100G single-polarization coherent light real-time communication method and system facing lightning disturbance

Abstract

The invention relates to the technical field of optical fiber communication, and provides a lightning stroke disturbance-oriented 100G single-polarization coherent light real-time communication method and system, and the method comprises the steps: detecting lightning stroke electromagnetic pulse time domain characteristics and stress wave propagation characteristics through an asymmetric sliding window, predicting a phase error through an exponential attenuation model, and carrying out the calculation of the phase error; acquiring stress wave delay by using a propagation velocity model; time sequence alignment is carried out, and a layered pre-compensation architecture is adopted to output compensation phase parameters; determining the position of a lightning stroke point according to the stress wave delay, reconstructing a three-dimensional stress field, converting the three-dimensional stress field into birefringence tensor distribution, and reversely solving to obtain a polarization control parameter; polarization control parameters are optimized through parameterization rapid estimation and FPGA parallel processing; and an asymmetric convergence control strategy is adopted to adjust compensation parameters and output a final communication signal. According to the invention, the signal recovery precision and the real-time processing capability of the 100G single-polarization coherent optical communication system in a lightning stroke environment are improved, and the environmental disturbance resistance requirement of high-speed coherent optical communication is met.

Description

Technical Field

[0001] This invention relates to the field of optical fiber communication technology, and in particular to a 100G single-polarization coherent optical real-time communication method and system for lightning strike disturbances. Background Technology

[0002] 100G single-polarization coherent optical communication technology, as a core technology of modern high-speed optical communication, is widely used in critical communication infrastructures such as backbone network transmission, data center interconnection, and submarine optical cables. Here, "G" is the unit of optical signal transmission rate, representing the number of gigabits per second (Gbps), used to measure the performance of an optical communication system; "100G" indicates that this single-polarization coherent optical communication technology supports a data transmission rate of 100 Gbps. This technology achieves high spectral efficiency and long-distance transmission capabilities through coherent detection and digital signal processing. However, in practical applications, 100G single-polarization coherent optical systems are susceptible to various communication failures such as phase errors, polarization drift, and signal quality degradation due to factors such as lightning electromagnetic pulse interference, optical cable stress changes, and environmental disturbances. Lightning disturbance, as a major environmental threat to optical communication systems, exerts complex spatiotemporal coupling effects on optical signal transmission through electromagnetic pulse and stress wave propagation mechanisms.

[0003] In existing technologies, the anti-disturbance processing of 100G single-polarization coherent optical communication systems mainly employs adaptive equalization and polarization control methods to achieve basic signal recovery. However, existing methods do not adequately consider the inherent correlation between the complex physical mechanisms of lightning disturbances and the characteristics of optical signal transmission. They struggle to organically integrate the temporal characteristics of lightning electromagnetic pulses, the spatial characteristics of stress wave propagation, and multi-dimensional parameters such as phase and polarization of the coherent optical system, resulting in the inability to achieve high-precision, high-real-time adaptive compensation for lightning disturbances. Specifically, traditional methods lack effective mechanisms for handling the temporal uncertainties of lightning disturbances, cannot accurately establish a physical correlation model between stress wave propagation delay and polarization state changes, and have significant shortcomings in multi-parameter collaborative optimization and real-time processing capabilities, making it difficult to meet the requirements of 100G high-speed coherent optical communication for anti-lightning disturbances. Summary of the Invention

[0004] In view of this, the present invention proposes a 100G single-polarization coherent optical real-time communication method and system for lightning strike disturbances. It solves the problem that existing methods do not adequately consider the inherent correlation between the complex physical mechanism of lightning strike disturbances and the characteristics of optical signal transmission, and it is difficult to organically integrate the temporal characteristics of lightning electromagnetic pulses, the spatial characteristics of stress wave propagation and the phase, polarization and other multi-dimensional parameters of coherent optical systems, which leads to the inability to achieve high-precision and high-real-time adaptive compensation for lightning strike disturbances.

[0005] The technical solution of this invention is implemented as follows: On one hand, this invention provides a 100G single-polarization coherent optical real-time communication method oriented towards lightning strike disturbances, comprising the following steps: Asymmetric sliding window was used to detect 100G single-polarization coherent optical signals to obtain the time-domain characteristics of lightning electromagnetic pulses, phase error baseline, and stress wave propagation characteristics. The time-domain characteristics of the lightning electromagnetic pulse are input into the exponential decay model to obtain the predicted phase error value, and the stress wave propagation characteristics are input into the propagation velocity model to obtain the stress wave propagation delay. The phase error prediction value and the phase error baseline are time-aligned using a lightning electromagnetic pulse characteristic spectrum matching algorithm to obtain time-aligned data. A hierarchical pre-compensation architecture is then used to process the time-aligned data and output the compensated phase parameters. The location and intensity of the lightning strike are determined based on the stress wave propagation delay. The three-dimensional stress field distribution is reconstructed by combining the optical cable geometric parameters. The three-dimensional stress field distribution is converted into a birefringence tensor distribution. The birefringence tensor distribution is then solved in reverse to obtain the polarization control parameters. The polarization control parameters are simplified by inputting a parameterized fast estimation algorithm, and the birefringence data of multiple fiber segments are processed simultaneously through an FPGA parallel processing architecture to output optimized polarization control parameters. An exponential decay model for lightning disturbance is adopted. The convergence parameters of the compensation phase parameter and the optimized polarization control parameter are adjusted according to the decay rate. The final communication signal is output through an asymmetric convergence control strategy.

[0006] Based on the above technical solutions, preferably, the asymmetric sliding window includes a short window and a long window. The short window is used to obtain the time-domain characteristics of the lightning electromagnetic pulse, the long window is used to obtain the phase error baseline, and the Stokes parameters of the polarization state are detected to obtain the stress wave propagation characteristics. The short window is set to a length of 8 to 16 symbol periods, and the long window is set to a length of 64 to 128 symbol periods. The short window samples the optical signal in a high-frequency update mode to capture the rapid change characteristics of the lightning electromagnetic pulse, and the long window samples the optical signal in a low-frequency stable mode to establish a statistical baseline for the phase error. By comparing the deviation between the detection results of the short window and the statistical baseline of the long window, the sudden temporal characteristics of the lightning electromagnetic pulse are identified. At the same time, a four-way parallel detection method is used to monitor the instantaneous change rate of the polarization state Stokes parameters S0, S1, S2, and S3 respectively to obtain the stress wave propagation characteristics.

[0007] Based on the above technical solutions, preferably, the identification of the sudden time-domain characteristics of lightning electromagnetic pulses includes: Calculate the rising slope and peak amplitude of the signal power within the short window. When the rising slope exceeds 10 times the normal signal change rate and the peak amplitude exceeds 5 times the average power, it is determined to be a lightning electromagnetic pulse characteristic. The acquisition of stress wave propagation characteristics includes: The first-order difference values ​​of the Stokes parameters of the four polarization states are calculated in real time, and the Stokes parameter change vector is established. When the magnitude of the Stokes parameter change vector exceeds a preset threshold and the change direction conforms to the polarization rotation law of stress wave propagation, the propagation speed and attenuation coefficient of the stress wave are extracted to obtain the stress wave propagation characteristics. The formula for calculating the magnitude of the Stokes parameter change vector is: ; in, The magnitude of the Stokes parameter variation vector; For the first First-order difference values ​​of the Stokes parameters; For the first The weighting coefficients for each Stokes parameter; These are the gradient compensation coefficients; For the first Spatial gradient of the Stokes parameters.

