A method for monitoring and early warning of the operation and transmission status of optical fiber communication

Abstract

This invention discloses a method for monitoring and early warning of the operational transmission status of optical fiber communication, belonging to the field of communication status monitoring technology. Starting with acquiring the pulse amplitude, the invention constructs first and second differential polarization angles and the difference in differential polarization angles using three optical pulses. It calculates the average residual angle using a sliding window and determines the presence of strong disturbances, ensuring that only the truly disturbed areas are processed. At the disturbance center, a prediction window and a verification window are combined to compensate and correct the cumulative rotation angle, suppressing single-point noise and residual trend errors. When the median deviation of the verified residual angle sequence exceeds the limit, the interval to be split is recursively split and secondary compensation is performed to remove the masking effect of the preceding disturbance on the subsequent polarization ground state. Finally, clustering and representative point selection determine the spatial location of each damage point. This achieves decoupling of multi-point disturbances from front to back, ensuring that the residual angle of each damage point is calculated based on the independent cumulative rotation angle of its respective interval, thereby detecting multiple damage points in a single measurement.

Description

Technical Field

[0001] This invention relates to the field of communication status monitoring technology, specifically a method for monitoring and early warning of the operation and transmission status of optical fiber communication. Background Technology

[0002] In actual optical fiber communication operation and transmission status monitoring scenarios, optical fiber links may suffer multiple external damages along their path. The polarization state exhibits non-uniform and abrupt spatial distribution characteristics with distance. Existing technologies have the following defects: First, existing optical fiber sensing methods based on polarization state changes do not decouple and quantify the abrupt change in polarization ground state caused by the first strong disturbance, the cumulative rotation angle propagation effect, and the residual polarization changes at subsequent locations. This makes it impossible to match the true spatial superposition law of polarization information in multi-point damage events. As a result, the information of the damage points after the first disturbance is masked by the nonlinear superposition of the polarization ground state. The multi-point positioning results lack objective separation basis, and the number of damage points detected is seriously insufficient. Furthermore, the existing scheme lacks a multi-point perturbation recursive splitting mechanism adapted to dynamic compensation of polarization ground state. It cannot implement online adaptive correction of subsequent sampling points based on the cumulative rotation angle of the detected perturbation center. It cannot effectively recover the residual polarization information of the subsequent damage point after the first strong perturbation occurs, and it is easy to miss subsequent damage points when multiple perturbations coexist due to the lack of iterative compensation. At the same time, there is no bidirectional feedback correction between the polarization analysis module and the damage point location module, which makes it impossible to use the fluctuation information of the verification residual angle for rolling optimization of the cumulative rotation angle, resulting in long-term mismatch between location and compensation.

[0003] Therefore, there is an urgent need for a fiber optic communication operation and transmission status monitoring and early warning method that can quantify the first differential polarization angle and the second differential polarization angle, calculate the rotation angle increment and the backward trend slope at the disturbance center, and realize dynamic compensation of cumulative rotation angle and recursive secondary compensation of the interval to be split, so as to solve the above problems. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for monitoring and early warning of the operational transmission status of optical fiber communication. It solves the problem in optical fiber sensing methods based on polarization state changes that a strong disturbance at the beginning of the optical fiber link will change the polarization ground state of all subsequent locations, causing subsequent disturbance information to be nonlinearly masked and multi-point disruptive events to be unable to be detected simultaneously.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: a method for monitoring and early warning of the operation and transmission status of optical fiber communication, comprising the following steps: performing polarization analysis on the backscattered optical signal corresponding to the probe optical signal, and calculating the pulse amplitude angle of each sampling point in combination with the transmission timing of the probe optical signal.

[0006] The first differential polarization angle, the second differential polarization angle, and the differential polarization angle difference are determined based on the pulse amplitude. At the same time, the average residual angle and the average differential polarization angle difference of the current detection window are calculated in combination with the pulse intensity.

[0007] If the absolute value of the average residual angle is less than a preset threshold, the current detection point is moved one sampling point to the right. Otherwise, it is determined that there is a strong disturbance in the current detection window, and the disturbance center is determined based on the first differential polarization angle.

[0008] The current cumulative rotation angle of the disturbance center is compensated, the verification residual angle sequence of the current detection window is determined, and the compensated cumulative rotation angle is corrected by combining the first differential polarization angle.

[0009] After correction, if the corresponding residual angle sequence of the verification meets the set conditions, the current detection window is determined as the interval to be split, and the disturbance center is re-determined and secondary compensation is performed based on the second differential polarization angle until there is no interval to be split.

[0010] The spatial location of each damage point is determined based on the final residual angle of each sampling point.

[0011] Compared with existing technologies, this invention has the following advantages: Starting with acquiring the pulse amplitude, it uses three light pulses to construct the first and second differential polarization angles and the difference in differential polarization angles, providing multi-dimensional polarization characteristics to distinguish different disturbance sources; it calculates the average residual angle through a sliding window and determines the existence of strong disturbances, ensuring that only the truly disturbed region is processed; at the disturbance center, it combines a prediction window and a verification window to compensate and correct the cumulative rotation angle, effectively suppressing single-point noise and residual trend errors; when the median deviation of the verified residual angle sequence exceeds the limit, it recursively splits the interval to be split and performs secondary compensation, gradually removing the masking effect of the preceding disturbance on the subsequent polarization ground state; finally, it determines the spatial location of each damage point through clustering and representative point selection. The entire step chain achieves sequential decoupling of multiple disturbances from front to back, so that the residual angle of each damage point is calculated based on the independent cumulative rotation angle of its respective interval, thereby detecting multiple damage points simultaneously in a single measurement. Attached Figure Description

[0012] Figure 1 This is a flowchart of the fiber optic communication operation transmission status monitoring and early warning method of the present invention;

[0013] Figure 2 This is a flowchart illustrating the calculation of pulse amplitude in the optical fiber communication operation transmission status monitoring and early warning method of the present invention;

[0014] Figure 3 This is a flowchart illustrating the process of determining the residual angle sequence of the current detection window in the optical fiber communication operation and transmission status monitoring and early warning method of the present invention. Detailed Implementation

[0015] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. Please refer to the accompanying drawings. Figure 1 The present invention provides a technical solution: a method for monitoring and early warning of the operation and transmission status of optical fiber communication, comprising the following steps: S1, performing polarization analysis on the backscattered light signal corresponding to the probe light signal, and calculating the pulse amplitude angle of each sampling point in combination with the emission timing of the probe light signal.

