CN116155371A - OTDR event detection method - Google Patents
OTDR event detection method Download PDFInfo
- Publication number
- CN116155371A CN116155371A CN202310111028.1A CN202310111028A CN116155371A CN 116155371 A CN116155371 A CN 116155371A CN 202310111028 A CN202310111028 A CN 202310111028A CN 116155371 A CN116155371 A CN 116155371A
- Authority
- CN
- China
- Prior art keywords
- event
- suspected
- reflection
- otdr
- data segment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
- H04B10/071—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using a reflected signal, e.g. using optical time domain reflectometers [OTDR]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/07—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems
- H04B10/075—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal
- H04B10/079—Arrangements for monitoring or testing transmission systems; Arrangements for fault measurement of transmission systems using an in-service signal using measurements of the data signal
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Optical Communication System (AREA)
Abstract
The invention discloses an OTDR event detection method, which relates to the technical field of data analysis, and comprises the following steps: acquiring an OTDR response curve of a communication optical fiber to be tested, and performing wavelet denoising treatment to obtain wavelet coefficients decomposed into layers; an improved simulated annealing algorithm is adopted to obtain the optimal threshold value of each layer of wavelet coefficient, threshold processing is carried out on each layer of wavelet coefficient, and a denoised OTDR response curve is obtained by reconstructing the signal; extracting an intermediate data segment of the denoised response curve, searching for suspected reflection events and non-suspected reflection events in the intermediate data segment, removing misjudgment points caused by noise in the suspected events according to the event interval distance and the event duration distance, and finally determining reflection event position information, non-reflection event position information and attenuation loss of each event. By adopting the method, a response curve with higher signal-to-noise ratio can be obtained, and the magnitude of each threshold value is intelligently obtained through an improved simulated annealing algorithm, so that the error probability of a report event is reduced.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to an OTDR event detection method.
Background
With the rapid development of the optical communication network in China, the communication optical fiber can often reach tens or even hundreds of kilometers, and the communication optical fiber is not only on land surface but also often distributed under water, and if the optical fiber is bent, broken and the like during communication, the communication quality and even the communication failure can be influenced. To ensure the quality of communication service, a device is designed to detect the working condition of the communication optical fiber, and to quickly detect and accurately locate the possible events on the optical fiber.
At present, an optical time domain reflectometer (Optical Time Domain Reflectometry, OTDR) is generally adopted for detecting communication optical fibers at home and abroad, and the OTDR is used as equipment for detecting the working condition of the communication optical fibers in an optical communication network and is used for detecting the power of back Rayleigh scattered light and Fresnel reflected light generated when light is transmitted in the optical fibers. In the prior art, the dynamic range and the initial end blind area position are usually obtained by manually comparing the horizontal coordinates and the vertical coordinates in the response curve graph; in addition, by analyzing the abrupt ascending or descending trend in the response curve graph, whether the event exists in the communication optical fiber can be judged, the occurrence position corresponding to each event is found out on the abscissa, and the event attenuation loss is calculated.
In practical detection, the OTDR system is affected by noise, and the noise level is typically reduced by storing and cumulatively averaging the received data multiple times by means of a cumulatively averaging method for the denoising process of the OTDR response curve. Therefore, researchers need to spend a great deal of time and effort on detecting data acquisition and accumulating average to improve the signal to noise ratio of a response curve, the accuracy of measuring the dynamic range, the initial dead zone and the event information accuracy of an OTDR system by a manual visual inspection and slope method is not high, the event is easy to miss or excessively report, and finally, a certain time is spent on manually recording the working condition.
Disclosure of Invention
The inventor aims at the problems and the technical requirements and provides an OTDR event detection method, which is realized based on a wavelet transformation algorithm and an improved simulated annealing algorithm, wherein the wavelet transformation algorithm is used for denoising an OTDR response curve, calculating a dynamic range and a starting end blind area, and solving signal singular points and mutation points so as to calculate event position information and event attenuation loss; the improved simulated annealing algorithm is used for selecting the threshold value in the processes of denoising treatment and solving signal singular points and abrupt points, and selecting a proper threshold value according to different signal lengths and each section of signal range information. The technical scheme of the invention is as follows:
an OTDR event detection method includes the following steps:
acquiring an OTDR response curve of a communication optical fiber to be tested, and performing wavelet denoising treatment to obtain wavelet coefficients decomposed into layers; wherein the abscissa of the OTDR response curve is the measurement distance, and the ordinate is the relative power intensity;
obtaining an optimal threshold value of each layer of wavelet coefficient by adopting an improved simulated annealing algorithm;
performing threshold processing on the wavelet coefficients of each layer according to the optimal threshold, and using the wavelet coefficients subjected to the threshold processing for reconstructing signals to obtain a denoised OTDR response curve;
extracting an intermediate data segment of the denoised OTDR response curve, searching a suspected reflection event and a non-suspected reflection event in the intermediate data segment, removing misjudgment points caused by noise in the suspected event according to the event interval distance and the event duration distance, and finally determining reflection event position information, non-reflection event position information and attenuation loss of each event;
the event interval distance and the event duration distance are fixed values set according to a resolution space, and the resolution space is a propagation distance corresponding to a pulse width.