[0008] Based on the above technical solutions, preferably, the step of inputting the time-domain characteristics of the lightning electromagnetic pulse into an exponential decay model to obtain a predicted phase error value, and inputting the stress wave propagation characteristics into a propagation velocity model to obtain the stress wave propagation delay, includes: A two-parameter exponential decay function is established based on the peak intensity and rise time of the lightning electromagnetic pulse. By fitting the energy dissipation law of the lightning disturbance in the optical fiber, the time-domain evolution trend of the phase error is predicted, and the predicted value of the phase error is obtained. Based on the elastic modulus, density distribution, and burial depth parameters of the optical cable, a layered medium model for stress wave propagation is established to obtain the propagation velocity model. According to the refraction and reflection characteristics of stress waves in different medium layers, the time delay of stress waves propagating from the lightning strike point to the optical fiber is calculated, and the propagation delay is corrected by combining the attenuation characteristics of stress waves.

[0009] Based on the above technical solutions, preferably, the two-parameter exponential decay function includes a first parameter and a second parameter; The first parameter is the lightning strike intensity attenuation coefficient, which is determined based on the relationship between the peak power of the lightning electromagnetic pulse and the nonlinear response of the optical fiber material; The second parameter is a time constant, which is determined by the convolution result of the spectral characteristics of the lightning electromagnetic pulse and the dispersion characteristics of the optical fiber. The layered medium model divides the optical cable burial environment into three propagation media: soil layer, protective layer, and optical cable core layer. The propagation speed of stress waves in each layer and the interface transmission coefficient are calculated respectively. The corrected stress wave propagation delay is obtained by accumulating the time delay of the multi-layer propagation path and multiplying the energy attenuation. The propagation speed of the soil layer is dynamically corrected according to the soil moisture content and compaction degree.

[0010] Based on the above technical solutions, preferably, the step of using a lightning electromagnetic pulse characteristic spectrum matching algorithm to perform time-series alignment of the predicted phase error value and the phase error baseline to obtain time-series aligned data, and then using a hierarchical pre-compensation architecture to process the time-series aligned data and output compensated phase parameters, includes: The lightning electromagnetic pulse feature spectrum matching algorithm constructs a spectrum fingerprint by extracting the main frequency component and harmonic components of the lightning electromagnetic pulse, and performs cross-correlation operation on the frequency domain characteristics of the spectrum fingerprint and the phase error baseline to determine the timing offset. The hierarchical pre-compensation architecture adopts a series-parallel hybrid structure. The first layer of the hierarchical pre-compensation architecture performs coarse compensation based on the lightning strike intensity, and the second layer of the hierarchical pre-compensation architecture performs fine compensation based on the residual error. The coarse compensation adopts a fast-response feedforward control method, and selects a preset compensation gain coefficient according to the lightning strike intensity level; The fine compensation adopts an adaptive feedback control method, which dynamically adjusts the compensation parameters by monitoring the statistical characteristics of the residual phase error in real time, and sets a buffer delay mechanism between the first layer and the second layer to avoid compensation conflicts.

[0011] Based on the above technical solutions, preferably, the spectral fingerprint includes three feature parameters, namely the center frequency, spectral width, and spectral asymmetry of the lightning electromagnetic pulse. The cross-correlation operation adopts a sliding window correlation detection method. When the correlation coefficient is greater than 0.8, the optimal timing offset is determined. The preset compensation gain coefficient of the coarse compensation is divided into three levels: weak, medium, and strong, according to the intensity of the lightning strike, corresponding to gain values ​​of 0.3, 0.6, and 0.9, respectively. The fine-compensation adaptive feedback control employs a minimum mean square error algorithm, and the step size factor of the minimum mean square error algorithm is dynamically adjusted according to the variance of the residual phase error. The buffer delay mechanism includes a delay time of 2 to 5 symbol periods. By monitoring the correlation between coarse compensation and fine compensation outputs, it determines whether there is a compensation conflict and adjusts the delay length accordingly. The calculation formula for the minimum mean square error algorithm is as follows: ; in, For the first The adaptive step size factor for the next iteration; This is the initial step size; The target variance; For the first The residual error variance of the next iteration; To prevent the elimination of zero factors; It is a correlation modulator; This is the signal correlation coefficient.

[0012] Based on the above technical solutions, preferably, the step of determining the lightning strike location and intensity based on the stress wave propagation delay, reconstructing the three-dimensional stress field distribution in conjunction with the optical cable geometric parameters, converting the three-dimensional stress field distribution into a birefringence tensor distribution, and performing inverse solving on the birefringence tensor distribution to obtain polarization control parameters includes: The stress wave propagation delay is input into the triangulation algorithm to determine the location and intensity of the lightning strike point. The space around the optical cable is discretized into stress calculation units using the finite element mesh generation method. The stress propagation equations are established in combination with the optical cable geometric parameters to reconstruct the three-dimensional stress field distribution. The three-dimensional stress field distribution is converted into a birefringence tensor distribution using the photoelastic coefficient matrix. The birefringence tensor distribution is then solved in reverse using an optimization iterative algorithm to obtain the polarization control parameters.

[0013] Based on the above technical solutions, preferably, the triangulation positioning algorithm determines the location and intensity of the lightning strike point by using the hyperbola intersection method based on the time difference of arrival of stress waves at multiple monitoring points. The finite element mesh generation method uses hexahedral mesh elements, with high-density mesh generation in the vicinity of the optical cable and low-density mesh generation in the far-away area. The elastic-optic coefficient matrix is ​​a symmetric matrix, and the elements of the symmetric matrix are determined based on the elastic modulus and refractive index anisotropy of the optical fiber material. The optimization iterative algorithm adopts the Newton-Raphson method, and switches to the gradient descent method when the iteration fails to converge.

[0014] On the other hand, the present invention also provides a 100G single-polarization coherent optical real-time communication system for lightning strike disturbances, the system comprising: The lightning disturbance detection module is used to detect 100G single-polarization coherent optical signals using an asymmetric sliding window, and to obtain the time-domain characteristics of lightning electromagnetic pulses, phase error baseline, and stress wave propagation characteristics. The disturbance prediction modeling module is used to input the time-domain characteristics of the lightning electromagnetic pulse into the exponential decay model to obtain the phase error prediction value, and to input the stress wave propagation characteristics into the propagation velocity model to obtain the stress wave propagation delay. The timing alignment compensation module is used to perform timing alignment between the predicted phase error value and the phase error baseline using a lightning electromagnetic pulse characteristic spectrum matching algorithm to obtain timing alignment data. The timing alignment data is then processed using a hierarchical pre-compensation architecture to output compensation phase parameters. The stress field reconstruction module is used to determine the location and intensity of the lightning strike point based on the stress wave propagation delay, reconstruct the three-dimensional stress field distribution by combining the optical cable geometric parameters, convert the three-dimensional stress field distribution into a birefringence tensor distribution, and perform inverse solving on the birefringence tensor distribution to obtain polarization control parameters. The parallel optimization processing module is used to simplify the input parameterized fast estimation algorithm of the polarization control parameters, and simultaneously process the birefringence data of multiple fiber segments through the FPGA parallel processing architecture to output optimized polarization control parameters. The adaptive convergence control module is used to adjust the convergence parameters of the compensation phase parameter and the optimized polarization control parameter according to the attenuation rate using the exponential decay model of lightning disturbance, and output the final communication signal through an asymmetric convergence control strategy.