[0016] Because backscattered optical signals are affected by the superposition of disturbances at multiple locations along the fiber during transmission, and the change in polarization state is essentially a vector rotation in three-dimensional space, the trajectory of this rotation cannot be fully described by the intensity or phase information of a single pulse. To subsequently separate the independent features of disturbances at different locations from the aliased polarization information, sufficiently detailed polarization state characterization parameters must first be obtained. Therefore, the transmission timing of three optical pulses is used, and a polarization diversity receiving module is employed to decompose the optical signal into two orthogonal polarization components, horizontal and vertical. Then, sliding time window integration sampling is used to obtain three sets of amplitude pairs corresponding to each sampling point. Specifically, as shown... Figure 2 As shown, the process of calculating the pulse amplitude at each sampling point is as follows: S101, the probe light signal is injected into the fiber optic link under test, and the backscattered light signal is received at the end of the fiber optic link under test. The backscattered light signal is decomposed into a horizontal polarization component and a vertical polarization component according to two orthogonal polarization directions by the polarization beam splitter in the polarization diversity receiver module. The horizontal polarization component and the vertical polarization component are respectively input into the balanced photodetector and converted into the first electrical signal and the second electrical signal.

[0017] S102. Based on the transmission timing of the first, second, and third optical pulses in the detection optical signal, the first and second electrical signals are sampled by sliding time window integration using the two-way propagation time interval between two adjacent sampling points as the time window. The voltage amplitude is read at the beginning of each time window, at the time after the first transmission time interval, and at the time after the second transmission time interval, respectively. Thus, at each sampling point, the first horizontal polarization amplitude and the first vertical polarization amplitude corresponding to the first optical pulse, the second horizontal polarization amplitude and the second vertical polarization amplitude corresponding to the second optical pulse, and the third horizontal polarization amplitude and the third vertical polarization amplitude corresponding to the third optical pulse are obtained.

[0018] S103. For each sampling point, calculate the first amplitude of the first optical pulse, the second amplitude of the second optical pulse, and the third amplitude of the third optical pulse. The first amplitude is defined as the arctangent function value of the ratio of the first vertical polarization amplitude to the first horizontal polarization amplitude. The second amplitude is defined as the arctangent function value of the ratio of the second vertical polarization amplitude to the second horizontal polarization amplitude. The third amplitude is defined as the arctangent function value of the ratio of the third vertical polarization amplitude to the third horizontal polarization amplitude.

[0019] S104. Output the first, second, and third amplitudes as the pulse amplitudes of each sampling point.

[0020] The horizontal and vertical polarization components refer to the two orthogonally linearly polarized light components that are separated from the backscattered light signal after passing through a polarization beam splitter. They are directly separated from the received light signal by the polarization beam splitter. Their relative amplitudes determine the polarization direction angle of the light, and the amplitude itself is related to the light intensity.

[0021] The magnitudes of each horizontal and vertical polarization amplitude are proportional to the intensity of the backscattered light from the corresponding optical pulse, and the ratio between the two determines the polarization angle of the optical pulse.

[0022] The first, second, and third arguments all range from -π / 2 to π / 2, representing the linear polarization direction angle of the polarization state at that pulse moment. The differences between the three arguments reflect the degree of rotation of the polarization state under different time delays.

[0023] In this embodiment, the three pulses correspond to three different time delays, which can be regarded as multiple measurements of the same location in the optical fiber under different polarization probes. By decomposing the backscattered light into horizontal and vertical components and reading three sets of polarization amplitudes according to the emission sequence of the three optical pulses, the three pulse amplitude angles of each sampling point can be calculated. This expands the polarization state of each sampling point in the optical fiber link from a single intensity information to three angular information. This brings two direct benefits: first, it preserves the complete directional information of polarization state changes, enabling subsequent differentiation of rotations along different axes; second, the differential relationship between the three amplitude angles is naturally insensitive to common-mode noise, while amplifying the polarization rotation differences introduced by different disturbance locations. Therefore, it provides a high-resolution raw data foundation for identifying multi-point damage events, avoiding the multi-point disturbance masking problem caused by information loss when relying solely on intensity or a single parameter polarization angle.

[0024] S2. Determine the first differential polarization angle, the second differential polarization angle, and the differential polarization angle difference based on the pulse amplitude angle, and calculate the average residual angle and the average differential polarization angle difference of the current detection window in combination with the pulse intensity.

[0025] Considering that polarization state changes in optical fibers include not only overall polarization rotation but also variations in rotation rate at different time scales, a single differential angle cannot capture this rate difference. Since strong disturbances in real-world optical fiber links can cause inconsistent changes in polarization rotation across different time intervals, and this inconsistency exhibits different characteristics at different disturbance locations, a parameter capable of quantifying the difference between two differential angles is needed. Using only a single differential parameter would confuse the information about the difference between the rotation angle and the differential polarization angle, making it impossible to accurately determine the location and intensity of the disturbance center. Therefore, a quantitative description of the polarization state at each sampling point is required.

