CN-122027025-A - Optical fiber transmission link quality monitoring and dispersion compensation optimizing method and system

CN122027025ACN 122027025 ACN122027025 ACN 122027025ACN-122027025-A

Abstract

The invention belongs to the technical field of optical fiber communication, and discloses an optical fiber transmission link quality monitoring and dispersion compensation optimizing method and system, which comprises extracting error vector amplitude, optical signal to noise ratio and constellation diagram distortion characteristics from a coherent receiver and fusing the characteristics into a comprehensive quality evaluation index, and (3) jointly estimating residual dispersion by using the clock recovery phase and equalizer coefficient distribution, adaptively optimizing the dispersion compensation filter parameters by adopting variable step gradient descent, and carrying out performance degradation trend analysis and early warning, and multi-wavelength channel joint monitoring and collaborative optimization.

Inventors

ZHANG YIDAN
SHI WENZAO
SHI ZHANQI
LI KAIXIN
ZHANG YILU

Assignees

福建师范大学

Dates

Publication Date: 20260512
Application Date: 20260409

Claims (10)

1. The method for monitoring the quality of the optical fiber transmission link and optimizing the dispersion compensation is characterized by comprising the following steps: S1, extracting optical signal quality parameters in real time, namely acquiring error vector amplitude, an optical signal to noise ratio estimated value and constellation diagram distortion characteristics from a digital signal processing unit of a coherent receiver, and carrying out weighted fusion on the error vector amplitude, the optical signal to noise ratio estimated value and the constellation diagram distortion characteristics to obtain a comprehensive transmission quality evaluation index; S2, extracting the phase change rate of a phase-locked error signal of a clock recovery loop in a digital signal processing unit, reading each tap coefficient of the adaptive equalizer, calculating the energy distribution gravity center offset, and carrying out joint mapping on the phase change rate and the energy distribution gravity center offset to obtain a residual dispersion estimated value; S3, a dispersion compensation parameter self-adaptive optimization step, namely dynamically adjusting each tap coefficient of the digital dispersion compensation filter by adopting a variable step gradient descent algorithm by taking the comprehensive transmission quality evaluation index as an optimization objective function, locking the current tap coefficient when the change rate of the comprehensive transmission quality evaluation index is smaller than a preset convergence threshold, and restarting the optimization process when the comprehensive transmission quality evaluation index is deteriorated to exceed the preset fluctuation threshold; Step S4, performance degradation trend analysis and early warning, namely periodically recording a historical data sequence of comprehensive transmission quality evaluation indexes and residual dispersion estimated values in a preset time window, performing linear regression analysis on the historical data sequence to obtain degradation slope, calculating the residual time expected to reach a fault threshold when the degradation slope exceeds a trend alarm threshold, and generating maintenance early warning information; And S5, a multi-wavelength channel joint monitoring step, namely respectively executing steps S1 to S4 on each wavelength channel in the dense wavelength division multiplexing system, converging quality indexes of each channel to form a full channel quality matrix, identifying degraded channels, and taking a dispersion compensation optimization result of each channel as a reference value for initializing adjacent channel compensation parameters.
2. The method according to claim 1, wherein in the step S1, the error vector magnitude is calculated by adopting a normalized root mean square deviation method, the range of the error vector magnitude is 0% to 100%, the effective range of the osnr estimation value is 10dB to 40dB, the range of the integrated transmission quality estimation index is 0 to 1, the range of the weight coefficient of the error vector magnitude in the weighted fusion is 0.3 to 0.5, the range of the weight coefficient of the osnr estimation value is 0.3 to 0.4, and the range of the weight coefficient of the constellation distortion feature is 0.1 to 0.3.
3. The method according to claim 1, wherein in the step S2, the phase change rate is obtained by performing differential operation on a phase-locked error signal in a continuous sampling period, the sensitivity is not lower than 0.5ps/nm, the energy distribution gravity center offset is a distance between a weighted sum value of a square of a modulus value of each tap coefficient and a tap sequence number and a center tap sequence number, and the joint mapping is calibrated by using a two-dimensional linear regression model and taking a pre-calibrated dispersion-characteristic comparison table as a reference.
4. The method according to claim 1, wherein in the step S3, the initial step factor of the variable step gradient descent algorithm ranges from 0.001 to 0.01, the adjustment range of the step factor ranges from 0.1 to 10 times of the initial step factor, the preset convergence threshold ranges from 0.0001 to 0.001, and the preset fluctuation threshold ranges from 1% to 5% of the current value of the integrated transmission quality evaluation index.