[0015] The 100G single-polarization coherent optical real-time communication method and system for lightning strike disturbances of the present invention have the following advantages over the prior art: (1) By integrating asymmetric sliding window detection and dual-model prediction mechanism, the lightning electromagnetic pulse characteristic spectrum matching algorithm is used for time alignment and hierarchical pre-compensation architecture processing. Combined with the three-dimensional stress field reconstruction of stress wave propagation delay and the inverse solution of birefringence tensor, and the multi-fiber segment collaborative optimization is achieved through FPGA parallel architecture and parameterized fast estimation, the signal recovery accuracy and real-time processing capability of 100G single polarization coherent optical communication system under lightning environment are improved. At the same time, the anti-environmental disturbance technical requirements of high-speed coherent optical communication are met through multi-physics field collaborative modeling and multi-parameter adaptive adjustment. (2) By adopting the short window high-frequency update mode and the long window low-frequency stable mode of the asymmetric sliding window, the short window of 8 to 16 symbol periods is used to capture the rapid changes of lightning electromagnetic pulse and the long window of 64 to 128 symbol periods is used to establish a statistical baseline. Combined with four-way parallel Stokes parameter monitoring and quantitative judgment criteria based on rising slope and peak amplitude, as well as the magnitude and direction analysis of the Stokes parameter change vector, the real-time identification accuracy of the 100G single polarization coherent optical system for the multi-physics field characteristics of lightning disturbance is improved. (3) By establishing a two-parameter exponential decay function based on peak intensity and rise time and integrating the nonlinear response relationship of optical fiber with the spectrum-dispersion convolution characteristics, the optical cable environment is divided into soil layer, protective layer and optical cable core layer using a layered medium model for stress wave propagation analysis. Combined with interface transmission coefficient calculation, multi-layer time delay accumulation and dynamic correction mechanism based on soil moisture content and compaction, the prediction accuracy of 100G single polarization coherent optical system for the time domain evolution trend of lightning disturbance and the calculation reliability of stress wave propagation delay are improved. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a 100G single-polarization coherent optical real-time communication method for lightning strike disturbances according to the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides a 100G single-polarization coherent optical real-time communication method for lightning strike disturbances, comprising the following steps: An asymmetric sliding window was used to detect 100G single-polarization coherent optical signals to obtain the time-domain characteristics of lightning electromagnetic pulses, phase error baseline, and stress wave propagation characteristics.

[0020] The time-domain characteristics of the lightning electromagnetic pulse are input into the exponential decay model to obtain the predicted phase error value, and the stress wave propagation characteristics are input into the propagation velocity model to obtain the stress wave propagation delay.

[0021] The predicted phase error value and the phase error baseline are time-aligned using a lightning electromagnetic pulse characteristic spectrum matching algorithm to obtain time-aligned data. A hierarchical pre-compensation architecture is then used to process the time-aligned data and output compensated phase parameters.

[0022] The location and intensity of the lightning strike are determined based on the stress wave propagation delay. The three-dimensional stress field distribution is reconstructed by combining the optical cable geometric parameters. The three-dimensional stress field distribution is then converted into a birefringence tensor distribution. The birefringence tensor distribution is then solved inversely to obtain the polarization control parameters.

[0023] The polarization control parameters are simplified by inputting a parameterized fast estimation algorithm, and the birefringence data of multiple fiber segments are processed simultaneously through an FPGA parallel processing architecture to output optimized polarization control parameters.

[0024] An exponential decay model for lightning disturbance is adopted. The convergence parameters of the compensation phase parameter and the optimized polarization control parameter are adjusted according to the decay rate. The final communication signal is output through an asymmetric convergence control strategy.

[0025] Specifically, this embodiment integrates asymmetric sliding window detection and dual-model prediction mechanisms, utilizes a lightning electromagnetic pulse characteristic spectrum matching algorithm for time alignment and hierarchical pre-compensation architecture processing, combines three-dimensional stress field reconstruction of stress wave propagation delay and inverse solution of birefringence tensor, and achieves multi-fiber segment collaborative optimization through FPGA parallel architecture and parameterized fast estimation. It adopts a lightning disturbance exponential decay model and asymmetric convergence control strategy to solve the spatiotemporal coupling disturbance problem between lightning electromagnetic pulse and stress wave propagation, improving the signal recovery accuracy and real-time processing capability of 100G single-polarization coherent optical communication system under lightning environment. At the same time, it meets the anti-environmental disturbance technical requirements of high-speed coherent optical communication through multi-physics field collaborative modeling and multi-parameter adaptive adjustment.

[0026] In one embodiment, the asymmetric sliding window includes a short window and a long window. The short window is used to obtain the time-domain characteristics of the lightning electromagnetic pulse, the long window is used to obtain the phase error baseline, and the Stokes polarization parameters are used to obtain the stress wave propagation characteristics.

[0027] The short window is set to a length of 8 to 16 symbol periods, and the long window is set to a length of 64 to 128 symbol periods. The short window samples the optical signal in a high-frequency update mode to capture the rapid changes in the lightning electromagnetic pulse, and the long window samples the optical signal in a low-frequency stable mode to establish a statistical baseline for the phase error.

[0028] By comparing the deviation between the detection results of the short window and the statistical baseline of the long window, the sudden temporal characteristics of the lightning electromagnetic pulse are identified. At the same time, a four-way parallel detection method is used to monitor the instantaneous change rate of the polarization state Stokes parameters S0, S1, S2, and S3 respectively to obtain the stress wave propagation characteristics.

[0029] In one embodiment, identifying the sudden temporal characteristics of a lightning electromagnetic pulse includes: Calculate the rising slope and peak amplitude of the signal power within the short window. When the rising slope exceeds 10 times the normal signal change rate and the peak amplitude exceeds 5 times the average power, it is determined to be a lightning electromagnetic pulse characteristic.

[0030] In one embodiment, acquiring the stress wave propagation characteristics includes: The first-order difference values ​​of the four polarization state Stokes parameters are calculated in real time, and the Stokes parameter change vector is established. When the magnitude of the Stokes parameter change vector exceeds a preset threshold and the change direction conforms to the polarization rotation law of stress wave propagation, the propagation velocity and attenuation coefficient of the stress wave are extracted to obtain the stress wave propagation characteristics.

[0031] In one specific embodiment, the formula for identifying the characteristics of the lightning electromagnetic pulse is as follows: ; in, The lightning electromagnetic pulse identification index; The slope of the signal power increase within the short window; This represents the normal rate of change of the signal. Peak amplitude, Average power, This is an adaptive weighting factor.

[0032] Specifically, the identification and determination formula for lightning electromagnetic pulse characteristics in this embodiment differs from traditional lightning detection, which relies solely on a single threshold. It introduces an adaptive weighting factor. This system enables a combined determination of lightning strike intensity and signal change rate. The weighting factor is dynamically adjusted based on the statistical characteristics of historical lightning events, avoiding misjudgment problems caused by fixed thresholds under different environmental conditions. At the same time, the system enhances the sensitivity to sudden electromagnetic pulses by multiplying the slope ratio and amplitude ratio.

[0033] The formula for calculating the magnitude of the Stokes parameter change vector is: ; in, The magnitude of the Stokes parameter variation vector; For the first First-order difference values ​​of the Stokes parameters; For the first The weighting coefficients for each Stokes parameter; These are the gradient compensation coefficients; For the first Spatial gradient of the Stokes parameters.

[0034] Specifically, the formula for calculating the magnitude of the Stokes parameter change vector in this embodiment differs from the classic Stokes parameter analysis, which only considers the changes in parameter values. It adds a spatial gradient term to the first-order difference. and gradient compensation coefficient This enables the capture of spatial non-uniform variations in polarization states. Weighting coefficients Differentiated processing based on the physical sensitivity of different Stokes parameters improves the spatial resolution and accuracy of stress wave propagation feature extraction.

