An optical signal wavelength optimization method, device, storage medium and processor

Abstract

The application discloses an optical signal wavelength optimization method and device, a storage medium and a processor. In the scheme, the wavelength distortion features in the original optical signal data are extracted; the target distortion distribution data constructed according to the distortion categories of the wavelength distortion features is deeply analyzed to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters, and mapping relationship data is determined according to the parameters and parameters of the optical communication system; wavelength correction coefficients corresponding to the optical signal are generated based on the mapping relationship data, the initial wavelength parameters are adjusted to obtain target wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material; the distortion change degree of the optical signal in the transmission environment of the optical communication system is simulated based on the target wavelength parameters; and the target wavelength parameters are adjusted according to the distortion change degree. The application solves the problem that the existing technology cannot effectively suppress wavelength distortion due to the difficulty in adapting to the dynamically changing transmission environment, and guarantees the signal transmission quality of the optical communication system.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method, apparatus, storage medium and processor for optimizing optical signal wavelength. Background Technology

[0002] In the field of optical communication, wavelength distortion of optical signals is a complex and critical technical challenge, directly affecting the transmission quality of optical signals and the performance of optical communication systems. With the increase in optical signal transmission distance and data rate, the effects of dispersion effects (such as material dispersion and waveguide dispersion) and nonlinear effects (such as four-wave mixing and stimulated Raman scattering) in optical fibers significantly amplify their impact on optical signals, leading to intensified wavelength distortion. This type of wavelength distortion not only reduces the quality of optical signals but also increases the complexity of optical communication systems, limiting the improvement of transmission capacity.

[0003] Existing technologies typically employ static compensation methods, such as fixed dispersion compensation modules or pre-compensation techniques, to suppress wavelength distortion. However, these methods are ill-suited to dynamic transmission environments, such as temperature fluctuations and adjustments to fiber optic link topology, thus failing to effectively suppress wavelength distortion.

[0004] Therefore, there is an urgent need for an adaptive wavelength optimization method to effectively suppress the increasingly serious wavelength distortion problem in high-speed, high-capacity optical communication systems and ensure the signal transmission quality and system performance of optical communication systems. Summary of the Invention

[0005] Based on the above problems, this application provides an optical signal wavelength optimization method, device, storage medium and processor, with the aim of solving the technical problem in the prior art that it is difficult to adapt to the dynamically changing transmission environment, resulting in the inability to effectively suppress wavelength distortion, and ensuring the signal transmission quality and system performance of the optical communication system.

[0006] The embodiments of this application disclose the following technical solutions:

[0007] In a first aspect, this application provides a method for optimizing the wavelength of an optical signal, the method comprising:

[0008] Raw optical signal data is acquired from the optical communication system; the raw optical signal data includes the initial wavelength parameters of the optical signal;

[0009] Wavelength distortion features are extracted from the original optical signal data, and target distortion distribution data is constructed based on the distortion categories of the wavelength distortion features. The distortion categories of the wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion. The target distortion distribution data is used to characterize the distribution of wavelength distortion in the time dimension or spatial dimension.

[0010] A pre-trained convolutional neural network model performs deep analysis on the target distortion distribution data to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters, the polarization mode dispersion characteristic parameters, and the parameters of the optical communication system, mapping relationship data is determined. The mapping relationship data is used to characterize the mapping relationship between the nonlinear crosstalk characteristic parameters and the polarization mode dispersion characteristic parameters and the complexity of the optical communication system.

[0011] Based on the mapping relationship data, the pre-trained generative adversarial network model generates wavelength correction coefficients corresponding to the optical signal, and adjusts the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters.

[0012] Based on the target wavelength parameters, the degree of distortion change of the optical signal in the transmission environment of the optical communication system is simulated; the degree of distortion change is used to measure the distortion performance of the optical communication system during signal transmission.

[0013] The target wavelength parameter is adjusted according to the degree of distortion change in order to optimize the wavelength of the optical signal.

[0014] In an optional implementation, adjusting the target wavelength parameter based on the degree of distortion change to optimize the wavelength of the optical signal includes:

[0015] If the degree of distortion change is greater than or equal to a preset threshold, then wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal are obtained.

[0016] Based on the wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal, the target wavelength parameters are adjusted to obtain the adjusted target wavelength parameters.

[0017] Polarization mode crosstalk analysis is performed on the adjusted target wavelength parameters to obtain the first signal compensation coefficient, and higher-order dispersion effect analysis is performed on the adjusted target wavelength parameters to obtain the second signal compensation coefficient.

[0018] The wavelength of the optical signal is optimized based on the first signal compensation coefficient and the second signal compensation coefficient.

[0019] In an optional implementation, the step of extracting wavelength distortion features from the original optical signal data and constructing target distortion distribution data based on the distortion categories of the wavelength distortion features includes:

[0020] Extract power spectral density data from the original optical signal data;

[0021] Based on the power spectral density data, the error between the actual power value and the ideal power value at each wavelength point of the optical signal is calculated to obtain the wavelength distortion characteristics;

[0022] The wavelength distortion features are classified into adjacent channel interference distortion and group velocity dispersion distortion.

[0023] Adjacent channel interference analysis is performed on the wavelength distortion characteristics to obtain the first distortion distribution data;

[0024] Group velocity dispersion analysis was performed on the wavelength distortion characteristics to obtain the second distortion distribution data;

[0025] The target distortion distribution data is constructed based on the first distortion distribution data and the second distortion distribution data.

[0026] In an optional implementation, the step of performing deep analysis on the target distortion distribution data using a pre-trained convolutional neural network model to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters, and determining mapping relationship data based on the nonlinear crosstalk characteristic parameters, the polarization mode dispersion characteristic parameters, and the parameters of the optical communication system, includes:

[0027] The target distortion distribution data is input into the convolutional neural network model, and the convolutional neural network model performs classification processing on the target distortion distribution data to classify the target distortion distribution data into nonlinear crosstalk distortion and polarization mode dispersion distortion;

[0028] The nonlinear crosstalk characteristic parameters are obtained by performing nonlinear crosstalk analysis on the target distortion distribution data using the convolutional neural network model.

[0029] The polarization mode dispersion characteristic parameters of the target distortion distribution data are obtained by performing polarization mode dispersion analysis on the convolutional neural network model.

[0030] Based on the nonlinear crosstalk characteristic parameters, the polarization mode dispersion characteristic parameters, and the parameters of the optical communication system, a regression analysis model is used to generate the complexity index of the optical communication system.

[0031] The mapping relationship data is generated based on the nonlinear crosstalk characteristic parameters, the polarization mode dispersion characteristic parameters, and the corresponding complexity index.

[0032] In an optional implementation, the pre-trained generative adversarial network model generates wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data, and adjusts the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters, including:

[0033] The mapping relationship data is input into the generative adversarial network model, and the generative adversarial network model matches the corresponding wavelength correction coefficient for the optical signal based on the mapping relationship data;

[0034] The initial wavelength parameters are corrected based on the wavelength correction coefficient to obtain the corrected wavelength parameters;

[0035] Dispersion analysis is performed on the corrected wavelength parameters based on the dispersion characteristics of the optical fiber material to obtain dispersion results; the dispersion results include dispersion values ​​and dispersion slopes.