The beneficial technical effects of the invention are as follows:
the method is suitable for an OTDR system to detect the working condition of the optical fiber in the optical communication network. The method can save the time of storing multiple times of received data, obtain an OTDR response curve with higher signal-to-noise ratio, calculate the mode maximum value of the curve through a wavelet transformation algorithm, calculate the position information of each singular point and each mutation point according to the mode maximum value, and further analyze the curve information to obtain more detailed information such as the size of a dynamic range, the range of a dead end blind area, the position information of a reflection event/non-reflection event, the attenuation loss of the event and the like; in the response curve analysis process, the optimal threshold value can be obtained intelligently through an improved simulated annealing algorithm, a corresponding annealing function can be selected according to the change rate of the objective function, the disturbance range of the threshold value is folded, the annealing temperature is controlled more effectively, and the algorithm has the opportunity of jumping out of a local optimal solution; the method is convenient and high in precision, the error probability of reporting the event is greatly reduced, a powerful technical support is provided for detecting the working condition of the communication optical fiber, and meanwhile, the intelligent detection of the communication optical fiber level is improved.
Drawings
Fig. 1 is a flowchart of an OTDR event detection method provided in the present application.
Fig. 2 is an OTDR response curve without denoising provided in the present application.
FIG. 3 is a flow chart of the improved simulated annealing algorithm provided herein for thresholding wavelet coefficients optimally.
Fig. 4 is an OTDR response curve after wavelet denoising and tail end noise removal provided herein.
Fig. 5 is a code diagram of an algorithm for obtaining a modulus maximum provided in the present application.
Fig. 6 is a schematic diagram of the mode maxima of the useful signal provided herein.
Fig. 7 is a code diagram of an algorithm for locating the start position of a reflection event provided herein.
Fig. 8 is a graph of the slope of the intermediate data segment after removal of the beginning dead zone and the end position provided herein.
Fig. 9 is a document screenshot of the automatic recording of the working condition of the communication fiber under test provided by the present application.
Detailed Description
The following describes the embodiments of the present invention further with reference to the drawings.
As shown in fig. 1, the present embodiment provides an OTDR event detection method, including the following steps:
step 1: acquiring an OTDR response curve of a communication optical fiber to be measured, comprising:
step 1.1: and detecting the communication optical fiber to be detected to obtain original received data, wherein the original received data comprises acquisition time (in mu s) and corresponding acquisition voltage (in v).
Because the original data received does not contain communication fiber signals in the data with the initial acquisition time less than zero, and belongs to the OTDR system idle acquisition, the data with the initial acquisition time can be regarded as a noise floor part before the communication fiber is not detected, and the step 1.2 is executed.
Step 1.2: and extracting a data segment with the acquisition time less than zero, and carrying out average processing on the segment data to obtain the bias voltage.
Step 1.3: the data segment with the acquisition time greater than zero is considered as a signal generation section, the data of the time is extracted, the acquisition time is converted into a measurement distance (in km) as an abscissa of an OTDR response curve, the acquisition voltage is converted into a relative power intensity (in dB) as an ordinate of the OTDR response curve, and the OTDR response curve of the communication optical fiber to be measured is obtained as shown in fig. 2. Wherein the relative power strength is expressed as:
wherein S represents the relative power intensity, V represents the acquisition voltage, V 0 Represents bias voltage, V 1 Representing the acquisition voltage of the first measurement signal.
Optionally, in step 1, further includes: other fixed values required by the method are defined, such as light speed, refractive index of communication optical fiber, pulse width of optical signal, propagation distance corresponding to one pulse width (hereinafter referred to as resolution space), etc.
Step 2: and carrying out wavelet denoising treatment on the OTDR response curve to obtain wavelet coefficients decomposed into layers.
After the OTDR measurement signal is obtained, wavelet denoising processing is required to be performed on the signal, wherein the selection of a wavelet basis function and a decomposition layer number is particularly important, and in the process of selecting the wavelet basis function, symmetry, tight support orthogonality, vanishing moment and support width are required to be considered. For an OTDR system, a wavelet basis function with good symmetry and tight support orthogonality is generally adopted, the good symmetry ensures that singular points and abrupt points do not shift after wavelet transformation is carried out on signals, and the high tight support orthogonality reduces the complexity of an algorithm and improves the calculation speed. After determining the wavelet basis function, the number of decomposition layers of wavelet noise reduction needs to be determined, if the number of decomposition layers is too small, the elimination of high-frequency noise is not thorough enough, if the number of decomposition layers is too large, the elimination is excessive, some useful signals are also taken as noise to be eliminated, and in an OTDR system, 3 to 5 layers are generally selected.