[0035] This embodiment solves the problems of insufficient environmental adaptability and limited spatial resolution of polarization state change analysis in traditional single-threshold lightning strike detection by integrating a composite judgment mechanism with adaptive weights and a Stokes parameter analysis method with spatial gradient compensation. It improves the real-time recognition accuracy and anti-interference capability of 100G coherent optical systems for the sudden characteristics of lightning electromagnetic pulses and the multidimensional polarization characteristics of stress wave propagation.

[0036] In one embodiment, the step of inputting the time-domain characteristics of the lightning electromagnetic pulse into an exponential decay model to obtain a predicted phase error value, and inputting the stress wave propagation characteristics into a propagation velocity model to obtain the stress wave propagation delay, includes: A two-parameter exponential decay function is established based on the peak intensity and rise time of the lightning electromagnetic pulse. By fitting the energy dissipation law of the lightning disturbance in the optical fiber, the time-domain evolution trend of the phase error is predicted, and the predicted value of the phase error is obtained.

[0037] Based on the elastic modulus, density distribution, and burial depth parameters of the optical cable, a layered medium model for stress wave propagation is established to obtain the propagation velocity model. According to the refraction and reflection characteristics of stress waves in different medium layers, the time delay of stress waves propagating from the lightning strike point to the optical fiber is calculated, and the propagation delay is corrected by combining the attenuation characteristics of stress waves.

[0038] In one embodiment, the two-parameter exponential decay function includes a first parameter and a second parameter.

[0039] The first parameter is the lightning strike intensity attenuation coefficient, which is determined based on the relationship between the peak power of the lightning electromagnetic pulse and the nonlinear response of the optical fiber material.

[0040] The second parameter is a time constant, which is determined by the convolution result of the spectral characteristics of the lightning electromagnetic pulse and the dispersion characteristics of the optical fiber.

[0041] The layered medium model divides the optical cable burial environment into three propagation media: soil layer, protective layer, and optical cable core layer. The propagation speed of stress waves in each layer and the interface transmission coefficient are calculated respectively. The corrected stress wave propagation delay is obtained by accumulating the time delay of the multi-layer propagation path and multiplying the energy attenuation. The propagation speed of the soil layer is dynamically corrected according to the soil moisture content and compaction degree.

[0042] In one specific embodiment, the formula for calculating the two-parameter exponential decay function is: ; in, for The predicted phase error at time t; This is the lightning strike intensity attenuation coefficient; It is a time constant; This is the oscillation amplitude modulation factor; It is the angular frequency of the oscillation; This is the initial phase offset.

[0043] Specifically, the two-parameter exponential decay function in this embodiment differs from the traditional exponential decay model, which is monotonically decaying. It introduces an oscillation modulation term on top of the basic exponential decay term. This simulates the resonance effect and multimode coupling phenomenon caused by lightning strike disturbances in optical fibers. (Oscillation amplitude modulation factor) and oscillation angular frequency Based on the dispersion characteristics and geometric parameters of optical fibers, the modeling accuracy of complex electromagnetic pulse attenuation laws has been improved.

[0044] The propagation delay correction formula for the layered medium model is as follows: ; in, For the corrected stress wave propagation delay; For the first Layer dielectric thickness; For the first Layer propagation speed; For the first Layer transmittance coefficient; For the first Interface reflection delay; This is the interface attenuation correction factor. The decay exponent, This represents the distance between the interfaces.

[0045] Specifically, the propagation delay correction formula in this embodiment differs from the classic propagation delay calculation, which ignores interface reflection and environmental dynamics. It introduces an interface reflection delay term. and exponential decay correction factor The study considers the reflection effect of multi-layer media interfaces and distance-dependent attenuation characteristics. This correction mechanism can adapt to dynamic changes in different soil conditions and burial depths, improving the environmental adaptability of stress wave propagation delay prediction.

[0046] This embodiment solves the problems of incomplete physical mechanisms in traditional monotonic attenuation modeling and neglect of interface effects in multi-medium propagation calculation by establishing a two-parameter attenuation model with oscillation modulation and a layered propagation model with interface reflection correction. This improves the physical accuracy of lightning disturbance energy evolution prediction and the environmental adaptability and engineering practicality of stress wave multi-medium propagation delay calculation.

[0047] In one embodiment, the step of using a lightning electromagnetic pulse characteristic spectrum matching algorithm to perform time-series alignment of the predicted phase error value and the phase error baseline to obtain time-series aligned data, and then processing the time-series aligned data using a hierarchical pre-compensation architecture to output compensated phase parameters includes: The lightning electromagnetic pulse characteristic spectrum matching algorithm constructs a spectral fingerprint by extracting the dominant frequency component and harmonic components of the lightning electromagnetic pulse, and performs cross-correlation operation on the frequency domain characteristics of the spectral fingerprint and the phase error baseline to determine the timing offset.

[0048] The hierarchical pre-compensation architecture adopts a series-parallel hybrid structure. The first layer of the hierarchical pre-compensation architecture performs coarse compensation based on the lightning strike intensity, and the second layer of the hierarchical pre-compensation architecture performs fine compensation based on the residual error.

[0049] The coarse compensation adopts a fast-response feedforward control method, and selects a preset compensation gain coefficient according to the lightning strike intensity level.

[0050] The fine compensation adopts an adaptive feedback control method, which dynamically adjusts the compensation parameters by monitoring the statistical characteristics of the residual phase error in real time, and sets a buffer delay mechanism between the first layer and the second layer to avoid compensation conflicts.

[0051] In one embodiment, the spectral fingerprint includes three feature parameters: the center frequency, spectral width, and spectral asymmetry of the lightning electromagnetic pulse. The cross-correlation operation adopts a sliding window correlation detection method, and when the correlation coefficient is greater than 0.8, the optimal timing offset is determined.

[0052] The preset compensation gain coefficient of the coarse compensation is divided into three levels: weak, medium, and strong, according to the intensity of the lightning strike, corresponding to gain values ​​of 0.3, 0.6, and 0.9, respectively.

[0053] The fine-compensation adaptive feedback control employs a minimum mean square error algorithm, and the step size factor of the minimum mean square error algorithm is dynamically adjusted according to the variance of the residual phase error.

[0054] The buffer delay mechanism includes a delay time of 2 to 5 symbol periods. By monitoring the correlation between coarse compensation and fine compensation outputs, it determines whether there is a compensation conflict and adjusts the delay length accordingly.

[0055] In one specific embodiment, the formula for constructing the spectral fingerprint is: ; in, These are spectral fingerprint features; The center frequency; For spectrum width weights; The bandwidth is the frequency spectrum. Weights for asymmetry; For spectral asymmetry; This is the harmonic intensity factor.

[0056] Specifically, the spectral fingerprint construction formula in this embodiment differs from traditional spectral feature extraction, which uses a single parameter. It constructs a composite fingerprint that integrates center frequency, spectral width, and asymmetry, and introduces a harmonic intensity factor. This achieves a multi-dimensional comprehensive description of the frequency domain characteristics of lightning electromagnetic pulses. Weighting parameters and Adaptive adjustments are made based on the spectral distribution characteristics of different lightning strike types, enhancing the accuracy of spectral matching identification.

[0057] In one specific embodiment, the calculation formula for the minimum mean square error algorithm is: ; in, For the first The adaptive step size factor for the next iteration; This is the initial step size; The target variance; For the first The residual error variance of the next iteration; To prevent the elimination of zero factors; It is a correlation modulator; This is the signal correlation coefficient.