[0036] The wavelength shift is calculated based on the dispersion value and the dispersion slope;

[0037] The corrected wavelength parameters are adjusted based on the wavelength offset to obtain the target wavelength parameters.

[0038] In an optional implementation, simulating the degree of distortion variation of the optical signal in the transmission environment of the optical communication system based on the target wavelength parameter includes:

[0039] Based on the target wavelength parameter, optical power leakage simulation analysis is performed on the optical signal in the transmission environment of the optical communication system to obtain the optical power leakage value corresponding to the optical signal.

[0040] If the optical power leakage value is greater than the preset leakage value, then the first compensation gain value is calculated based on the target wavelength parameter using the power compensation model;

[0041] The degree of distortion change is determined based on the first compensation gain value.

[0042] In an optional implementation, determining the degree of distortion change based on the first compensation gain value includes:

[0043] The target wavelength parameters are adjusted based on the first compensation gain value to obtain the compensated target wavelength parameters;

[0044] Based on the compensated target wavelength parameters, the optical signal is subjected to dispersion compensation error analysis in the optical communication system transmission environment to obtain the compensation error corresponding to the optical signal.

[0045] Based on the compensation error, a parameter correction coefficient is generated, and based on the parameter correction coefficient, the above-compensated target wavelength parameter is corrected to obtain the corrected target wavelength parameter.

[0046] The degree of distortion change is determined based on the corrected target wavelength parameters and the initial wavelength parameters.

[0047] A second aspect of this application provides an optical signal wavelength optimization device, the device comprising:

[0048] A data acquisition module is used to acquire raw optical signal data from an optical communication system; the raw optical signal data includes the initial wavelength parameters of the optical signal.

[0049] The feature extraction module is used to extract wavelength distortion features from the original optical signal data and construct target distortion distribution data based on the distortion categories of the wavelength distortion features. The distortion categories of the wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion. The target distortion distribution data is used to characterize the distribution of wavelength distortion in the time dimension or spatial dimension.

[0050] The analysis module is used to perform deep analysis on the target distortion distribution data using a pre-trained convolutional neural network model to obtain nonlinear crosstalk feature parameters and polarization mode dispersion feature parameters. Based on the nonlinear crosstalk feature parameters, the polarization mode dispersion feature parameters, and the parameters of the optical communication system, the mapping relationship data is determined. The mapping relationship data is used to characterize the mapping relationship between the nonlinear crosstalk feature parameters and the polarization mode dispersion feature parameters and the complexity of the optical communication system.

[0051] The adjustment module is used to generate wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data by the pre-trained generative adversarial network model, and adjust the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters.

[0052] The simulation module is used to simulate the degree of distortion change of the optical signal in the transmission environment of the optical communication system based on the target wavelength parameter; the degree of distortion change is used to measure the distortion performance of the optical communication system during signal transmission.

[0053] An optimization module is used to adjust the target wavelength parameter according to the degree of distortion change, so as to optimize the wavelength of the optical signal.

[0054] In a third aspect, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned optical signal wavelength optimization method.

[0055] In a fourth aspect, this application provides a processor for running a computer program, which executes the above-described optical signal wavelength optimization method during operation.

[0056] Compared with the prior art, this application has the following beneficial effects:

[0057] In this technical solution, by acquiring raw optical signal data from the optical communication system in real time, the initial wavelength parameters of the optical signal can be comprehensively obtained, providing accurate basic data support for subsequent wavelength analysis and compensation, and ensuring the dynamic perception capability of the optical communication system of the optical signal state. Then, wavelength distortion features are extracted from the raw optical signal data, and target distortion distribution data is constructed based on the distortion categories of the wavelength distortion features. Since the distortion categories of wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion, the target distortion distribution data is used to characterize the distribution of wavelength distortion in the time or spatial dimensions. Therefore, by classifying and extracting adjacent channel interference distortion and group velocity dispersion distortion, different types of distortion sources can be accurately identified. Combined with the distribution features in the time or spatial dimensions, the dynamic evolution law of wavelength distortion can be more comprehensively characterized, providing a basis for subsequent targeted compensation.

[0058] Then, a pre-trained convolutional neural network model performs in-depth analysis of the target distortion distribution data, enabling rapid identification of complex interferences in optical signal transmission (such as mutual interference between signals and irregular changes during light wave propagation) to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and parameters of the optical communication system, the mapping relationship data between the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and the complexity of the optical communication system is determined. This accurately identifies the correlation between the above-mentioned interference problems and system settings, providing a data foundation for subsequent accurate correction of the above-mentioned interference. Then, a pre-trained generative adversarial network model generates wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data. Based on the wavelength correction coefficients and the dispersion characteristics of the optical fiber material, the initial wavelength parameters are adaptively adjusted to obtain the target wavelength parameters. This achieves dynamic response to changes in the transmission environment, significantly improves the flexibility and real-time performance of the compensation strategy, and effectively addresses signal distortion problems in complex scenarios.

[0059] Then, based on the target wavelength parameters, the degree of distortion of the optical signal in the transmission environment of the optical communication system is simulated. Since the degree of distortion is used to measure the distortion performance of the optical communication system during signal transmission, adjusting the target wavelength parameters according to the degree of distortion can quantitatively evaluate the compensation effect and further optimize the wavelength parameters to continuously improve the transmission quality of the optical signal. This solves the technical problem in the existing technology that it is difficult to adapt to the dynamically changing transmission environment, resulting in the inability to effectively suppress wavelength distortion, and ensures the signal transmission quality and system performance of the optical communication system. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0061] Figure 1 A flowchart illustrating an optical signal wavelength optimization method provided in this application embodiment;

[0062] Figure 2 A flowchart illustrating the process of constructing target distortion distribution data is provided in this application embodiment;

[0063] Figure 3 A flowchart illustrating the process of constructing mapping relationship data provided in this application embodiment;

[0064] Figure 4 A flowchart illustrating a wavelength parameter adjustment process provided in this application embodiment;

[0065] Figure 5 This is a schematic diagram of an optical signal wavelength optimization device provided in an embodiment of this application. Detailed Implementation

[0066] As described above, existing technologies typically employ static compensation methods (such as fixed dispersion compensation modules or pre-compensation techniques) to suppress wavelength distortion. However, these methods are ill-suited to dynamically changing transmission environments (such as temperature fluctuations and fiber optic link topology adjustments), thus failing to effectively suppress wavelength distortion. Therefore, an adaptive wavelength optimization method is urgently needed to effectively suppress the increasingly severe wavelength distortion problem in high-speed, high-capacity optical communication systems, ensuring signal transmission quality and system performance.

