniform.

[0095] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0096] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this application. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for optical power equalization control, characterized in that, The method includes: Obtain the output optical power of each core in a multi-core optical fiber and obtain the distribution matrix; Based on the deviation between the output optical power of each fiber core and the preset target power, a compensation coefficient for each fiber core is generated. The input optical power of each fiber core is adjusted according to the compensation coefficient. The compensation coefficient is dynamically adjusted until the power error of each fiber core is less than the threshold.

2. The optical power equalization control method as described in claim 1, characterized in that, The steps for obtaining the output optical power of each core in a multi-core optical fiber and obtaining the distribution matrix include: Acquire the output optical power of a portion of the fiber core transmitted by the optical power monitoring module; The output optical power of each fiber core is determined based on the output optical power of some fiber cores and the coupling matrix formula; whereby the coupling matrix formula is: Let J represent the output optical power of the j-th fiber core, and K represent the total number of fiber cores. This represents the coupling amplitude coefficient from fiber core i to fiber core j. This represents the time-varying phase difference from the i-th fiber core to the j-th fiber core. This represents the input optical power of the i-th fiber core. This represents the random measurement noise or model error of fiber core j.

3. The optical power equalization control method as described in claim 2, characterized in that, After determining the output optical power of each fiber core based on the output optical power of a portion of the fiber cores and the coupling matrix formula, the method further includes: When there is a correlation error between the output optical power determined by the coupling matrix formula and the actual measured output optical power, environmental compensation is performed on the output optical power determined by the coupling matrix formula.

4. The optical power equalization control method as described in claim 3, characterized in that, The output optical power after environmental compensation is , in, This represents the systematic error caused by environmental factors, where a0, a1, a2, and a3 represent setpoint coefficients, and T represents temperature. Indicates strain, Indicates phase change, Indicates the wavelength of light. Represents the temperature coefficient of refractive index. Indicates the amount of temperature change. Indicates the refractive index of a material. Indicates the coefficient of thermal expansion. Indicates the change in length. Represents the dynamic coupling factor. This represents the nominal coupling coefficient.

5. The optical power equalization control method as described in claim 1, characterized in that, After generating compensation coefficients for each fiber core based on the deviation between the output optical power of each fiber core and the preset target power, the method further includes: The correlation matrix is ​​determined based on the fiber core spacing and refractive index distribution; The compensation coefficient is adjusted based on the correlation matrix.

6. The optical power equalization control method as described in claim 5, characterized in that, The correlation matrix satisfies the formula: in, Represents the correlation matrix. Let represent the covariance of the output optical power between fiber core i and fiber core j. This represents the standard deviation of the i-th fiber core. This represents the standard deviation of the j-th fiber core.

7. The optical power equalization control method as described in claim 1, characterized in that, The output optical power of each fiber core includes the output optical power transmitted by the optical power monitoring module and the predicted optical power. After adjusting the input optical power of each fiber core according to the compensation coefficient, the method further includes: Obtain the correlation coefficient between the predicted optical power and the output optical power transmitted by the optical power monitoring module; Determine the difference between the correlation coefficient at the current time and the correlation coefficient at the previous time. When the difference value is greater than the threshold, the optical power monitoring module is controlled to monitor the output optical power of the fiber core with the largest change.

8. The optical power equalization control method as described in claim 7, characterized in that, The correlation coefficient satisfies the formula: in, This represents the predicted estimate of the j-th fiber core. This represents the historical average value of fiber core j. This represents the correlation coefficient between the optical power of fiber core i and fiber core j. This represents the input optical power of the i-th fiber core. This represents the historical average of the i-th fiber core.

9. The optical power equalization control method as described in claim 1, characterized in that, The step of dynamically correcting the compensation coefficient until the power error of each fiber core is less than the threshold includes: The compensation coefficients are dynamically corrected using the Huber loss function and gradient sparsity constraints.

10. A power equalization control system, characterized in that, The system includes a controller for executing the optical power equalization control method described above.

Images