[0058] Specifically, the minimum mean square error algorithm in this embodiment differs from the classic LMS algorithm, which uses a fixed step size. It introduces an adaptive mechanism based on the error variance ratio and a signal correlation correction term. This achieves dynamic optimization of the step size. When the residual error is large, the step size is automatically increased to accelerate convergence, and when the solution is close to the optimum, the step size is decreased to improve accuracy. The correlation correction term further avoids convergence oscillation problems caused by signal correlation.

[0059] This embodiment solves the problems of incomplete feature description in traditional single-parameter spectrum matching and the contradiction between convergence speed and accuracy in fixed-step-size LMS algorithm by constructing a multi-dimensional composite spectral fingerprint and an adaptive step-size optimization algorithm for correlation correction. It improves the multi-dimensional accuracy of lightning electromagnetic pulse frequency domain feature recognition and the convergence efficiency and stability of adaptive compensation for residual phase error.

[0060] In one embodiment, the process of determining the lightning strike location and intensity based on the stress wave propagation delay, reconstructing the three-dimensional stress field distribution using optical cable geometric parameters, converting the three-dimensional stress field distribution into a birefringence tensor distribution, and inversely solving the birefringence tensor distribution to obtain polarization control parameters includes: The stress wave propagation delay is input into the triangulation algorithm to determine the location and intensity of the lightning strike. The space around the optical cable is discretized into stress calculation units using the finite element mesh generation method. The stress propagation equations are established in combination with the optical cable geometric parameters to reconstruct the three-dimensional stress field distribution.

[0061] The three-dimensional stress field distribution is converted into a birefringence tensor distribution using the photoelastic coefficient matrix. The birefringence tensor distribution is then solved in reverse using an optimization iterative algorithm to obtain the polarization control parameters.

[0062] The triangulation algorithm determines the location and intensity of the lightning strike point by using the hyperbola intersection method based on the time difference of arrival of stress waves at multiple monitoring points.

[0063] The finite element mesh generation method uses hexahedral mesh elements, employing high-density mesh generation in the vicinity of the optical cable and low-density mesh generation in the distant region.

[0064] The elastic-optic coefficient matrix is ​​a symmetric matrix, and the elements of the symmetric matrix are determined based on the elastic modulus and refractive index anisotropy of the optical fiber material.

[0065] The optimization iterative algorithm adopts the Newton-Raphson method, and switches to the gradient descent method when the iteration fails to converge.

[0066] In one specific embodiment, the formula for calculating the coordinates of the lightning strike point in the triangulation algorithm is: ; ; in, The coordinates of the lightning strike point; Number of monitoring points; For the first Weight of each monitoring point; For the first Coordinates of each monitoring point; Let the time difference confidence function be... For confidence decay parameters, For the first The time difference between the monitoring points.

[0067] Specifically, the lightning strike point coordinate calculation formula in this embodiment differs from the traditional triangulation method which uses equal weighting. Instead, it introduces a weighting function based on time difference confidence. and monitoring point weights This approach enables differentiated processing of the reliability of different monitoring points. The confidence function is designed based on the statistical distribution of time measurement errors, with monitoring points having smaller time differences receiving higher weights, thus improving the accuracy of lightning strike location and noise resistance.

[0068] The transformation optimization formula for the birefringence tensor is: ; in, These are the components of the birefringence tensor; It is the fourth-order elastic-optical coefficient tensor; These are the components of the stress tensor; Adjust the weights for gradient correction; It is a gradient coupling tensor; For stress gradient; This is a temperature correction factor; This is a temperature correction term.

[0069] Specifically, the birefringence tensor transformation optimization formula in this embodiment differs from the classical elasto-optical effect formula, which only considers the direct effect of stress. It adds a stress gradient term to the basic elasto-optical tensor transformation. and temperature correction item The effects of stress space inhomogeneity on the second-order effects of birefringence and temperature variations were considered. Gradient coupling tensor This describes the anisotropic projection of the stress gradient in different directions, with a temperature correction factor. It compensates for the effect of changes in ambient temperature on the elastic coefficient.

[0070] This embodiment solves the problems of measurement error sensitivity in traditional equal-weighted positioning and neglect of spatial non-uniformity in classical elasto-optic modeling by establishing a confidence-weighted multi-point triangulation algorithm and a gradient-corrected birefringence tensor transformation model. It significantly improves the anti-interference accuracy of lightning strike point location determination and the physical modeling integrity and environmental adaptability of stress field to birefringence tensor transformation.

[0071] In one embodiment, the step of simplifying the input parameterization fast estimation algorithm for the polarization control parameters, and simultaneously processing birefringence data from multiple fiber segments using an FPGA parallel processing architecture to output optimized polarization control parameters includes: The polarization control parameters are input into a parameterized fast estimation algorithm. The high-dimensional polarization control parameters are converted into low-dimensional parameter representations through a dimensionality reduction mapping method. The polarization transformation function is simplified using a piecewise linear approximation method to obtain simplified polarization control parameters.

[0072] In one specific embodiment, the dimensionality reduction mapping method uses principal component analysis to extract the main feature components of the polarization control parameters, and retains the top few principal components with the largest contribution rates as low-dimensional parameter representations.

[0073] The piecewise linear approximation method divides the numerical range of the polarization control parameters into multiple linear intervals, and uses different linear functions for approximation within each linear interval. Through continuity constraints at the interval boundaries, the smooth transition of the simplified polarization control parameters is ensured.

[0074] The simplified polarization control parameters are input into the FPGA parallel processing architecture, and the birefringence data of multiple fiber segments are processed simultaneously using a multi-channel parallel computing method. Through pipeline processing mechanism and cache optimization strategy, optimized polarization control parameters are output.

[0075] In one specific embodiment, the multi-channel parallel computing method distributes the birefringence data of multiple fiber segments to different parallel computing units, and each parallel computing unit independently processes the polarization compensation calculation of the corresponding fiber segment.

[0076] The pipelined processing mechanism divides the polarization compensation calculation process into three stages: data preprocessing, parameter calculation, and result postprocessing. Data is transferred between each stage through a cache queue. The cache optimization strategy adopts a dual-caching mechanism, which preloads the next batch of data into the backup cache while the current cache is being calculated.

[0077] Specifically, this embodiment solves the problems of excessively high real-time computational complexity of high-dimensional polarization control parameters and insufficient processing efficiency of multi-fiber segment birefringence compensation by adopting a principal component analysis-based dimensionality reduction mapping method and a piecewise linear approximation simplification strategy. It utilizes the multi-channel computation and three-stage pipeline mechanism of the FPGA parallel processing architecture, combined with a double-buffer optimization strategy, interval boundary continuity constraints, and a data allocation mechanism for independent parallel computing units. This improves the simplification accuracy of complex polarization transformation functions and the real-time performance of multi-fiber parallel processing in 100G single-polarization coherent optical systems. At the same time, through parameterized fast estimation and hardware parallel optimization, it meets the low latency and high throughput requirements of high-speed coherent optical communication systems for polarization control parameter optimization.

[0078] In one embodiment, the step of employing a lightning disturbance exponential decay model, adjusting the convergence parameters of the compensation phase parameter and the optimized polarization control parameter according to the decay rate, and outputting the final communication signal through an asymmetric convergence control strategy includes: The compensated phase parameter and the optimized polarization control parameter are input into the lightning disturbance exponential decay model. The dynamic convergence coefficient is calculated based on the decay rate of the lightning disturbance. The convergence parameters of the compensated phase parameter and the optimized polarization control parameter are adjusted using a two-parameter adaptive adjustment method to obtain the adjusted compensated phase parameter and the adjusted optimized polarization control parameter.