5. The method for monitoring the quality of an optical fiber transmission link and compensating for dispersion according to claim 1, wherein in the step S1, the calculation process of the error vector amplitude includes collecting complex sample values of the received signal after equalization in each symbol period, taking complex differences between the complex sample values and ideal constellation points of the latest decision as error vectors, carrying out root mean square statistics on the modulus values of the error vectors in N continuous symbol periods, and dividing the modulus values by the root mean square amplitude of the ideal constellation points, wherein the value range of N is 1024 to 65536, and the extraction process of the constellation distortion characteristics includes taking the ratio of the major axis and the minor axis of a dispersion ellipse of each constellation cluster as radial distortion factors and taking the standard deviation of the angular offset of the center of each constellation cluster relative to the ideal position as angle distortion factors.
6. The method according to claim 1, wherein in the step S2, the online estimation of the accumulated amount of chromatic dispersion further comprises determining that the equalizer is operating in a non-optimal state when the absolute value of the energy distribution center of gravity offset exceeds 25% of the total number of taps of the adaptive equalizer, and taking the phase change rate as a main reference and increasing the weight coefficient thereof in the joint mapping to not lower than 0.7, so as to reduce the influence of the non-optimal state of the equalizer on the accuracy of the chromatic dispersion estimation.
7. The method according to claim 1, wherein in the step S3, the step adjustment rule of the variable step gradient descent algorithm includes that when the signs of the integrated transmission quality evaluation index change amounts of two successive iterations are the same, the step factor is multiplied by the acceleration factor to be amplified, the value range of the acceleration factor is 1.05 to 1.2, and when the signs of the integrated transmission quality evaluation index change amounts of two successive iterations are opposite, the step factor is multiplied by the deceleration factor to be reduced, and the value range of the deceleration factor is 0.5 to 0.8.
8. The method according to claim 1, wherein in the step S4, the length of the predetermined time window ranges from 1h to 24h, the sampling interval ranges from 1min to 10min, the linear regression analysis further includes calculating a decision coefficient to evaluate the regression fit, and the decision index change does not have a significant trend and skips the early warning judgment when the decision coefficient is lower than 0.6, and the maintenance early warning information includes a degradation index type, a current value, a degradation slope, a remaining time expected to reach a failure threshold, and a recommended maintenance operation type.
9. The method for monitoring the quality of an optical fiber transmission link and optimizing the dispersion compensation according to claim 1, wherein in the step S5, the full channel quality matrix is a two-dimensional matrix with a wavelength channel number as a row index and a quality index type as a column index, and the initializing the dispersion compensation optimization result of each channel as a reference value of the adjacent channel compensation parameters specifically includes reading tap coefficients of a dispersion compensation filter of an existing channel having the smallest wavelength interval with the newly added channel when the newly added wavelength channel is turned on, and performing linear extrapolation correction on the tap coefficients according to a product of the wavelength interval and the dispersion slope to obtain initial tap coefficients of the newly added channel.
10. The optical fiber transmission link quality monitoring and dispersion compensation optimizing system is used for realizing the optical fiber transmission link quality monitoring and dispersion compensation optimizing method according to any one of claims 1-9, and is characterized by comprising an optical signal quality parameter real-time extraction module, a dispersion accumulation on-line estimation module, a dispersion compensation parameter self-adaptive optimizing module, a performance degradation trend analysis and early warning module and a multi-wavelength channel joint monitoring module; The optical signal quality parameter real-time extraction module is used for acquiring error vector amplitude, an optical signal to noise ratio estimated value and constellation diagram distortion characteristics from a digital signal processing unit of the coherent receiver, and carrying out weighted fusion on the three to obtain a comprehensive transmission quality evaluation index; the dispersion accumulation on-line estimation module is used for acquiring a phase-locked error signal of the clock recovery loop, extracting a phase change rate, reading each tap coefficient of the adaptive equalizer, calculating energy distribution gravity center offset, and carrying out joint mapping on the phase change rate and the energy distribution gravity center offset to obtain a residual dispersion estimation value; The dispersion compensation parameter self-adaptive optimization module is used for dynamically adjusting each tap coefficient of the digital dispersion compensation filter by adopting a variable step gradient descent algorithm to continuously track an optimal dispersion compensation working point by taking the comprehensive transmission quality evaluation index as an optimization objective function according to the residual dispersion estimation value; The performance degradation trend analysis and early warning module is used for periodically recording historical data sequences of various quality indexes, performing linear regression analysis on the historical data sequences to obtain degradation slopes, and generating maintenance early warning information when the degradation slopes exceed trend warning thresholds; The multi-wavelength channel joint monitoring module is used for converging the quality indexes of the wavelength channels to form a full channel quality matrix, identifying degraded channels and taking the dispersion compensation optimization result of each channel as a reference value for initializing the adjacent channel compensation parameters.