[0067] Through research, the inventors proposed a method for optimizing optical signal wavelengths. This method first acquires raw optical signal data in real time from the optical communication system, comprehensively obtaining the initial wavelength parameters of the optical signal. This provides accurate basic data support for subsequent wavelength analysis and compensation, ensuring the optical communication system's dynamic perception capability of the optical signal state. Then, wavelength distortion features are extracted from the raw optical signal data, and target distortion distribution data is constructed based on the distortion categories of these features. Since the distortion categories of wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion, the target distortion distribution data is used to characterize the distribution of wavelength distortion in the time or spatial dimensions. Therefore, by classifying and extracting adjacent channel interference distortion and group velocity dispersion distortion, different types of distortion sources can be accurately identified. Combined with the distribution characteristics in the time or spatial dimensions, the dynamic evolution law of wavelength distortion can be more comprehensively characterized, providing a basis for subsequent targeted compensation.

[0068] Then, a pre-trained convolutional neural network model performs in-depth analysis of the target distortion distribution data, enabling rapid identification of complex interferences in optical signal transmission (such as mutual interference between signals and irregular changes during light wave propagation) to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and parameters of the optical communication system, the mapping relationship data between the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and the complexity of the optical communication system is determined. This accurately identifies the correlation between the above-mentioned interference problems and system settings, providing a data foundation for subsequent accurate correction of the above-mentioned interference. Then, a pre-trained generative adversarial network model generates wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data. Based on the wavelength correction coefficients and the dispersion characteristics of the optical fiber material, the initial wavelength parameters are adaptively adjusted to obtain the target wavelength parameters. This achieves dynamic response to changes in the transmission environment, significantly improves the flexibility and real-time performance of the compensation strategy, and effectively addresses signal distortion problems in complex scenarios.

[0069] Then, based on the target wavelength parameters, the degree of distortion of the optical signal in the transmission environment of the optical communication system is simulated. Since the degree of distortion is used to measure the distortion performance of the optical communication system during signal transmission, adjusting the target wavelength parameters according to the degree of distortion can quantitatively evaluate the compensation effect and further optimize the wavelength parameters to continuously improve the transmission quality of the optical signal. This solves the technical problem in the existing technology that it is difficult to adapt to the dynamically changing transmission environment, resulting in the inability to effectively suppress wavelength distortion, and ensures the signal transmission quality and system performance of the optical communication system.

[0070] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0071] Method Implementation Examples

[0072] This application provides an embodiment of an optical signal wavelength optimization method. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although the flowchart shows a logical order, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0073] See Figure 1 The figure is a flowchart of an optical signal wavelength optimization method provided in an embodiment of this application, as shown below. Figure 1 As shown, the method includes the following steps:

[0074] Step S101: Acquire raw optical signal data from the optical communication system.

[0075] In step S101, the original optical signal data includes the initial wavelength parameters of the optical signal (such as wavelength range, transmission rate, etc.).

[0076] In one optional embodiment, the optical communication system can serve as the execution subject of the optical signal wavelength optimization method of this application. For ease of description, the optical communication system will be referred to as the system below.

[0077] In this embodiment of the application, the system can comprehensively acquire the initial wavelength parameters of the optical signal by acquiring raw optical signal data from the optical communication system in real time, providing accurate basic data support for subsequent wavelength analysis and compensation, and ensuring the dynamic perception capability of the optical communication system of the optical signal state.

[0078] Step S102: Extract wavelength distortion features from the original optical signal data, and construct target distortion distribution data based on the distortion categories of the wavelength distortion features.

[0079] In step S102, the distortion categories of wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion, and the target distortion distribution data is used to characterize the distribution of wavelength distortion in the time dimension or spatial dimension.

[0080] In this embodiment, the system can extract wavelength distortion features from the original optical signal data and classify them into adjacent channel interference distortion and group velocity dispersion distortion. Then, based on the distortion category of the wavelength distortion features, target distortion distribution data is constructed to characterize the distribution of wavelength distortion in the time or spatial dimensions. This allows for the accurate identification of different types of distortion sources by classifying and extracting adjacent channel interference distortion and group velocity dispersion distortion. Combined with the distribution features in the time or spatial dimensions, the system can more comprehensively characterize the dynamic evolution of wavelength distortion, thereby providing a basis for subsequent targeted compensation.

[0081] See Figure 2 The figure is a flowchart of a process for constructing target distortion distribution data according to an embodiment of this application. The process includes the following steps:

[0082] Step S1021: Extract power spectral density data from the original optical signal data;

[0083] Step S1022: Based on the power spectral density data, calculate the error between the actual power value and the ideal power value at each wavelength point of the optical signal to obtain the wavelength distortion characteristics;

[0084] Step S1023: Classify the wavelength distortion features to classify them into adjacent channel interference distortion and group velocity dispersion distortion.

[0085] Step S1024: Perform adjacent channel interference analysis on the wavelength distortion characteristics to obtain the first distortion distribution data;

[0086] Step S1025: Perform group velocity dispersion analysis on the wavelength distortion characteristics to obtain the second distortion distribution data;

[0087] Step S1026: Construct target distortion distribution data based on the first distortion distribution data and the second distortion distribution data.

[0088] Specifically, extracting wavelength distortion features in an optical communication system and classifying them for adjacent channel interference and group velocity dispersion to obtain target distortion distribution data can be achieved through the following example: First, the system takes a set of optical signal wavelength data (i.e., raw optical signal data) collected from the optical communication system, with a wavelength range of 1550nm to 1560nm and a sampling interval of 0.1nm, and extracts power spectral density data from the raw optical signal data using a spectral analyzer. For each wavelength point, the deviation (i.e., error) between its actual power value and the ideal power value is calculated to obtain the wavelength distortion features; for example, if the actual power value measured at a wavelength of 1550.5nm is -2.3dB, while the ideal power value is 0dB, then the recorded wavelength distortion feature is determined to be 2.3.

[0089] Subsequently, the system can perform classification processing for adjacent channel interference. A spectrum separation algorithm is used to identify the power leakage between adjacent channels as a feature, with a leakage threshold of 1.5 dB. If the leakage power at the wavelength of 1551.0 nm is 1.8 dB, the system can classify it as being affected by adjacent channel interference. A Fast Fourier Transform (FFT) algorithm is then used to decompose the corresponding power spectrum, extract the interference frequency components, and quantify the interference intensity, obtaining the first distortion distribution data. This first distortion distribution data shows that the interference is mainly concentrated within a 0.2 nm bandwidth. Next, the system can perform classification processing for group velocity dispersion. A dispersion compensation algorithm is used to quantify the dispersion effect by measuring the pulse broadening. If the pulse broadening at 1552.0 nm is 10 ps, ​​exceeding the threshold of 5 ps, it is classified as being affected by group velocity dispersion. The system can use the finite difference method to calculate the dispersion coefficient, obtaining a dispersion coefficient of 20 ps / nm / km, which is then used as the second distortion distribution data. Finally, the first and second distortion distribution data are integrated into the target distortion distribution data to generate a two-dimensional matrix. The rows of this two-dimensional matrix represent wavelength points, and the columns represent distortion types and values. For example, in the two-dimensional matrix, 1550.5nm corresponds to a distortion value of 2.3, and the type is adjacent channel interference. A heat map is generated using data visualization tools to visually display the distortion distribution. The analysis shows that the distortion is mainly concentrated in the range of 1550nm to 1552nm, providing a basis for subsequent optimization.

[0090] Step S103: The pre-trained convolutional neural network model performs in-depth analysis on the target distortion distribution data to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters and the parameters of the optical communication system, the mapping relationship data is determined.