[0079] In one specific embodiment, the lightning disturbance exponential decay model establishes an exponential decay function based on the time decay law of lightning intensity, and determines the decay rate by monitoring the real-time change rate of lightning disturbance energy.

[0080] The dual-parameter adaptive adjustment method uses two independent parameters: phase convergence coefficient and polarization convergence coefficient. The phase convergence coefficient and polarization convergence coefficient are adjusted differently according to the dynamic convergence coefficient. When the lightning disturbance is in the rapid decay stage, the convergence coefficient is increased, and when the lightning disturbance is in the slow decay stage, the convergence coefficient is decreased.

[0081] The adjusted compensated phase parameters and the adjusted optimized polarization control parameters are processed by an asymmetric convergence control strategy. The final communication signal is output through coordinated control using a fast and slow dual-speed convergence mechanism and a stability judgment algorithm.

[0082] In one specific embodiment, the fast and slow dual-speed convergence mechanism employs different convergence speeds for phase compensation and polarization compensation, using a fast convergence mode for phase compensation and a slow convergence mode for polarization compensation.

[0083] The stability judgment algorithm determines the system stability by monitoring the fluctuation range of the compensation parameters within a continuous time window. When the fluctuation range is less than a preset threshold, the system is considered stable. The collaborative control maintains the synchronous update of the two compensation parameters through a timing synchronization mechanism to avoid signal quality degradation caused by compensation timing mismatch.

[0084] Specifically, this embodiment establishes an exponential decay function based on the time decay law of lightning strike intensity and a two-parameter adaptive adjustment method. It utilizes a differentiated adjustment mechanism of phase convergence coefficient and polarization convergence coefficient, combined with an asymmetric control strategy of fast and slow dual-speed convergence, a stability judgment algorithm based on fluctuation range monitoring, and a collaborative control strategy of timing synchronization mechanism. This solves the problems of mismatched convergence speed of compensation parameters and timing synchronization of multi-parameter collaborative control during the attenuation process of lightning disturbance. It improves the parameter convergence accuracy and system stability of 100G single-polarization coherent optical system during the dynamic attenuation process of lightning disturbance. At the same time, through dynamic convergence coefficient calculation and asymmetric convergence control, it meets the adaptive compensation and real-time stable output requirements of high-speed coherent optical communication under complex lightning environments.

[0085] The present invention also provides a 100G single-polarization coherent optical real-time communication system for lightning strike disturbances, the system comprising: The lightning disturbance detection module is used to detect 100G single-polarization coherent optical signals using an asymmetric sliding window, and to obtain the time-domain characteristics of lightning electromagnetic pulses, phase error baseline, and stress wave propagation characteristics.

[0086] The disturbance prediction modeling module is used to input the time-domain characteristics of the lightning electromagnetic pulse into the exponential decay model to obtain the phase error prediction value, and to input the stress wave propagation characteristics into the propagation velocity model to obtain the stress wave propagation delay.

[0087] The timing alignment compensation module is used to perform timing alignment between the predicted phase error value and the phase error baseline using a lightning electromagnetic pulse characteristic spectrum matching algorithm to obtain timing alignment data. The timing alignment data is then processed using a hierarchical pre-compensation architecture to output compensated phase parameters.

[0088] The stress field reconstruction module is used to determine the location and intensity of the lightning strike point based on the stress wave propagation delay, reconstruct the three-dimensional stress field distribution by combining the optical cable geometric parameters, convert the three-dimensional stress field distribution into a birefringence tensor distribution, and perform inverse solving on the birefringence tensor distribution to obtain the polarization control parameters.

[0089] The parallel optimization processing module is used to simplify the input parameterized fast estimation algorithm of the polarization control parameters, and simultaneously process the birefringence data of multiple fiber segments through the FPGA parallel processing architecture to output optimized polarization control parameters.

[0090] The adaptive convergence control module is used to adjust the convergence parameters of the compensation phase parameter and the optimized polarization control parameter according to the attenuation rate using the exponential decay model of lightning disturbance, and output the final communication signal through an asymmetric convergence control strategy.

[0091] Specifically, this embodiment presents a 100G single-polarization coherent optical real-time communication system for lightning strike disturbances. By constructing and integrating six modules—lightning strike disturbance detection, disturbance prediction modeling, timing alignment compensation, stress field reconstruction, parallel optimization processing, and adaptive convergence control—it combines multi-dimensional disturbance detection using an asymmetric sliding window with predictive modeling of a two-parameter exponential decay model. It integrates timing alignment using spectrum matching with phase compensation processing using a hierarchical pre-compensation architecture, inverse solving of the birefringence tensor for three-dimensional stress field reconstruction with multi-segment optimization calculations using an FPGA parallel architecture, and an asymmetric convergence control strategy for the lightning strike disturbance attenuation model. This solves the problems of multi-physics coupling disturbance identification, real-time prediction compensation, and adaptive parameter optimization in 100G single-polarization coherent optical communication systems under complex lightning strike environments. It improves the lightning strike disturbance resistance, signal recovery accuracy, and real-time processing performance of high-speed coherent optical communication systems, achieving a breakthrough in high-reliability optical communication technology for harsh electromagnetic environments.

[0092] In one specific embodiment, a 100G single-polarization coherent optical communication system is used for long-distance data transmission on a backbone fiber optic communication link. The link is approximately 200 kilometers long, with the optical cable buried at a depth of 1.2 meters, and passes through several lightning-prone areas. The system configuration includes: a 100Gbps transmission rate, single-polarization coherent detection, a symbol rate of 32Gbaud, and a modulation format of Quadrature Phase Shift Keying (QPSK). The specific implementation process includes: Step 1, Lightning Disturbance Detection: The system detected a strong lightning strike, and the lightning disturbance detection module activated the asymmetric sliding window detection mechanism. The short window was set to 12 symbol periods, approximately 375 picoseconds, and the long window was set to 96 symbol periods, approximately 3 nanoseconds.

[0093] At the instant of the lightning strike, the signal power rise rate detected within a short window reached 15 times the normal rate of change, and the peak amplitude exceeded 8 times the average power. Based on the lightning electromagnetic pulse identification formula, the calculated lightning electromagnetic pulse identification index exceeded a preset threshold, and the system successfully identified the characteristics of the lightning electromagnetic pulse.

[0094] Meanwhile, the four-channel parallel detection method detected significant changes in the polarization state Stokes parameters: the S0 parameter decreased by about 12%, the S1 and S2 parameters changed by 8% and 6% respectively, and the S3 parameter changed by 4%. By calculating the magnitude of the Stokes parameter change vector, the stress wave propagation speed was determined to be approximately 3200 meters per second, with an attenuation coefficient of 0.15.

[0095] Step 2, Perturbation Prediction Modeling: The time-domain characteristics of the detected lightning electromagnetic pulse are input into a two-parameter exponential decay function through the disturbance prediction modeling module. Based on the lightning strike intensity and rise time, the lightning strike intensity decay coefficient is determined to be 0.025, the time constant is 15 microseconds, the peak power of the lightning strike intensity is approximately 500 watts, and the rise time is approximately 2 microseconds.