Description

Optical fiber transmission link quality monitoring and dispersion compensation optimizing method and system Technical Field The invention belongs to the technical field of optical fiber communication, and particularly relates to an optical fiber transmission link quality monitoring and dispersion compensation optimizing method and system. Background With the rapid development of 5G bearer networks and data center interconnection services, optical fiber transmission systems continue to evolve toward higher rates, longer distances, and greater capacities. In long-distance coherent optical transmission systems, chromatic dispersion is one of the key factors limiting transmission performance. The dispersion effect causes the optical pulses to spread stepwise during transmission, causing intersymbol interference and significantly degrading the signal quality at the receiving end. For standard single mode optical fibers, the dispersion coefficient at 1550nm band is about 17ps/nm/km, and the accumulated dispersion can reach 17000ps/nm at 1000km transmission distance, so that the accumulated dispersion can be completely unrecoverable without compensation. At present, a dispersion compensation technology based on digital signal processing has become a core technical means in a coherent optical communication system, and accumulated dispersion is electrically compensated by utilizing a digital filter at a receiving end, so that the traditional optical domain dispersion compensation scheme such as a dispersion compensation optical fiber can be effectively replaced, and the cost and complexity of the system are reduced. However, long-haul fiber optic transmission links present a number of challenges in a practical operating environment. On the one hand, the dispersion coefficient of the optical fiber slowly drifts under the influence of environmental temperature change, material aging and other factors, wherein the dispersion coefficient change rate caused by temperature is about 0.003ps/nm/km/°c, and under the condition that the temperature difference between day and night is 20 ℃, the dispersion drift amount of a 1000km link can reach 60ps/nm, so that the dispersion compensation parameter which is initially set gradually deviates from the optimal working point, and continuous degradation of transmission performance is caused. On the other hand, link quality monitoring schemes in existing optical fiber communication systems generally adopt an offline measurement or periodic inspection mode, for example, optical time domain reflectometry is used for optical fiber link test or transmission quality is estimated through bit error rate statistics, and these methods cannot timely sense progressive degradation of transmission quality, and often cannot find problems after performance degradation has affected service quality. In addition, in the dense wavelength division multiplexing system, a scene of co-fiber transmission of a plurality of wavelength channels exists, dispersion characteristics among the channels are different due to different wavelengths, and higher requirements are put on unified monitoring and collaborative optimization of all the channels. The Chinese patent with publication number CN116318406B discloses a signal compensation method and system for an optical fiber communication system, wherein the method reduces distortion and noise by preprocessing before signal transmission, eliminates distortion and noise in the transmission process by performing subsequent processing after the signal is transmitted to a receiving end, and detects and classifies physical breakage of an optical fiber cable by combining an image recognition technology to implement targeted signal compensation. The scheme provides a signal compensation strategy in three dimensions of forward compensation, backward compensation and line loss identification, but has the following defects that firstly, the scheme focuses on cable breakage detection and basic signal forward and backward compensation in a physical layer, no refined index reflecting transmission quality is involved in extracting error vector amplitude, optical signal to noise ratio, constellation diagram distortion and the like from digital signal processing parameters in real time, secondly, an online estimation mechanism of dispersion accumulation amount is not established, link dispersion change cannot be dynamically perceived under the condition of no service interruption, and compensation parameters cannot be adaptively adjusted, thirdly, the scheme lacks analysis and prediction capability of long-term evolution trend of the transmission quality, and does not have the function of giving early warning in the early stage of performance degradation, and fourthly, the scheme does not consider the joint monitoring and collaborative optimization requirements of multi-wavelength channels under a dense wavelength division multiplexing scene. Therefore, a method and system for on-li