[0091] In step S103, the mapping relationship data is used to characterize the mapping relationship between nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters and the complexity of the optical communication system.

[0092] In this embodiment, the system can perform in-depth analysis of target distortion distribution data using a pre-trained convolutional neural network model. It can quickly identify complex interferences in optical signal transmission (such as mutual interference between signals and irregular changes during light wave propagation) to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters combined with the parameters of the optical communication system, it determines the mapping relationship data between the nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters and the complexity of the optical communication system. It can accurately identify the correlation between the above-mentioned interference problems and system settings, providing a data basis for subsequent accurate correction of the above-mentioned interference.

[0093] See Figure 3The figure is a flowchart of a process for constructing mapping relationship data according to an embodiment of this application. The process includes the following steps:

[0094] Step S1031: Input the target distortion distribution data into the convolutional neural network model, and let the convolutional neural network model classify the target distortion distribution data into nonlinear crosstalk distortion and polarization mode dispersion distortion.

[0095] Step S1032: The nonlinear crosstalk analysis of the target distortion distribution data is performed by the convolutional neural network model to obtain the nonlinear crosstalk characteristic parameters;

[0096] Step S1033: The polarization mode dispersion analysis of the target distortion distribution data is performed by the convolutional neural network model to obtain the polarization mode dispersion characteristic parameters;

[0097] Step S1034: Based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters and optical communication system parameters, a regression analysis model is used to generate the complexity index of the optical communication system.

[0098] Step S1035: Generate mapping relationship data based on nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters and corresponding complexity indices.

[0099] Specifically, in optical communication systems, deep analysis of target distortion distribution data and the construction of mapping relationships can be achieved through the following example: The system uses a pre-trained convolutional neural network model to deeply mine wavelength distortion features from a set of target distortion distribution data with wavelengths ranging from 1545nm to 1555nm. This model contains a network model with multiple convolutional and pooling layers. The input data is the target distortion distribution data, with a matrix dimension of 50×10. Each row of the matrix represents a wavelength point, and each column represents different distortion parameters. The convolutional neural network model extracts hidden layer feature vectors from the target distortion distribution data. For example, the feature value output at 1546.4nm is 0.85, indicating the presence of significant nonlinear features. Subsequently, a convolutional neural network model was used to perform polarization mode dispersion analysis on the target distortion distribution data. A signal demodulation algorithm was employed to match the hidden layer feature vectors with a pre-defined nonlinear model. If the correlation coefficient between the feature value at 1547.2 nm and the model reached 0.9, it was determined to be dominated by nonlinear crosstalk, and its influence bandwidth was calculated to be 0.3 nm, which was then used as a polarization mode dispersion feature parameter. Following this, the convolutional neural network model was used to perform polarization mode dispersion analysis on the target distortion distribution data. A polarization state decomposition algorithm was used to quantify the polarization difference group delay. For example, a delay value of 8 ps was measured at 1548.0 nm, exceeding the threshold of 6 ps, and was classified as a feature affected by polarization mode dispersion. The polarization mode distribution was further calculated using matrix factorization, yielding a principal axis offset angle of 15 degrees, which was then used as a polarization mode dispersion feature parameter. Finally, the system can generate a complexity index of the optical communication system based on nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and parameters of the optical communication system (such as the number of channels being 40) through a regression analysis model. For example, the analysis shows that the complexity index at 1546.4nm is 3.2, which is positively correlated with the nonlinear crosstalk characteristic value of 0.85. Based on the correlation between the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and the corresponding complexity index, mapping relationship data is generated.

[0100] Step S104: The pre-trained generative adversarial network model generates wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data, and adjusts the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters.

[0101] In this embodiment, the system can generate wavelength correction coefficients corresponding to optical signals based on mapping relationship data using a pre-trained generative adversarial network model, and adaptively adjust the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters. This achieves dynamic response to changes in the transmission environment, significantly improves the flexibility and real-time performance of the compensation strategy, and effectively addresses signal distortion problems in complex scenarios.

[0102] See Figure 4The figure is a flowchart of a wavelength parameter adjustment process provided in an embodiment of the application. The process includes the following steps:

[0103] Step S1041: Input the mapping relationship data into the generative adversarial network model, and use the mapping relationship data of the generative adversarial network model to match the corresponding wavelength correction coefficients for the optical signal;

[0104] Step S1042: Correct the initial wavelength parameters based on the wavelength correction coefficient to obtain the corrected wavelength parameters;

[0105] Step S1043: Perform dispersion analysis on the corrected wavelength parameters based on the dispersion characteristics of the optical fiber material to obtain dispersion results; the dispersion results include dispersion values ​​and dispersion slopes.

[0106] Step S1044: Calculate the wavelength shift based on the dispersion value and dispersion slope;

[0107] Step S1045: Adjust the corrected wavelength parameters based on the wavelength offset to obtain the target wavelength parameters.

[0108] Specifically, in an optical communication system, generating wavelength correction coefficients for optical signals based on mapping data and adjusting the initial wavelength parameters according to these coefficients and the dispersion characteristics of the fiber material can be achieved through the following example: First, mapping data with a wavelength range of 1560nm to 1570nm is input into a pre-trained generative adversarial network (GAN) model. The GAN generates wavelength correction coefficients for the optical signals based on this data, for example, outputting a wavelength correction coefficient of 0.75 at 1562.5nm. This coefficient is then used to correct the initial wavelength parameters, resulting in the corrected wavelength parameters. The discriminator network of the GAN model is used to evaluate the feasibility of the generated correction strategy; its training iterations are set to 800, and the learning rate is 0.002. Next, the system can combine the dispersion characteristics of the optical fiber material to analyze the dispersion value and dispersion slope of the corrected wavelength parameters. For example, the dispersion value measured at 1563.0 nm is 18 ps / nm / km, and the dispersion slope is 0.08 ps / nm² / km. The adjusted wavelength offset is calculated to be 0.15 nm through numerical simulation algorithm, and the corrected wavelength parameters are adjusted based on the wavelength offset to obtain the target wavelength parameters.

[0109] Step S105: Based on the target wavelength parameters, simulate the degree of distortion of the optical signal in the transmission environment of the optical communication system.

[0110] In step S105, the degree of distortion change is used to measure the distortion performance of the optical communication system during signal transmission.

[0111] In this embodiment, the system can perform optical power leakage simulation analysis on the optical signal in the optical communication system transmission environment based on the target wavelength parameter to obtain the optical power leakage value corresponding to the optical signal; if the optical power leakage value is greater than the preset leakage value, the system calculates the first compensation gain value based on the target wavelength parameter through the power compensation model; and then determines the degree of distortion change based on the first compensation gain value.

[0112] Optionally, the system determines the degree of distortion change based on the first compensation gain value, including the following process: adjusting the target wavelength parameter based on the first compensation gain value to obtain the compensated target wavelength parameter; then, based on the compensated target wavelength parameter, performing dispersion compensation error analysis on the optical signal in the optical communication system transmission environment to obtain the compensation error corresponding to the optical signal; then, generating parameter correction coefficients based on the compensation error, and correcting the above-mentioned compensated target wavelength parameter based on the parameter correction coefficients to obtain the corrected target wavelength parameter; finally, determining the degree of distortion change based on the corrected target wavelength parameter and the initial wavelength parameter.