[0096] Based on optical cable parameters, including an elastic modulus of 70 gigapascals (GPa) and a density of 2200 kg / m³, a layered medium model was established. The calculated propagation speed of the stress wave was 1800 m / s in the soil layer, 2800 m / s in the protective layer, and 4200 m / s in the optical cable core. Considering the interface transmission coefficient and the multi-layer propagation paths, the stress wave propagation delay was calculated to be 45 milliseconds. Step 3, Timing Alignment Compensation: The timing alignment compensation module employs a lightning electromagnetic pulse characteristic spectrum matching algorithm to extract the dominant frequency component, harmonic components, and spectral asymmetry of the lightning pulse, constructing a spectral fingerprint. The dominant frequency component is approximately 2.5 MHz. Through cross-correlation calculations, the timing offset between the predicted phase error and the baseline is determined to be 3.2 microseconds.

[0097] The hierarchical pre-compensation architecture is initiated. The first layer of coarse compensation is based on the strong lightning strike intensity level and selects a preset gain coefficient of 0.9 for rapid response. The second layer of fine compensation uses the minimum mean square error algorithm, and after 6 iterations, the residual phase error is reduced to 8% of the original error.

[0098] Step 4, Stress Field Reconstruction: Based on the stress wave arrival time difference data from 5 monitoring points distributed along the link, the stress field reconstruction module uses a triangulation algorithm to determine the location of the lightning strike point: approximately 85 meters from the optical cable, with an intensity level of III.

[0099] Using the finite element method (FEM), the space surrounding the optical cable was divided into approximately 15,000 hexahedral mesh elements. A high-density mesh was used near the cable, with each element measuring 0.1 meters, while a low-density mesh was used further away, with each element measuring 2 meters. The reconstructed three-dimensional stress field showed a maximum stress of approximately 50 megapascals (MPa) with a radius of approximately 20 meters.

[0100] The birefringence tensor distribution is obtained by transforming the elastic-optical coefficient matrix. The Newton-Raphson iterative algorithm converges after 8 iterations, and the polarization control parameters are obtained.

[0101] Step 5, Parallel optimization processing: The polarization control parameters are input into a parameterized fast estimation algorithm through a parallel optimization processing module. Principal component analysis is used to reduce the dimensionality of the 120-dimensional polarization control parameters to 15 dimensions, retaining approximately 95% of the information. A piecewise linear approximation method is employed to divide the parameter range into eight linear intervals for simplification.

[0102] An FPGA parallel processing architecture was implemented, dividing the 200-kilometer link into 20 fiber segments, each assigned to one of 20 parallel computing units for simultaneous processing. A pipelined mechanism divided the processing into three stages: data preprocessing, parameter calculation, and result post-processing, with processing times of 0.8 microseconds, 1.2 microseconds, and 0.5 microseconds respectively. A dual-buffering mechanism ensured continuous data stream processing, keeping the total processing latency below 5 microseconds.

[0103] Step 6, Adaptive Convergence Control: The adaptive convergence control module dynamically adjusts the convergence parameters based on the attenuation rate of the lightning disturbance, which is approximately an exponential decay of 0.08 per second. The phase convergence coefficient is set to 0.75, and the polarization convergence coefficient is set to 0.45 to achieve differentiated adjustment.

[0104] In the fast-slow dual-speed convergence mechanism, phase compensation employs a fast convergence mode, while polarization compensation employs a slow convergence mode. The convergence time for the fast convergence mode is approximately 2 milliseconds, and the convergence time for the slow convergence mode is approximately 8 milliseconds. A stability judgment algorithm monitors the fluctuation range of the compensation parameters within a continuous 10-millisecond time window; the system is considered stable when the fluctuation is less than 3%.

[0105] This embodiment presents a 100G single-polarization coherent optical real-time communication system resistant to lightning strike disturbances. Through the aforementioned processing steps, signal recovery can be completed within 15 milliseconds after a lightning strike. Signal quality indicators show that the bit error rate improved from 10⁻⁶ at the moment of the lightning strike to 10⁻¹², the optical signal-to-noise ratio recovered from 18 dB to 28 dB, and the polarization degree recovered from 0.3 to 0.95. Compared with traditional methods, this embodiment shortens the signal recovery time by 65%, improves the signal quality recovery accuracy by 40%, and achieves a system availability of 99.97%, meeting the anti-disturbance technical requirements of high-speed coherent optical communication systems under lightning strike conditions.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A 100G single polarization coherent optical real-time communication method for lightning disturbance, characterized in that, The method comprises the following steps: The asymmetric sliding window is used for detecting the 100G single polarization coherent optical signal, the lightning electromagnetic pulse time domain characteristics, the phase error baseline and the stress wave propagation characteristics are obtained; The lightning electromagnetic pulse time domain characteristics are input into the exponential decay model to obtain the phase error prediction value, and the stress wave propagation characteristics are input into the propagation velocity model to obtain the stress wave propagation delay; The lightning electromagnetic pulse characteristic spectrum matching algorithm is used to time-align the phase error prediction value and the phase error baseline to obtain time-aligned data, the hierarchical pre-compensation architecture is used to process the time-aligned data, and the compensation phase parameter is output; The lightning point position and intensity are determined according to the stress wave propagation delay, the three-dimensional stress field distribution is reconstructed in combination with the optical cable geometric parameters, the three-dimensional stress field distribution is converted into the birefringence tensor distribution, the polarization control parameter is obtained by inversely solving the birefringence tensor distribution; The polarization control parameter is input into the parameterized fast estimation algorithm for simplified processing, the birefringence data of multiple optical fiber segments are simultaneously processed through the FPGA parallel processing architecture, and the optimized polarization control parameter is output. The lightning disturbance exponential decay model is used to adjust the convergence parameters of the compensation phase parameter and the optimized polarization control parameter according to the decay rate, and the final communication signal is output through the asymmetric convergence control strategy.

2. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 1, characterized in that, The asymmetric sliding window comprises a short window and a long window, the lightning electromagnetic pulse time domain characteristics are obtained through the short window, the phase error baseline is obtained through the long window, and the stress wave propagation characteristics are obtained by detecting the polarization state Stokes parameter; The length of the short window is set to 8-16 symbol periods, the length of the long window is set to 64-128 symbol periods, the short window samples the optical signal in a high-frequency update mode to capture the sharp change characteristics of the lightning electromagnetic pulse, and the long window samples the optical signal in a low-frequency stable mode to establish a statistical baseline of the phase error; The deviation degree of the detection result of the short window and the statistical baseline of the long window is compared to identify the burst time domain characteristics of the lightning electromagnetic pulse, and the instantaneous change rates of the polarization state Stokes parameters S0, S1, S2 and S3 are monitored respectively through a four-way parallel detection mode to obtain the stress wave propagation characteristics.

3. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 2, characterized in that, The identification of the burst time domain characteristics of the lightning electromagnetic pulse comprises: The rising slope and peak amplitude of the signal power in the short window are calculated, when the rising slope exceeds 10 times the normal signal change rate and the peak amplitude exceeds 5 times the average power, the lightning electromagnetic pulse characteristics are determined; The stress wave propagation characteristics comprise: The first-order difference values of the four polarization state Stokes parameters are calculated in real time to establish a Stokes parameter change vector, when the module length of the Stokes parameter change vector exceeds a preset threshold and the change direction conforms to the polarization rotation rule of the stress wave propagation, the propagation velocity and the attenuation coefficient of the stress wave are extracted to obtain the stress wave propagation characteristics; The module length calculation formula of the Stokes parameter change vector is: ； wherein is a modulus of the Stokes parameter variation vector; is a first-order difference value of the th Stokes parameter; is a weighting coefficient of the th Stokes parameter; is a gradient compensation coefficient; is a spatial gradient of the th Stokes parameter.

4. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 1, characterized in that, The lightning electromagnetic pulse time domain characteristic is input into an exponential decay model to obtain a phase error prediction value, and the stress wave propagation characteristic is input into a propagation speed model to obtain a stress wave propagation delay, including: A double-parameter exponential decay function is established according to the peak intensity and the rise time of the lightning electromagnetic pulse, the energy dissipation law of the lightning disturbance in the optical fiber is fitted, the time domain evolution trend of the phase error is predicted, and a phase error prediction value is obtained; A layered medium model of stress wave propagation is established based on the elastic modulus, the density distribution and the burial depth parameters of the optical cable, the propagation speed model is obtained, the time delay of the stress wave from the lightning point to the optical fiber is calculated according to the refraction and reflection characteristics of the stress wave in different medium layers, and the propagation delay is corrected in combination with the attenuation characteristics of the stress wave.

5. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 4, characterized in that, The double-parameter exponential decay function includes a first parameter and a second parameter; The first parameter is a lightning intensity attenuation coefficient, which is determined according to the nonlinear response relationship between the peak power of the lightning electromagnetic pulse and the optical fiber material; The second parameter is a time constant, which is determined according to the convolution result of the spectral characteristics of the lightning electromagnetic pulse and the dispersion characteristics of the optical fiber; The layered medium model divides the optical cable burial environment into three propagation media of soil layer, protection layer and optical cable core layer, respectively calculates the propagation speed and the interface transmission coefficient of the stress wave in each layer, and obtains the corrected stress wave propagation delay through the product operation of the time delay accumulation and the energy attenuation of the multi-layer propagation path, and the propagation speed of the soil layer is dynamically corrected according to the soil water content and the compaction degree.

6. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 1, characterized in that, The lightning electromagnetic pulse characteristic spectrum matching algorithm is used to time-align the phase error prediction value and the phase error baseline to obtain time-aligned data, and a layered pre-compensation architecture is used to process the time-aligned data to output a compensation phase parameter, including: The lightning electromagnetic pulse characteristic spectrum matching algorithm constructs a spectral fingerprint by extracting the main frequency component and the harmonic component of the lightning electromagnetic pulse, and performs cross-correlation operation on the spectral fingerprint and the frequency domain characteristics of the phase error baseline to determine the time sequence offset; The layered pre-compensation architecture adopts a series-parallel hybrid structure, a first layer of the layered pre-compensation architecture performs coarse compensation according to the lightning intensity, and a second layer of the layered pre-compensation architecture performs fine compensation according to the residual error; The coarse compensation adopts a fast response feedforward control mode, and a preset compensation gain coefficient is selected according to the lightning intensity level; The fine compensation adopts an adaptive feedback control mode, compensation parameters are dynamically adjusted by monitoring the statistical characteristics of the residual phase error in real time, a buffer delay mechanism is set between the first layer and the second layer to avoid compensation conflict.

7. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 6, characterized in that, The spectral fingerprint includes three characteristic parameters, namely the center frequency, the spectral width and the spectral asymmetry of the lightning electromagnetic pulse, and the cross-correlation operation adopts a sliding window correlation detection mode, and when the correlation coefficient is greater than 0.8, the best time sequence offset is determined; The preset compensation gain coefficient of the coarse compensation is divided into three levels of weak, medium and strong according to the lightning intensity, and corresponds to gain values of 0.3, 0.6 and 0.9 respectively. The fine compensation adaptive feedback control adopts a least mean square error algorithm, and a step factor of the least mean square error algorithm is dynamically adjusted according to a variance of residual phase error; The buffer delay mechanism includes a delay time of 2 to 5 symbol periods, and whether there is a compensation conflict is judged by monitoring correlation of coarse compensation and fine compensation outputs, and the delay length is adjusted; A calculation formula of the least mean square error algorithm is: ； in, For the first The adaptive step size factor for the next iteration; This is the initial step size; The target variance; For the first The residual error variance of the next iteration; To prevent the elimination of zero factors; It is a correlation modulator; This is the signal correlation coefficient.

8. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 1, characterized in that, The lightning strike point position and intensity are determined according to the stress wave propagation delay, a three-dimensional stress field distribution is reconstructed in combination with optical cable geometric parameters, the three-dimensional stress field distribution is converted into a birefringent tensor distribution, and polarization control parameters are obtained by inversely solving the birefringent tensor distribution, including: The lightning strike point position and intensity are determined by inputting the stress wave propagation delay into a triangular positioning algorithm, an optical cable surrounding space is discretized into stress calculation units by using a finite element mesh division method, and a stress propagation equation set is established in combination with the optical cable geometric parameters to reconstruct the three-dimensional stress field distribution; The three-dimensional stress field distribution is converted into the birefringent tensor distribution through a photoelastic coefficient matrix, and the birefringent tensor distribution is inversely solved by using an optimization iterative algorithm to obtain the polarization control parameters.

9. The 100G single polarization coherent optical real-time communication method against lightning disturbance according to claim 8, characterized in that, The triangular positioning algorithm determines the lightning strike point position and intensity based on a time difference of stress waves arriving at a plurality of monitoring points through a hyperbolic curve intersection method; The finite element mesh division method uses a hexahedral mesh unit, high-density mesh division is used in a region near the optical cable, and low-density mesh division is used in a region far away; The photoelastic coefficient matrix is a symmetric matrix, and symmetric matrix elements are determined according to anisotropic characteristics of an elastic modulus and a refractive index of an optical fiber material; The optimization iterative algorithm uses a Newton-Raphson method, and switches to a gradient descent method when iteration does not converge.

10. A 100G single polarization coherent optical real-time communication system against lightning disturbance, used for performing the 100G single polarization coherent optical real-time communication method against lightning disturbance according to any one of claims 1-9, characterized in that, The system comprises: A lightning strike disturbance detection module is configured to detect a 100G single polarization coherent optical signal by using an asymmetric sliding window, and obtain lightning electromagnetic pulse time domain characteristics, a phase error baseline, and stress wave propagation characteristics; A disturbance prediction modeling module is configured to input the lightning electromagnetic pulse time domain characteristics into an exponential decay model to obtain a phase error prediction value, and input the stress wave propagation characteristics into a propagation velocity model to obtain a stress wave propagation delay; A timing alignment compensation module is configured to perform timing alignment on the phase error prediction value and the phase error baseline by using a lightning electromagnetic pulse characteristic spectrum matching algorithm to obtain timing alignment data, and process the timing alignment data by using a hierarchical pre-compensation architecture to output compensation phase parameters; A stress field reconstruction module is configured to determine a lightning strike point position and intensity according to the stress wave propagation delay, reconstruct a three-dimensional stress field distribution in combination with optical cable geometric parameters, convert the three-dimensional stress field distribution into a birefringent tensor distribution, and inversely solve the birefringent tensor distribution to obtain polarization control parameters. A stress field reconstruction module is configured to determine a lightning strike point position and intensity according to the stress wave propagation delay, reconstruct a three-dimensional stress field distribution in combination with optical cable geometric parameters, convert the three-dimensional stress field distribution into a birefringent tensor distribution, and inversely solve the birefringent tensor distribution to obtain polarization control parameters. A parallel optimization processing module is configured to input the polarization control parameter into a parameterized fast estimation algorithm for simplified processing, simultaneously process birefringence data of multiple fiber sections through a Field-Programmable Gate Array (FPGA) parallel processing architecture, and output an optimized polarization control parameter. An adaptive convergence control module is configured to adopt a lightning disturbance exponential decay model, adjust convergence parameters of the compensation phase parameter and the optimized polarization control parameter according to a decay rate, and output a final communication signal through an asymmetric convergence control strategy.

Images