[0026] The process of determining the first differential polarization angle, the second differential polarization angle, and the differential polarization angle difference is as follows: S201, for each sampling point, the three phase angles in the pulse phase angle are sorted according to the emission timing of the probe light signal to determine the first phase angle, the second phase angle, and the third phase angle.

[0027] S202. Subtract the second angle from the first angle to obtain the first differential polarization angle, and subtract the second angle from the third angle to obtain the second differential polarization angle.

[0028] S203. Divide the absolute value of the difference between the first differential polarization angle and the second differential polarization angle by the sum of the first differential polarization angle and the second differential polarization angle to obtain the differential polarization angle difference degree.

[0029] The first differential polarization angle represents the change in the linear polarization direction angle of the polarization state within the time interval between two adjacent optical pulses, theoretically ranging from negative π to positive π. A positive value indicates counterclockwise rotation, and a negative value indicates clockwise rotation. The larger the absolute value, the faster the polarization rotation within that time interval, making it more sensitive to local disturbances in the optical fiber; abrupt changes in the first differential polarization angle may occur near the disturbance center.

[0030] The second differential polarization angle represents the change in the linear polarization direction angle of the polarization state over a time span of one optical pulse, and its numerical range is also from negative π to positive π. Compared with the first differential polarization angle, the second differential polarization angle corresponds to a longer time interval, and therefore is more sensitive to slowly changing polarization rotations, while it may exhibit different numerical characteristics for local drastic disturbances.

[0031] The differential polarization angle difference value ranges from 0 to 1. A value of 0 indicates that the first differential polarization angle is equal to the second differential polarization angle, meaning that the polarization rotation is completely consistent within adjacent time intervals and within a time interval separated by one pulse, indicating uniform rotation. A value of 1 indicates that the first and second differential polarization angles are opposites, meaning that the rotation directions are opposite within the two time intervals, indicating extremely non-uniform rotation. A value between 0 and 1 indicates that there are different degrees of difference in the rotation amount; the larger the value, the more significant the difference, reflecting a higher degree of non-uniformity in polarization rotation.

[0032] It should be noted that the sum of the first and second differential polarization angles may be zero. In this case, special handling is required, namely, setting the difference in differential polarization angles to 0. Since noise may exist in actual calculations causing the ratio to slightly exceed the range of 0 to 1, the system will perform truncation: if the result is less than 0, it will be set to 0; if it is greater than 1, it will be set to 1.

[0033] This embodiment utilizes the temporal relationship of the amplitudes of three optical pulses to construct a first differential polarization angle and a second differential polarization angle, and then obtains the differential polarization angle difference degree through their combination. The reason for this design is that the first differential polarization angle reflects the polarization rotation between two adjacent detector bases, while the second differential polarization angle reflects the polarization rotation separated by one pulse interval. The two have different sensitivities to polarization rotation changes at different time scales. Through their ratio relationship, a dimensionless difference degree parameter can be extracted, thereby decomposing the change in polarization state into two independent dimensions: cumulative rotation angle and rotation difference degree.

[0034] In this way, when determining whether a strong disturbance exists, not only can the rotation angle (such as the first differential polarization angle) be used, but the change in the differential polarization angle difference can also be used to help identify the disturbance type, avoiding misjudging a simple change in the differential polarization angle difference as a sudden change in the rotation angle, thereby improving the accuracy of locating each disturbance point in a multi-point damage event.

[0035] Because the intensity of backscattered light varies at different locations in the optical fiber (stronger signal near the beginning, weaker signal further away), and polarization state measurement itself introduces random noise, directly using the raw differential angle of each sampling point to determine the presence of disturbance would significantly increase the misjudgment rate in weak signal regions. Furthermore, the cumulative effect of polarization state rotation (i.e., cumulative rotation angle) needs to be gradually established starting from a disturbance-free region; otherwise, the incremental rotation caused by subsequent disturbances cannot be accurately isolated. Therefore, averaging is necessary. Specifically, the process of calculating the average residual angle and average differential polarization angle difference of the current detection window is as follows: S204, set the initial value of the cumulative rotation angle; the initial value of the cumulative rotation angle is zero.

[0036] S205. Set the current detection point as the first sampling point at the beginning of the optical fiber, set the detection window length W, and take W consecutive sampling points from the current detection point as the starting point of the window to form the current detection window.

[0037] S206. Within the current detection window, the sum of the pulse intensities of the first and second light pulses in the probe light signal is used as the intensity confidence factor. The pulse intensity is the sum of the squares of the horizontal polarization amplitude and the squares of the vertical polarization amplitude. The initial residual angle sequence is obtained by subtracting the current cumulative rotation angle from the first differential polarization angle of each sampling point. Each value in the initial residual angle sequence is weighted using the intensity confidence factor, and the weight is equal to the ratio of the intensity confidence factor to the sum of the intensity confidence factors of all sampling points within the current detection window, resulting in a weighted residual angle sequence.

[0038] S207. The average residual angle is obtained by removing outliers and sorting and truncating the weighted residual angle sequence.

[0039] S208. Calculate the arithmetic mean of the differential polarization angle difference of each sampling point within the current detection window to obtain the average differential polarization angle difference.

[0040] The cumulative rotation angle is the total polarization rotation angle caused by various disturbances from the beginning of the optical fiber to the current detection window. The initial value is set to zero, and it is continuously accumulated and corrected based on the compensation results from the disturbance center. It is a cumulative quantity, measured in radians, and its magnitude represents the total amplitude of the polarization rotation. Positive and negative signs indicate the direction of rotation.

[0041] The current detection point is the starting sampling point number of the current detection window. It is initially set to the first sampling point at the beginning of the optical fiber. It is a location index, and its value represents the number of sampling points counting from the beginning of the optical fiber. The detection window length W is the number of consecutive sampling points contained in each detection window. It is a positive integer, preset by the designer based on the system's spatial resolution and signal-to-noise ratio requirements. A larger value results in more stable statistics but decreased positioning accuracy; a smaller value results in more sensitive positioning but is more susceptible to noise interference.