[0113] Specifically, in an optical communication system, simulating the distortion variation of an optical signal in the transmission environment based on the target wavelength parameter can be achieved through the following example: First, optical power leakage simulation analysis is performed on the optical signal in the transmission environment of the optical communication system. Using a power distribution analysis algorithm, a power leakage value of -2.3dBm is detected at 1564.8nm, exceeding the threshold of -3.0dBm. The system can automatically call the power compensation model, calculate the compensation gain (i.e., the first compensation gain value) as 1.2dB, and adjust the target wavelength parameter based on the first compensation gain value to obtain the compensated target wavelength parameter. Subsequently, dispersion compensation error analysis is performed on the optical signal in the transmission environment of the optical communication system. Using an error quantization algorithm, the compensation error is measured to be 0.5ps / nm at 1565.2nm. The system can use an adaptive filtering algorithm to generate a parameter correction coefficient of 0.92 based on the compensation error, automatically adjust the compensation module parameters, and correct the above-mentioned compensated target wavelength parameter based on the parameter correction coefficient to obtain the corrected target wavelength parameter. Finally, the degree of distortion variation is determined based on the corrected target wavelength parameter and the initial wavelength parameter to ensure that distortion is minimized.

[0114] Step S106: Adjust the target wavelength parameters according to the degree of distortion change in order to optimize the wavelength of the optical signal.

[0115] In this embodiment, the system can acquire wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal when the degree of distortion change is greater than or equal to a preset threshold. Then, based on the wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal, the target wavelength parameters are adjusted to obtain the adjusted target wavelength parameters. The system can then perform polarization mode crosstalk analysis on the adjusted target wavelength parameters to obtain a first signal compensation coefficient, and perform high-order dispersion effect analysis on the adjusted target wavelength parameters to obtain a second signal compensation coefficient. Finally, the system can further optimize the wavelength of the optical signal based on the first and second signal compensation coefficients to continuously improve the transmission quality of the optical signal. This solves the technical problem in the prior art where wavelength distortion cannot be effectively suppressed due to the difficulty in adapting to dynamically changing transmission environments, thus ensuring the signal transmission quality and system performance of the optical communication system.

[0116] Specifically, in optical communication systems, when the degree of distortion change does not reach a preset threshold, the target wavelength parameters can be automatically adjusted based on real-time feedback of wavelength drift interference data and temperature-dependent dispersion data. First, the system collects real-time optical signals with a wavelength range of 1555nm to 1565nm at a sampling interval of 0.05nm, generating a wavelength drift interference matrix (200×6 dimensions) (i.e., wavelength drift interference data). This matrix is ​​then input into a deep learning prediction model based on a convolutional neural network. The model undergoes 1000 training iterations with a learning rate of 0.001, predicting a wavelength drift of 0.12nm at 1557.8nm. Next, the system, combining temperature-dependent dispersion data, determines the dispersion value at 1558.2 nm to be 16.5 ps / nm / km at 25°C, and 17.0 ps / nm / km at 30°C, with a dispersion slope of 0.07 ps / nm² / km. The system uses a polynomial fitting algorithm to calculate the dispersion shift caused by temperature change as 0.03 nm. Then, based on the wavelength drift of 0.12 nm and the dispersion shift of 0.03 nm, the system automatically updates the wavelength parameters to obtain the adjusted target wavelength parameters. Subsequently, the system performs polarization mode crosstalk analysis on the adjusted target wavelength parameters. Using an optical power monitoring algorithm, a power attenuation value of -1.8 dBm is detected at 1559.5 nm, which does not exceed the threshold of -2.5 dBm. The gain adjustment model is invoked to calculate a gain compensation value of 0.9 dB (i.e., the first signal compensation coefficient), and a power distribution curve is generated and stored in the system log. If temperature changes cause missing dispersion data, the system can perform high-order dispersion effect analysis on the adjusted target wavelength parameters, automatically invoke a time-series-based interpolation algorithm, and combine historical data to complete the dispersion value at 1560.1 nm to 16.8 ps / nm / km (i.e., the second signal compensation coefficient). Finally, for the combined effects of wavelength drift and dispersion, the system adopts an adaptive equalization algorithm, calculates an equalization coefficient of 0.88 at 1561.3 nm, and automatically adjusts the transmission module parameters to further optimize the wavelength of the optical signal, ensuring signal integrity.

[0117] In one alternative embodiment, the system can generate a compensation mechanism design based on the optimized wavelength parameters, perform signal compensation processing for polarization mode crosstalk and higher-order dispersion effects, and update the wavelength parameters of the optical signal through multiple rounds of iterative calculation.

[0118] Specifically, in optical communication systems, for signal compensation against polarization mode crosstalk (PMC) and higher-order dispersion effects, the system first acquires signals with wavelengths ranging from 1545 nm to 1555 nm at a sampling interval of 0.04 nm, generating a PMC matrix (250×8 dimensions). Principal component analysis (PCA) is used to extract the main interference components, and the PMC value at 1547.6 nm is calculated to be 0.15 dB. Next, the system uses a random forest regression model, trained for 800 iterations with a learning rate of 0.002, to predict a signal phase shift of 0.08 rad at 1548.3 nm caused by PMC. Subsequently, to address higher-order dispersion effects, the system measured the second-order dispersion value at 1549.1 nm at 25°C to be 0.045 ps / nm² / km, which increased to 0.048 ps / nm² / km when the temperature rose to 28°C. Using the Lagrange interpolation algorithm, the dispersion increment caused by temperature change was calculated to be 0.002 ps / nm² / km, and the result was stored in the parameter database. To address signal distortion caused by polarization mode crosstalk, the system invoked an adaptive filtering algorithm, calculating a filtering coefficient of 0.92 at 1550.7 nm, generating a polarization compensation curve, and storing it in the log. If higher-order dispersion data was missing, the system used Kalman filtering time-series prediction to complete the dispersion value at 1551.4 nm, resulting in 0.046 ps / nm² / km. Finally, the system integrated polarization mode crosstalk and higher-order dispersion effects, employing a gradient descent optimization algorithm to iteratively calculate a compensation coefficient of 0.85 at 1552.9 nm, updating the wavelength parameters of the optical signal.

[0119] Furthermore, the system can analyze the changing trends of transmission quality indicators affected by modulation format and optical signal bandwidth in the optimization parameter set, thereby obtaining the optimization parameter set.