[0042] The intensity confidence factor is the sum of the intensity of the first and second optical pulses at each sampling point within the current detection window. The pulse intensity is defined as the sum of the squares of the horizontal polarization amplitude and the squares of the vertical polarization amplitude corresponding to the optical pulse. The intensity confidence factor reflects the total energy of the signal at the sampling point; higher energy results in more reliable measurements. It is a positive number, typically larger near the beginning of the fiber and smaller at the end.

[0043] The initial residual angle sequence is a series of angle values ​​obtained by subtracting the current cumulative rotation angle from the first differential polarization angle of each sampling point within the current detection window. It represents the remaining local deflection angle after removing the accumulated rotation. Theoretically, the numerical range is from -π to +π, with smaller absolute values ​​indicating less deviation from the cumulative rotation baseline. The weighted residual angle sequence is obtained by assigning a weight to each value in the initial residual angle sequence. The magnitude of the weighted residual angles is of the same order of magnitude as the initial residual angles, but the contribution of weak signal points is reduced after weighting.

[0044] The average residual angle is obtained as follows: First, obviously abnormal values ​​(such as those exceeding three standard deviations) are removed. Then, the remaining values ​​are sorted from smallest to largest, and a certain percentage is removed from both ends (such as removing the minimum and maximum values ​​of 5%). The arithmetic mean of the middle part is then calculated. The numerical value represents the average residual rotation angle within the current detection window. If its absolute value is close to 0, it indicates no disturbance; if it is large, it indicates the presence of strong disturbance.

[0045] The average differential polarization angle difference ranges from 0 to 1, representing the average level of polarization rotation difference within the current detection window. The larger the value, the stronger the non-uniformity of polarization rotation within the window.

[0046] This embodiment introduces a sliding detection window mechanism, using the fiber optic start point as a reference and employing statistics from multiple sampling points within the window to smooth noise. Furthermore, to further improve statistical reliability, the different contributions of signal strength at different sampling points to measurement accuracy are considered, and a weighting factor based on pulse intensity is introduced. The contamination of mean calculation by outliers (such as burst noise or local transient interference) is also considered, therefore outlier removal and truncated averaging methods are used. In environments with strong and multi-point disturbances, robust baselines for the mean residual angle and mean differential polarization angle difference can be obtained, thus reliably determining whether a genuine disturbance exists within the window.

[0047] S3. If the absolute value of the average residual angle is less than the preset threshold, the current detection point is moved back one sampling point. Otherwise, it is determined that there is a strong disturbance in the current detection window, and the disturbance center is determined based on the first differential polarization angle.

[0048] The process of determining the disturbance center based on the first differential polarization angle is as follows: calculate the variance of the first differential polarization angle of each sampling point within the current detection window, and take the position where the variance reaches its maximum value as the disturbance center.

[0049] In this embodiment, the location where the variance reaches its maximum value is taken as the perturbation center because the first differential polarization angle reflects the change in polarization rotation angle between adjacent sampling points. In an optical fiber link, when a strong perturbation point exists, the polarization state near that location undergoes the most drastic abrupt change, resulting in the largest difference in the first differential polarization angle values ​​on both sides of the perturbation center (one side being the stable value before the perturbation, and the other side being the new value after the perturbation). This causes the dispersion of the first differential polarization angle sequence within the local window centered at that location to reach its peak, and variance is precisely the statistical measure of data dispersion. Therefore, the sampling point with the largest variance corresponds to the location with the most drastic change in polarization rotation, which is the perturbation center.

[0050] S4. Compensate for the current cumulative rotation angle of the disturbance center, determine the verification residual angle sequence of the current detection window, and correct the compensated cumulative rotation angle in combination with the first differential polarization angle.

[0051] It should be noted that the location of the disturbance center is determined by the maximum variance, but the first difference polarization angle at that point may be distorted by noise or sampling time deviation. Directly using the difference between this point and the previous sampling point as the rotation angle increment would introduce single-point error. Furthermore, the rotation of the polarization state after a disturbance is not instantaneous; residual trend changes may still exist over several subsequent sampling points (e.g., slow recovery or oscillation after a mechanical disturbance). Therefore, a prediction window is needed to compensate for the current cumulative rotation angle of the disturbance center. Specifically, for example... Figure 3 As shown, the process of determining the verification residual angle sequence of the current detection window is as follows: S401, taking the disturbance center as the boundary, the difference between the first differential polarization angle corresponding to the disturbance center and the first differential polarization angle of the previous sampling point of the disturbance center is used as the original rotation angle increment. If the disturbance center is the first sampling point at the beginning of the optical fiber, the original rotation angle increment is set to zero.

[0052] S402. A prediction window is formed by taking P consecutive sampling points behind the perturbation center, where P is an integer between 3 and 7. The difference between the first differential polarization angle and the current cumulative rotation angle of each sampling point within the prediction window is calculated to obtain a prediction difference sequence. The prediction slope is obtained by linear fitting the prediction difference sequence. The prediction window is used to estimate the short-term trend of polarization rotation after the perturbation.

[0053] S403. The original rotation angle increment and the predicted slope are linearly weighted and summed to obtain the corrected rotation angle increment. For example, the weight of the original rotation angle increment is 0.6, and the weight of the predicted slope is 0.4.

[0054] S404. Add the corrected rotation angle increment to the current cumulative rotation angle to obtain the compensated cumulative rotation angle.

[0055] S405. Take M consecutive sampling points behind the disturbance center to form a verification window, where M is an integer between 15 and 50. For each sampling point in the verification window, calculate the difference between the second differential polarization angle and the compensated cumulative rotation angle to obtain the verification residual angle value of each sampling point.