[0120] Specifically, in optical communication systems, based on optimized wavelength parameters, the system automatically generates an optimized parameter set by analyzing the impact of modulation format and optical signal bandwidth on transmission quality indicators. First, the system collects signal data within the wavelength range of 1560nm to 1570nm at sampling intervals of 0.05nm. For different modulation formats such as QPSK and 16-QAM, the bit error rate (BER) trend is calculated. The results show that at 1562.5nm, the BER is 1.2e-4 under QPSK modulation, while it increases to 3.5e-4 under 16-QAM, indicating that higher-order modulation formats are more sensitive to noise. Subsequently, the system uses a Support Vector Machine (SVM) algorithm with a radial basis function (RBF) kernel and 500 training samples. The prediction shows that when the optical signal bandwidth expands from 10GHz to 20GHz at 1564.8nm, the signal eye diagram opening decreases by 0.18, indicating that bandwidth expansion leads to a decrease in transmission quality. Next, the system analyzes the spectral characteristics using Fourier transform, measuring a spectral width of 0.12 nm at 1566.3 nm when the bandwidth is 15 GHz. It then uses a linear regression model to predict the impact of bandwidth on signal jitter, calculating the jitter increment to be 0.03 ps. Further, considering the effects of modulation format and bandwidth, the system invokes a genetic algorithm at 1568.9 nm, setting the population size to 100, the number of iterations to 200, and optimizing the modulation depth parameter to 0.75, generating an optimized parameter set.

[0121] Furthermore, the system can adjust the wavelength configuration strategy of the optical module by optimizing the parameter set, monitor the changes in transmission quality indicators such as optical amplifier noise and dispersion compensation error in real time, and determine whether the system complexity meets the preset path reduction requirements.

[0122] Specifically, in an optical communication system, to optimize the wavelength configuration strategy of the optical module based on the set of parameters, the system first automatically collects signal data in the wavelength range of 1550nm to 1560nm with a sampling interval of 0.1nm. Considering the impact of optical amplifier noise, the signal-to-noise ratio (SNR) is calculated. Analysis shows that the SNR is 25.6dB at 1553.2nm, while it drops to 23.8dB at 1555.7nm, indicating that noise accumulation increases with wavelength shift. Subsequently, the system calls the random forest algorithm, setting the number of trees to 150 and the maximum depth to 20. It predicts that when the noise power increases from 0.02mW to 0.05mW at 1558.4nm, the SNR will further decrease by 1.5dB. Analysis shows that the interference of noise on transmission quality increases non-linearly. Next, addressing the dispersion compensation error, the system calculated the residual dispersion at 1551.9nm using a time-domain analysis algorithm, finding it to be 12.3 ps / nm. A multinomial regression model was used to predict the impact of the error on signal distortion, yielding a distortion increment of 0.07. Analysis of historical data confirmed a positive correlation between error accumulation and wavelength shift. To achieve real-time monitoring, a neural network-based prediction model was deployed at 1554.6nm, with 3 hidden layers and 64 nodes per layer. Real-time calculation of transmission quality indicators revealed a signal delay increment of 0.04 ps due to the dispersion compensation error, automatically generating adjustment suggestions and updating the parameter library. Simultaneously, the system assessed complexity, calculating resource utilization and algorithm execution time. The average processing time of the current monitoring logic was found to be 0.32 seconds, lower than the preset 0.5-second threshold. Based on business requirements analysis, the system adjusted the monitoring frequency from once per second to 0.8 times per second to further reduce resource load and ensure compliance with the requirement to reduce path complexity. Relevant data was synchronously stored in the system log, forming a complete closed loop from parameter adjustment to complexity optimization.

[0123] Furthermore, if the system fails to meet the preset path requirements when the system complexity reduction is achieved, it can re-analyze the interaction between fiber dispersion interaction and optical signal bandwidth through a long short-term memory network to obtain a new correction strategy and determine the final wavelength optimization parameters.

[0124] Specifically, in optical communication systems, when the reduction in system complexity fails to meet the preset path requirements, the system automatically initiates a Long Short-Term Memory (LSTM) network for deep analysis to evaluate the interaction between fiber dispersion and optical signal bandwidth. First, the system collects optical signal data at 0.2 nm intervals within the wavelength range of 1548 nm to 1552 nm, calculating the correlation between dispersion values ​​and bandwidth occupancy. Analysis shows that at 1549.5 nm, the dispersion value is 15.7 ps / nm, and the bandwidth occupancy is 82.3%. However, at 1551.0 nm, the dispersion value increases to 16.4 ps / nm, and the bandwidth occupancy decreases to 79.8%, indicating that increased dispersion constrains bandwidth resources. Next, the system deploys an LSTM model with a time step of 10 and 128 hidden units. Training the model using historical data predicts that if the bandwidth occupancy is adjusted to 80.5% at 1550.3 nm, the dispersion value may decrease to 15.9 ps / nm. Analysis shows a non-linear correlation between bandwidth and dispersion. Subsequently, the system generates a correction strategy based on the prediction results, optimizes the wavelength parameters using the gradient descent algorithm, sets the learning rate to 0.01, and performs 200 iterations. The calculation shows that adjusting the signal bandwidth to 81.2% at 1549.8nm can control the dispersion value within 15.5ps / nm. Finally, the system combines the optimized wavelength parameters with service requirements and automatically verifies their impact on transmission stability. Simulation analysis confirms that the bit error rate decreases from 0.00012 to 0.00009 after adjustment, meeting service reliability requirements.

[0125] The method provided in this application achieves comprehensive acquisition of the initial wavelength parameters of optical signals, providing accurate basic data support for subsequent wavelength analysis and compensation, and ensuring the dynamic perception capability of the optical communication system of the optical signal state. It also enables the accurate identification of different types of distortion sources by classifying and extracting adjacent channel interference distortion and group velocity dispersion distortion. Combined with the distribution characteristics in time or space dimensions, it can more comprehensively characterize the dynamic evolution law of wavelength distortion, providing a basis for subsequent targeted compensation. Furthermore, it enables the rapid identification of complex interferences in optical signal transmission (such as mutual interference between signals and irregular changes during light wave propagation) to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and parameters of the optical communication system, it determines the mapping between the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters, and the complexity of the optical communication system. The mapping relationship data can accurately identify the correlation between the aforementioned interference problems and system settings, providing a data foundation for subsequent accurate correction of the interference. Then, a pre-trained generative adversarial network model generates wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data. The initial wavelength parameters are adaptively adjusted according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters. This achieves dynamic response to changes in the transmission environment, significantly improving the flexibility and real-time performance of the compensation strategy and effectively addressing signal distortion problems in complex scenarios. Then, the target wavelength parameters are adjusted according to the degree of distortion change, which can quantitatively evaluate the compensation effect and further optimize the wavelength parameters to continuously improve the transmission quality of the optical signal. This solves the technical problem in existing technologies where wavelength distortion cannot be effectively suppressed due to the difficulty in adapting to dynamically changing transmission environments, ensuring the signal transmission quality and system performance of the optical communication system.

[0126] Device Examples

[0127] This application provides an optical signal wavelength optimization device, wherein... Figure 5 This is a schematic diagram of the structure of an optical signal wavelength optimization device provided in an embodiment of this application, as shown below. Figure 5 As shown, the device includes: a data acquisition module 11, a feature extraction module 12, an analysis module 13, an adjustment module 14, a simulation module 15, and an optimization module 16. From Figure 5 You can see the connections between several modules.

[0128] The data acquisition module 11 is used to acquire raw optical signal data from the optical communication system; the raw optical signal data includes the initial wavelength parameters of the optical signal.