[0056] S406. Arrange all the residual angle values ​​in the order of the sampling points to form a residual angle sequence.

[0057] The original rotation angle increment represents the magnitude of the polarization angle change between adjacent sampling points at the disturbance center. Theoretically, the value range is from negative π to positive π. The larger the absolute value, the more violent the instantaneous rotation caused by the disturbance.

[0058] The predicted difference sequence reflects the amount of rotation remaining at each point within the prediction window after removing the accumulated rotation. The value can be positive or negative, and its trend (such as linear increase or decrease) is described by the prediction slope.

[0059] The prediction slope is the slope value obtained by linearly fitting the prediction difference sequence (such as the least squares method). It represents the rate of change of the residual rotation angle within the prediction window. The value can be positive or negative. The larger the absolute value, the faster the polarization state is rotating after the perturbation.

[0060] The corrected rotation angle increment integrates abrupt change and trend information, and its value lies between the original increment and the predicted slope, indicating a more reliable rotation increment estimate. The compensated cumulative rotation angle represents the total polarization rotation from the fiber optic start to the area behind the disturbance center after this disturbance compensation, and its value increases or decreases as the disturbance accumulates.

[0061] The residual angle values ​​reflect the residual system bias after compensation. If the values ​​deviate entirely from zero, it indicates inaccurate compensation; if they fluctuate randomly around zero, it indicates good compensation. The residual angle sequence is a sequence of all residual angle values ​​arranged in the order of sampling points. It is used to determine whether a secondary correction is needed.

[0062] This embodiment uses the difference between the first differential polarization angle of multiple sampling points after perturbation and the current cumulative rotation angle for linear fitting to capture the trend slope. The original single-point increment is then weighted and fused with the trend slope. The aim is to retain the abrupt change information at the perturbation center while using the subsequent trend to correct possible single-point deviations. Furthermore, to verify the compensation effect, a second differential polarization angle (instead of the first) is used to construct a verification residual angle sequence. This is because the second differential polarization angle corresponds to a longer time interval and is more sensitive to deviations in the cumulative rotation angle, allowing for a more rigorous test of whether the compensated residual angle tends to zero. This achieves a robust compensated cumulative rotation angle even in the presence of noise and residual fluctuations after perturbation, providing a reliable benchmark for subsequent multi-point splitting.

[0063] In addition, using the second differential polarization angle to construct the verification residual angle sequence can detect whether there is still a systematic deviation after compensation with higher sensitivity: if the cumulative rotation angle after compensation is accurate, then the difference between the second differential polarization angle and the cumulative rotation angle at each sampling point in the verification window should fluctuate randomly around zero; if there is a deviation, the verification residual angle sequence will show an obvious overall shift or trend.

[0064] Since the residual angle sequence obtained in S406 is calculated based on the compensated cumulative rotation angle and the second differential polarization angle, if the compensation is accurate, the residual angle sequence should fluctuate randomly around zero with a small variance. However, in actual fiber optic links, there may be a continuous residual rotation trend behind the disturbance center (e.g., the disturbance has not completely ended or measurement noise causes overall deviation), or the polarization rotation caused by the disturbance may have a proportional relationship change between different pulse pairs (e.g., the response amplitudes of the first differential polarization angle and the second differential polarization angle are inconsistent due to polarization mode dispersion or nonlinear effects). Therefore, correction is required. The process of correcting the compensated cumulative rotation angle is as follows: S407, calculate the arithmetic mean of the residual angle sequence. If the absolute value of the arithmetic mean is greater than the preset deviation tolerance threshold, subtract the arithmetic mean from the compensated cumulative rotation angle to obtain the first-corrected cumulative rotation angle.

[0065] S408. Based on the cumulative rotation angle after one correction, recalculate the difference between the second differential polarization angle of each sampling point in the verification window and the cumulative rotation angle after one correction to obtain the first correction verification residual angle sequence, and calculate the variance of the first correction verification residual angle sequence.

[0066] S409. If the variance is greater than the preset fluctuation diffusion tolerance threshold, calculate the ratio of the median of the second differential polarization angle to the median of the first differential polarization angle of all sampling points within the verification window, and use the ratio as the attenuation coefficient.

[0067] S410. Multiply the corrected rotation angle increment by the attenuation coefficient to obtain the second corrected rotation angle increment, and replace the corrected rotation angle increment with the second corrected rotation angle increment. Then, re-execute the steps starting from adding the corrected rotation angle increment to the current cumulative rotation angle to obtain the second corrected cumulative rotation angle.

[0068] S411. Output the final accumulated rotation angle as the result after compensating and correcting the current accumulated rotation angle of the disturbance center.

[0069] Among them, the deviation tolerance threshold is a preset positive threshold, such as 0.05 radians, used to determine whether the absolute value of the arithmetic mean is large enough to require correction. It determines the sensitivity of the correction. The smaller the threshold, the more frequent the correction. The larger the threshold, the greater the residual deviation allowed.

[0070] The fluctuation diffusion tolerance threshold is a preset positive threshold used to determine whether the variance is too large.

[0071] The attenuation coefficient theoretically reflects the ratio of the responses of the two differential polarization angles within the same window. If the ratio is close to 1, it indicates that the two amplitudes are comparable; if it is significantly less than 1 or greater than 1, it indicates the presence of nonlinearity or proportional deviation, which is used to correct for the rotation angle increment.

[0072] This embodiment employs a two-stage correction strategy: First, the overall offset is eliminated by subtracting the arithmetic mean of the sequence, a simple and effective zero-bias correction. Second, if the variance of the corrected sequence is still too large, it indicates that the compensation amount itself may have an amplitude error. In this case, the ratio of the median of the second differential polarization angle to the first differential polarization angle is used as an attenuation coefficient to readjust the rotation angle increment. This approach can adapt to different types of disturbances (instantaneous changes or slow variations) and different noise levels, ensuring that the compensation result of the cumulative rotation angle tends to be accurate under various conditions.

[0073] S5. After correction, if the corresponding residual angle sequence meets the set conditions, the current detection window is determined as the interval to be split, and after re-determining the perturbation center, secondary compensation is performed based on the second differential polarization angle until there is no interval to be split.

[0074] Since the preceding steps compensated and corrected the cumulative rotation angle of the disturbance center when dealing with the first strong disturbance, other damage points may still exist in the region behind the disturbance (i.e., the side away from the fiber start end), and the compensation for the first disturbance may not completely eliminate its change to the polarization ground state of the subsequent sampling points. If the cumulative rotation angle after the first compensation is directly used to calculate the residual angle of all subsequent sampling points, the residual angles corresponding to the damage points may still be masked.

[0075] Therefore, the concept of a splitting interval is introduced: the presence of insufficiently compensated residual fluctuations within the current window is determined by verifying the median deviation of the residual angle sequence. The median deviation is more robust to outliers than the variance and can reliably indicate the dispersion of the residual angle sequence. When the dispersion exceeds a preset threshold, it indicates that the current detection window still contains at least one disturbance that has not been independently processed. This window needs to be split, and secondary compensation is performed using a second differential polarization angle (more sensitive to cumulative rotation angle deviation) based on the new disturbance center within this window. This recursive splitting is to achieve the individual removal of multi-point destructive events, ensuring that each destructive point can receive accurate cumulative rotation angle compensation within its local window.

[0076] The condition is that the median deviation of the calibration residual angle sequence is greater than a preset residual fluctuation threshold. The median deviation is calculated as follows: first, the median of the calibration residual angle sequence is calculated; then, the absolute deviation of each calibration residual angle value from the median is calculated; finally, the median of all absolute deviations is taken as the median deviation.

[0077] The disturbance center is redefined by recalculating the variance of the first difference polarization angle of each sampling point within the interval to be split, and taking the position where the variance reaches its maximum value as the new disturbance center.

[0078] Specifically, the process of performing secondary compensation based on the second differential polarization angle until there is no interval to be split is as follows: S501, calculate the difference between the second differential polarization angle of the M consecutive sampling points behind the new disturbance center and the current cumulative rotation angle to obtain the secondary compensation verification residual angle sequence.

[0079] S502. Calculate the arithmetic mean of the residual angle sequence after secondary compensation, and subtract the arithmetic mean from the current cumulative rotation angle to obtain the cumulative rotation angle after secondary compensation.

[0080] S503. Update the final residual angle of each sampling point in the sampling point within the interval to be split according to the cumulative rotation angle after secondary compensation.

[0081] S504. Take the endpoint of the interval to be split as the new current detection point, and then return to execute step S4 (that is, compensate for the current cumulative rotation angle of the disturbance center, determine the verification residual angle sequence of the current detection window, and correct the compensated cumulative rotation angle in combination with the first differential polarization angle) and the judgment condition in step S5, that is, re-determine whether the corrected verification residual angle sequence meets the set conditions. If it does, continue splitting until no new intervals to be split are generated.

[0082] The median deviation is a robust statistic that measures the dispersion of a residual angle sequence. It is calculated by first finding the median of the sequence, then calculating the absolute deviation of each value from the median, and finally taking the median of these absolute deviations. Since the median deviation has the same dimension as the residual angle (radians), it represents the magnitude of the typical absolute deviation in the sequence. A larger median deviation indicates more severe fluctuations in the residual angle.

[0083] The residual fluctuation threshold is a preset positive threshold used to determine whether the median deviation is too large, for example, 0.1 radians. When the median deviation exceeds this threshold, it is considered that there are still insufficiently compensated disturbances within the current detection window, requiring further splitting. The interval to be split is a range of continuous sampling points that requires secondary compensation, representing a section of the optical fiber containing at least two independent disturbances or one disturbance region that has not yet been fully compensated.

[0084] The final residual angle is the residual angle value used to determine the location of the damage point after all compensations (including the original compensation and the secondary compensation of each subsequent interval to be split) are applied to each sampling point. The absolute value of the final residual angle reflects the amount of polarization rotation remaining at the sampling point after removing all known disturbances, and is used to compare with the alarm threshold.

[0085] The endpoint of the interval to be split is the position of the last sampling point of the current interval to be split, which is determined by the starting point of the current detection window and the window length W (the starting point of the window is the current detection point, and the endpoint is the starting point + W - 1).

[0086] This embodiment uses the median deviation as a criterion to automatically identify whether there are still residual fluctuations that have not been eliminated within the current detection window, thereby triggering recursive processing of the interval to be split. Its advantages are: First, using the median deviation instead of the variance effectively resists local spike noise that may exist in the residual angle sequence, making the splitting judgment more robust; Second, in the secondary compensation, the difference between the second differential polarization angle and the current cumulative rotation angle is directly used, and the cumulative rotation angle is corrected only by subtracting the arithmetic mean, avoiding complex weighting and slope prediction, resulting in high computational efficiency and sufficient effectiveness for already split local intervals; Third, by using the endpoint of the interval to be split as the new reference position and recursively executing, each destruction point can be processed sequentially from front to back, ultimately ensuring that the final residual angle of all sampling points is calculated based on the independent cumulative rotation angle of their respective intervals, thereby simultaneously detecting the spatial positions of multiple destruction points.

[0087] S6. Determine the spatial location of each damage point based on the final residual angle of each sampling point.

[0088] Considering that after multi-point splitting and compensation, the final residual angles corresponding to multiple damage points on the optical fiber will form local peaks near their respective locations, but the same damage event may cause the residual angles of multiple consecutive sampling points to exceed the alarm threshold, and the distance between different damage points may be close to or less than the Rayleigh coherence length of the optical fiber (at which point the two damage points cannot be physically distinguished as independent events). If all sampling points exceeding the threshold are simply output as damage points, a large number of redundant alarms and false locations will be generated.