[0129] The feature extraction module 12 is used to extract wavelength distortion features from the original optical signal data and construct target distortion distribution data based on the distortion categories of the wavelength distortion features. The distortion categories of the wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion. The target distortion distribution data is used to characterize the distribution of wavelength distortion in the time dimension or spatial dimension.

[0130] Analysis module 13 is used to perform in-depth analysis of the target distortion distribution data by a pre-trained convolutional neural network model to obtain nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters. Based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters and the parameters of the optical communication system, the mapping relationship data is determined. The mapping relationship data is used to characterize the mapping relationship between the nonlinear crosstalk characteristic parameters and polarization mode dispersion characteristic parameters and the complexity of the optical communication system.

[0131] The adjustment module 14 is used to generate wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data from the pre-trained generative adversarial network model, and adjust the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameters.

[0132] The simulation module 15 is used to simulate the degree of distortion of the optical signal in the transmission environment of the optical communication system based on the target wavelength parameter; the degree of distortion is used to measure the distortion performance of the optical communication system during signal transmission.

[0133] The optimization module 16 is used to adjust the target wavelength parameters according to the degree of distortion change in order to optimize the wavelength of the optical signal.

[0134] Optionally, the optimization module includes:

[0135] The acquisition unit is used to acquire wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal if the degree of distortion change is greater than or equal to a preset threshold.

[0136] The parameter adjustment unit is used to adjust the target wavelength parameters based on the wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal, so as to obtain the adjusted target wavelength parameters.

[0137] The compensation analysis unit is used to perform polarization mode crosstalk analysis on the adjusted target wavelength parameters to obtain the first signal compensation coefficient, and to perform higher-order dispersion effect analysis on the adjusted target wavelength parameters to obtain the second signal compensation coefficient.

[0138] The optimization unit is used to optimize the wavelength of the optical signal based on the first signal compensation coefficient and the second signal compensation coefficient.

[0139] Optionally, the feature extraction module includes:

[0140] An extraction unit is used to extract power spectral density data from the raw optical signal data;

[0141] The first calculation unit is used to calculate the error between the actual power value and the ideal power value of the optical signal at each wavelength point based on the power spectral density data, and to obtain the wavelength distortion characteristics.

[0142] The first classification processing unit is used to classify the wavelength distortion features into adjacent channel interference distortion and group velocity dispersion distortion.

[0143] The interference analysis unit is used to perform adjacent channel interference analysis on wavelength distortion characteristics to obtain the first distortion distribution data.

[0144] The first dispersion analysis unit is used to perform group velocity dispersion analysis on wavelength distortion characteristics to obtain second distortion distribution data.

[0145] The construction unit is used to construct the target distortion distribution data based on the first distortion distribution data and the second distortion distribution data.

[0146] Optionally, the analysis module includes:

[0147] The second classification processing unit is used to input the target distortion distribution data into the convolutional neural network model, and the convolutional neural network model performs classification processing on the target distortion distribution data to classify the target distortion distribution data into nonlinear crosstalk distortion and polarization mode dispersion distortion.

[0148] The nonlinear crosstalk analysis unit is used to perform nonlinear crosstalk analysis on the target distortion distribution data using a convolutional neural network model to obtain nonlinear crosstalk characteristic parameters.

[0149] The polarization mode dispersion analysis unit is used to perform polarization mode dispersion analysis on the target distortion distribution data using a convolutional neural network model, and obtain polarization mode dispersion characteristic parameters.

[0150] The first generation unit is used to generate the complexity index of the optical communication system based on the nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters and the parameters of the optical communication system through a regression analysis model.

[0151] The second generation unit is used to generate mapping relationship data based on nonlinear crosstalk characteristic parameters, polarization mode dispersion characteristic parameters and corresponding complexity exponents.

[0152] Optionally, the adjustment module includes:

[0153] The matching unit is used to input the mapping relationship data into the generative adversarial network model, and the generative adversarial network model matches the corresponding wavelength correction coefficients for the optical signal based on the mapping relationship data.

[0154] The correction unit is used to correct the initial wavelength parameters based on the wavelength correction coefficient to obtain the corrected wavelength parameters;

[0155] The second dispersion analysis unit is used to perform dispersion analysis on the corrected wavelength parameters based on the dispersion characteristics of the optical fiber material, and obtain dispersion results; the dispersion results include dispersion values ​​and dispersion slopes.

[0156] The second calculation unit is used to calculate the wavelength offset based on the dispersion value and the dispersion slope;

[0157] The first adjustment unit is used to adjust the corrected wavelength parameters based on the wavelength offset to obtain the target wavelength parameters.

[0158] Optionally, the simulation module includes:

[0159] The simulation analysis unit is used to perform optical power leakage simulation analysis on optical signals in the transmission environment of optical communication systems based on target wavelength parameters, and to obtain the optical power leakage value corresponding to the optical signal.

[0160] The third calculation unit is used to calculate the first compensation gain value based on the target wavelength parameter through the power compensation model if the optical power leakage value is greater than the preset leakage value.

[0161] The determination unit is used to determine the degree of distortion change based on the first compensation gain value.

[0162] Optionally, the determined unit includes:

[0163] The second adjustment unit is used to adjust the target wavelength parameters based on the first compensation gain value to obtain the compensated target wavelength parameters.

[0164] The dispersion compensation error analysis unit is used to perform dispersion compensation error analysis on the optical signal in the transmission environment of the optical communication system based on the compensated target wavelength parameters, and to obtain the corresponding compensation error of the optical signal.

[0165] The correction unit is used to generate parameter correction coefficients based on the compensation error, and to correct the above-compensated target wavelength parameters based on the parameter correction coefficients to obtain the corrected target wavelength parameters.

[0166] The degree determination unit is used to determine the degree of distortion change based on the corrected target wavelength parameters and the initial wavelength parameters.

[0167] Storage Media Examples

[0168] This application provides a computer-readable storage medium storing a program, which, when executed by a processor, implements some or all of the steps in the optical signal wavelength optimization method described in the foregoing method embodiments of this application. The storage medium can be any medium capable of storing program code, such as a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0169] Processor Implementation

[0170] This application provides a processor for running a program, wherein, during program execution, some or all of the steps in the optical signal wavelength optimization method described in the foregoing method embodiments are performed.

[0171] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiments. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate. The components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment solution according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0172] The above is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of wavelength optimization of an optical signal, characterized by, The method comprises the following steps: collecting original optical signal data from an optical communication system; the original optical signal data contains initial wavelength parameters of an optical signal; extracting wavelength distortion features from the original optical signal data, and constructing target distortion distribution data according to distortion categories of the wavelength distortion features; the distortion categories of the wavelength distortion features include adjacent channel interference distortion and group velocity dispersion distortion, and the target distortion distribution data is used to represent the distribution of wavelength distortion in the time dimension or the space dimension; performing deep analysis on the target distortion distribution data by a pre-trained convolutional neural network model to obtain nonlinear crosstalk feature parameters and polarization mode dispersion feature parameters, and determining mapping relationship data according to the nonlinear crosstalk feature parameters, the polarization mode dispersion feature parameters and parameters of the optical communication system; the mapping relationship data is used to represent the mapping relationship between the nonlinear crosstalk feature parameters, the polarization mode dispersion feature parameters and the complexity of the optical communication system; generating wavelength correction coefficients corresponding to the optical signal based on the mapping relationship data by a pre-trained generative adversarial network model, and adjusting the initial wavelength parameters according to the wavelength correction coefficients and the dispersion characteristics of fiber materials to obtain target wavelength parameters; simulating the distortion change degree of the optical signal in the transmission environment of the optical communication system based on the target wavelength parameters; the distortion change degree is used to measure the distortion performance of the optical communication system in the signal transmission process; adjusting the target wavelength parameters according to the distortion change degree to optimize the wavelength of the optical signal.