[0089] Therefore, a clustering and merging strategy is adopted: first, candidate damage points are screened by threshold, and then the number of sampling points corresponding to the Rayleigh coherence length of the fiber is used to determine which candidate points belong to the same damage event (when the spatial distance is too close, they should be regarded as the response interval of the same damage source). Then, the point with the largest absolute value of the residual angle in each cluster set is selected as the representative position of the damage event. Under the limitation of spatial resolution, the exact position of each independent damage event can be output in the most robust way, avoiding false alarms and duplicate reports.

[0090] The specific process is as follows: S601, compare the absolute value of the final residual angle corresponding to each sampling point with the preset multi-point damage event alarm threshold, and mark the sampling points whose absolute values ​​exceed the multi-point damage event alarm threshold as candidate damage points.

[0091] S602. Scan all candidate damage points from front to back along the fiber position. When the spatial distance between two adjacent candidate damage points is less than half the number of sampling points corresponding to the Rayleigh coherence length of the fiber, these two adjacent candidate damage points are grouped into the same candidate damage point set.

[0092] S603. In each set of candidate failure points, the candidate failure point with the largest absolute value of the final residual angle is taken as the representative failure point of the candidate failure point set.

[0093] S604. When the number of representative damage points is greater than or equal to 2, and the spatial distance between any two representative damage points is greater than the number of sampling points corresponding to the Rayleigh coherence length of the optical fiber, it is determined to be a multi-point damage event, and the spatial positions of all representative damage points are output as the spatial positions of each damage point.

[0094] S605. When the number of representative destruction points is 1, it is determined to be a single-point destruction event, and the spatial location of the representative destruction point is output.

[0095] S606. When the number of destruction points is 0, it is determined as no destruction event.

[0096] The multi-point damage event alarm threshold is a preset positive threshold value used to determine whether the absolute value of the final residual angle meets the damage alarm condition, such as 0.2 radians. The smaller the value, the higher the sensitivity but the more false alarms there will be; the larger the value, the more likely it is to miss an alarm.

[0097] The Rayleigh coherence length of an optical fiber is the coherence length of the backscattered Rayleigh light in the fiber, determined by the linewidth of the light source, and typically ranges from a few meters to tens of meters. It represents the physical limit of spatial resolution; if the distance between two destruction points is less than this length, they cannot be distinguished in the detection signal. The number of sampling points corresponding to the Rayleigh coherence length is the integer (or rounded down) obtained by dividing the Rayleigh coherence length by the system's spatial resolution (i.e., the fiber length between adjacent sampling points), representing the minimum sampling interval between two spatially indistinguishable destruction points.

[0098] This embodiment uses absolute value comparison, neighborhood clustering, representative point selection, and spacing judgment to transform continuous residual angle waveforms into a discrete list of destructive events. Only when the number of representative destructive points is greater than or equal to two and the distance between them is greater than the Rayleigh coherence length is it confirmed as a true multi-point destructive event; if there is only one representative point, it is a single-point destructive event; if there are no representative points, there is no destructive event. The advantages are: First, using alarm thresholds to filter candidate points avoids misjudging minor noise fluctuations as destructive events; second, clustering and merging based on the Rayleigh coherence length ensures that the same physical destructive event (which may cause multiple consecutive sampling point responses) outputs only one representative location, avoiding duplicate alarms; third, by statistically analyzing the number of representative points and their spacing, it is possible to clearly distinguish between single-point destructive, multi-point destructive, and no-destructive situations, providing a clear basis for subsequent operation and maintenance decisions; fourth, the representative point is selected based on the point with the largest absolute value of the residual angle within the set, which is closest to the actual disturbance center, improving positioning accuracy. In this way, the entire monitoring and early warning method can ultimately output the spatial location of one or more damage points in the fiber optic link stably and accurately, solving the technical bottleneck that multiple damage events cannot be detected simultaneously.

[0099] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0100] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0101] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0102] The above description is merely a 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.

[0103] Finally, 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 method for monitoring and early warning of the operational transmission status of optical fiber communication, characterized in that, Includes the following steps: Polarization analysis is performed on the backscattered light signal corresponding to the probe light signal, and the pulse amplitude angle of each sampling point is calculated in combination with the emission timing of the probe light signal; The first differential polarization angle, the second differential polarization angle, and the differential polarization angle difference are determined based on the pulse amplitude. At the same time, the average residual angle and the average differential polarization angle difference of the current detection window are calculated in combination with the pulse intensity. If the absolute value of the average residual angle is less than the preset threshold, the current detection point will be moved to the next sampling point. Otherwise, it is determined that there is a strong disturbance in the current detection window, and the disturbance center is determined based on the first differential polarization angle; The current cumulative rotation angle of the disturbance center is compensated, the verification residual angle sequence of the current detection window is determined, and the compensated cumulative rotation angle is corrected by combining the first differential polarization angle. After correction, if the corresponding residual angle sequence of the verification meets the set conditions, the current detection window is determined as the interval to be split, and the disturbance center is re-determined and secondary compensation is performed based on the second differential polarization angle until there is no interval to be split. The spatial location of each damage point is determined based on the final residual angle of each sampling point.

2. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The process of calculating the pulse amplitude at each sampling point is as follows: The backscattered light signal is decomposed into a horizontal polarization component and a vertical polarization component, which are then converted into a first electrical signal and a second electrical signal, respectively. Based on the emission timing of the three optical pulses in the detection optical signal, the horizontal polarization amplitude and vertical polarization amplitude corresponding to the three optical pulses at each sampling point are extracted from the first and second electrical signals. The arctangent function value of the ratio of the vertical polarization amplitude to the horizontal polarization amplitude is used as the amplitude of the corresponding optical pulse; The three amplitudes are output as pulse amplitudes.

3. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The process of determining the first differential polarization angle, the second differential polarization angle, and the degree of difference in differential polarization angles is as follows: For each sampling point, the three phases of the pulse are sorted according to the emission sequence of the probe light signal to determine the first, second, and third phases; The first differential polarization angle is obtained by subtracting the second differential polarization angle from the first differential polarization angle, and the second differential polarization angle is obtained by subtracting the second differential polarization angle from the third differential polarization angle. The difference in polarization angles is obtained by dividing the absolute value of the difference between the first and second differential polarization angles by the sum of the first and second differential polarization angles.

4. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 3, characterized in that, The process of calculating the average residual angle and average differential polarization angle difference of the current detection window is as follows: Set the initial value for the cumulative rotation angle; Set the current detection point as the first sampling point at the beginning of the optical fiber, set the detection window length W, and take W consecutive sampling points from the current detection point as the starting point of the window to form the current detection window; Within the current detection window, the sum of the pulse intensities of the first and second light pulses in the probe light signal is used as the intensity confidence factor; The initial residual angle sequence is obtained by subtracting the current cumulative rotation angle from the first differential polarization angle of each sampling point. The weighted residual angle sequence is obtained by weighting each value in the initial residual angle sequence using the intensity confidence factor. The average residual angle is obtained by removing outliers and sorting and truncating the weighted residual angle sequence; Calculate the arithmetic mean of the differential polarization angle difference for each sampling point to obtain the average differential polarization angle difference.

5. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The process of determining the disturbance center based on the first differential polarization angle is as follows: Calculate the variance of the first differential polarization angle of each sampling point within the current detection window, and take the position where the variance reaches its maximum value as the perturbation center.

6. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The process of determining the verification residual angle sequence of the current detection window is as follows: If the disturbance center is the first sampling point at the beginning of the optical fiber, the original rotation angle increment is set to zero; otherwise, the difference between the first differential polarization angle corresponding to the disturbance center and the first differential polarization angle of the sampling point before the disturbance center is used as the original rotation angle increment. P consecutive sampling points are taken from behind the disturbance center to form a prediction window. The difference between the first differential polarization angle and the current cumulative rotation angle of each sampling point in the prediction window is calculated to obtain the prediction difference sequence. The prediction slope is obtained by linear fitting of the prediction difference sequence. The original rotation angle increment is linearly weighted and summed with the predicted slope to obtain the corrected rotation angle increment. Add the corrected rotation angle increment to the current cumulative rotation angle to obtain the compensated cumulative rotation angle; M consecutive sampling points are taken from behind the disturbance center to form a verification window. For each sampling point in the verification window, the difference between the second differential polarization angle and the compensated cumulative rotation angle is calculated to obtain the verification residual angle value. Arrange all the residual angle values ​​in the order of the sampling points to form a residual angle sequence.

7. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 6, characterized in that, The process of correcting the cumulative rotation angle of the compensation is as follows: If the absolute value of the arithmetic mean of the residual angle sequence is greater than the preset deviation tolerance threshold, the arithmetic mean is subtracted from the compensated cumulative rotation angle to obtain the first-corrected cumulative rotation angle. Based on the cumulative rotation angle after one correction, the difference between the second differential polarization angle of each sampling point in the verification window and the cumulative rotation angle after one correction is recalculated to obtain the first correction verification residual angle sequence, and the variance of the first correction verification residual angle sequence is calculated. If the variance is greater than the preset fluctuation diffusion tolerance threshold, the ratio of the median of the second differential polarization angle to the median of the first differential polarization angle of all sampling points in the verification window will be used as the attenuation coefficient. Multiply the corrected rotation angle increment by the attenuation coefficient to obtain the second corrected rotation angle increment, and replace the corrected rotation angle increment with the second corrected rotation angle increment. After re-correction, the cumulative rotation angle after the second correction is obtained. The final accumulated rotation angle will be output.

8. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The condition is that the median deviation of the residual angle sequence is greater than a preset residual fluctuation threshold; the median deviation is calculated as follows: First, calculate the median of the check residual angle sequence. Then, calculate the absolute deviation of each check residual angle value from the median. Finally, take the median of all absolute deviations as the median of the deviations.

9. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The process of secondary compensation based on the second differential polarization angle is as follows: The residual angle sequence for secondary compensation verification is obtained based on M consecutive sampling points behind the new disturbance center. Subtract the arithmetic mean of the residual angle sequence after secondary compensation from the current cumulative rotation angle to obtain the cumulative rotation angle after secondary compensation. Update the final residual angle of each sampling point within the interval to be split according to the cumulative rotation angle after secondary compensation.

10. The method for monitoring and early warning of optical fiber communication operation and transmission status according to claim 1, characterized in that, The process of determining the spatial location of each damage point is as follows: Sampling points whose absolute value of the final residual angle exceeds the preset multi-point damage event alarm threshold are marked as candidate damage points; When the spatial distance between two adjacent candidate failure points is less than half the number of sampling points corresponding to the Rayleigh coherence length of the optical fiber, these two adjacent candidate failure points are grouped into the same candidate failure point set. In each set of candidate failure points, the candidate failure point with the largest absolute value of the final residual angle is selected as the representative failure point of the candidate failure point set; When the number of representative damage points is greater than or equal to 2, and the spatial distance between any two representative damage points is greater than the number of sampling points corresponding to the Rayleigh coherence length of the optical fiber, it is determined to be a multi-point damage event, and the spatial positions of all representative damage points are output as the spatial positions of each damage point. When the number of destruction points is 1, it is determined to be a single-point destruction event, and the spatial location of the destruction point is output. When the number of destruction points is 0, it is determined as a non-destruction event.

Images