2. The method of claim 1, wherein, The adjusting the target wavelength parameters according to the distortion change degree to optimize the wavelength of the optical signal comprises: if the distortion change degree is greater than or equal to a preset threshold, obtaining wavelength drift interference data and temperature-dependent dispersion data corresponding to the optical signal; adjusting the target wavelength parameters based on the wavelength drift interference data and the temperature-dependent dispersion data corresponding to the optical signal to obtain adjusted target wavelength parameters; performing polarization mode crosstalk analysis on the adjusted target wavelength parameters to obtain a first signal compensation coefficient, and performing high-order dispersion effect analysis on the adjusted target wavelength parameters to obtain a second signal compensation coefficient; optimizing the wavelength of the optical signal according to the first signal compensation coefficient and the second signal compensation coefficient.

3. The method of claim 1, wherein, The extracting the wavelength distortion features from the original optical signal data and constructing target distortion distribution data according to the distortion categories of the wavelength distortion features comprises: extracting power spectral density data from the original optical signal data; calculating the error between the actual power value and the ideal power value of each wavelength point of the optical signal based on the power spectral density data to obtain the wavelength distortion features; classifying the wavelength distortion features to classify the wavelength distortion features into adjacent channel interference distortion and group velocity dispersion distortion; performing adjacent channel interference analysis on the wavelength distortion features classified as adjacent channel interference distortion to obtain first distortion distribution data; Perform group velocity dispersion analysis on the wavelength distortion characteristics classified as group velocity dispersion distortion to obtain second distortion distribution data; Based on the first distortion distribution data and the second distortion distribution data, the target distortion distribution data is constructed.

4. The method of claim 1, wherein, The target distortion distribution data is input into the convolutional neural network model, and the convolutional neural network model classifies the target distortion distribution data to classify the target distortion distribution data into nonlinear crosstalk distortion and polarization mode dispersion distortion; The target distortion distribution data classified as nonlinear crosstalk distortion is analyzed by the convolutional neural network model to obtain nonlinear crosstalk feature parameters; The target distortion distribution data classified as polarization mode dispersion distortion is analyzed by the convolutional neural network model to obtain polarization mode dispersion feature parameters; The complexity index of the optical communication system is generated based on the nonlinear crosstalk feature parameters, the polarization mode dispersion feature parameters and the parameters of the optical communication system through a regression analysis model; Based on the nonlinear crosstalk feature parameters, the polarization mode dispersion feature parameters and the corresponding complexity index, the mapping relationship data is generated. The wavelength correction coefficient corresponding to the optical signal is generated based on the mapping relationship data by the pre-trained generative adversarial network model, and the initial wavelength parameter is adjusted according to the wavelength correction coefficient and the dispersion characteristics of the optical fiber material to obtain the target wavelength parameter, including:

5. The method of claim 1, wherein, The mapping relationship data is input into the generative adversarial network model, and the generative adversarial network model matches the corresponding wavelength correction coefficient based on the mapping relationship data for the optical signal; Based on the wavelength correction coefficient, the initial wavelength parameter is corrected to obtain a corrected wavelength parameter; Based on the dispersion characteristics of the optical fiber material, dispersion analysis is performed on the corrected wavelength parameter to obtain a dispersion result; the dispersion result includes a dispersion value and a dispersion slope; Based on the dispersion value and the dispersion slope, a wavelength offset is calculated; Based on the wavelength offset, the corrected wavelength parameter is adjusted to obtain the target wavelength parameter. Based on the target wavelength parameter, the distortion change degree of the optical signal in the optical communication system transmission environment is simulated, including:

6. The method of claim 5, wherein, Based on the target wavelength parameter, optical power leakage simulation analysis is performed on the optical signal in the optical communication system transmission environment to obtain an optical power leakage value corresponding to the optical signal; If the optical power leakage value is greater than a preset leakage value, a first compensation gain value is calculated based on the target wavelength parameter through a power compensation model; The distortion change degree is determined according to the first compensation gain value. The distortion change degree is determined according to the first compensation gain value, including:

7. The method of claim 6, wherein, ​ adjust the target wavelength parameter based on the first compensation gain value to obtain a compensated target wavelength parameter; perform dispersion compensation error analysis on the optical signal in the optical communication system transmission environment based on the compensated target wavelength parameter to obtain a compensation error corresponding to the optical signal; generate a parameter correction coefficient based on the compensation error and correct the compensated target wavelength parameter based on the parameter correction coefficient to obtain a corrected target wavelength parameter; determine the distortion variation degree based on the corrected target wavelength parameter and the initial wavelength parameter.

8. An optical signal wavelength optimization device, characterized in that, comprises: a data acquisition module configured to acquire original optical signal data from an optical communication system; the original optical signal data comprises an initial wavelength parameter of an optical signal; a feature extraction module configured to extract a wavelength distortion feature from the original optical signal data and construct target distortion distribution data according to a distortion category of the wavelength distortion feature; the distortion category of the wavelength distortion feature comprises adjacent channel interference distortion and group velocity dispersion distortion, and the target distortion distribution data is used to represent a distribution of wavelength distortion in a time dimension or a space dimension; an analysis module configured to perform deep analysis on the target distortion distribution data by a pre-trained convolutional neural network model to obtain a nonlinear crosstalk feature parameter and a polarization mode dispersion feature parameter, and determine mapping relationship data according to the nonlinear crosstalk feature parameter, the polarization mode dispersion feature parameter, and parameters of the optical communication system; the mapping relationship data is used to represent a mapping relationship between the nonlinear crosstalk feature parameter, the polarization mode dispersion feature parameter, and complexity of the optical communication system; an adjustment module configured to generate a wavelength correction coefficient corresponding to the optical signal based on the mapping relationship data by a pre-trained generative adversarial network model, and adjust the initial wavelength parameter based on the wavelength correction coefficient and dispersion characteristics of an optical fiber material to obtain a target wavelength parameter; a simulation module configured to simulate a distortion variation degree of the optical signal in the optical communication system transmission environment based on the target wavelength parameter; the distortion variation degree is used to measure distortion performance of the optical communication system in a signal transmission process; an optimization module configured to adjust the target wavelength parameter according to the distortion variation degree to optimize a wavelength of the optical signal.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and when the computer program is run by a processor, an optical signal wavelength optimization method according to any one of claims 1-7 is implemented.

10. A processor, comprising: A computer program is used to run, and when the computer program is run, an optical signal wavelength optimization method according to any one of claims 1-7 is executed.

Images