In the embodiment, the denoising effect, algorithm complexity and calculation speed of a wavelet denoising algorithm in an OTDR system are comprehensively considered, db4 wavelet in Daubechies is selected as a wavelet basis function from three main wavelet basis functions of Daubechies, symlets and Biorthogonal, and the number of signal decomposition layers is 3. And decomposing the measurement signal with noise into multiple layers on different scales to obtain wavelet coefficients of each layer.
Step 3: and obtaining the optimal threshold value of each layer of wavelet coefficient by adopting an improved simulated annealing algorithm.
At this time, the measurement signal with noise is decomposed by multiple layers, and the size of each layer of threshold needs to be determined to process each layer of wavelet coefficient, and in this embodiment, an improved simulated annealing algorithm is adopted to optimize the wavelet coefficient optimal threshold, and the method for optimizing the wavelet coefficient of each layer of threshold is the same, as shown in fig. 3, and specifically includes the following sub-steps:
step 3.1: setting algorithm initial parameters including initial temperature T 0 And termination temperature T k The method comprises the steps of carrying out a first treatment on the surface of the And setting control parameters including the number of cycles N at each temperature step S 。
In the process of setting algorithm parameters, the initial temperature T 0 And termination temperature T k Should be set according to the actual situation, if the initial temperature T 0 Too high or termination temperature T k Too low, the algorithm program wastes unnecessary time, if the initial temperature T 0 Too low or termination temperature T k Too high may cause the algorithm program to mature early and fall into a locally optimal solution. Number of cycles per temperature step N S Typically set to 20 times.
Step 3.2: giving the layerWavelet coefficient threshold value certain initial value rho i (where i represents the number of layers) and calculates a corresponding objective function f (ρ) based on the initial threshold i )。
Step 3.3: setting a proper threshold generating function to generate a new threshold rho by disturbance in a threshold initial value neighborhood range i ' and calculates the new threshold objective function f (ρ) i '), let n=n+1.
Step 3.4: comparing the increment delta f of the objective function before and after the comparison to determine whether to accept the new threshold value, if delta f is less than or equal to 0, accepting the new threshold value rho i ' if not, it is determined whether to accept the new threshold according to the Metropolis criterion. Where Δf=f (ρ i ′)-f(ρ i )。
Step 3.5: if the cycle number N reaches the set cycle number N at the current temperature T S Further judging whether a termination condition is met, if not, entering the step 3.6, and if so, entering the step 3.9; if the cycle number N does not reach the set cycle number N at the current temperature T S The operation of perturbing the threshold neighborhood to generate a new threshold is performed again, i.e. step 3.3 is returned.
Step 3.6: if the current temperature T does not reach the set end temperature T k The rate of change deltac of the objective function is calculated to select the appropriate annealing function. The calculation formula of the change rate delta C of the objective function is as follows:
the annealing function is typically a linear function T' =k 1 T, where the slope k 1 And adjusting according to the magnitude of delta C. I.e. when ΔC is greater than the set value, the slope k is adjusted 1 The value of (2) is larger than 1 so as to realize temperature rise; otherwise, the slope k is adjusted 1 The value of (2) is smaller than 1 so as to realize cooling. In the present embodiment, the slope k 1 Three optional settings are set: 0.80, 0.96 and 1.025, respectively corresponding to rapid cooling, normal cooling and heating, and selecting according to the magnitude of delta C, thereby being capable of more effectively controlling the annealing temperature and enabling the algorithm to have the opportunity of jumping out of the local optimal solution。
Step 3.7: and (3) carrying out annealing operation according to the selected annealing function, and resetting the value of n.
Step 3.8: folding the disturbance range of the threshold, namely, reducing the upper limit and the lower limit of the neighborhood range of the threshold according to a certain proportion, and carrying out the operation of generating a new threshold by disturbance in the neighborhood range of the threshold again, namely, returning to the step 3.3.
The new threshold value can be further determined by adding the step, so that the operation time of the algorithm is shortened, and the calculation speed is increased. Optionally, in this embodiment, the folding ratio of the disturbance range is set to 0.92.
Step 3.9: if the current temperature T reaches the set end temperature T k And outputting the optimal threshold value of the wavelet coefficient of the layer, and ending the algorithm.
Step 4: and carrying out threshold processing on the wavelet coefficients of each layer according to the optimal threshold, using the wavelet coefficients subjected to the threshold processing for reconstructing signals to obtain a denoised OTDR response curve, removing unnecessary noise sections at the tail end, and reserving a measuring part, wherein the obtained OTDR response curve is shown in figure 4.
The traditional threshold judgment function is divided into a hard threshold function and a soft threshold function, but the mode of forced zero setting in the hard threshold function can cause the signal to generate additional oscillation so as to be discontinuous, and the soft threshold function is easy to lose signal details, thereby influencing the authenticity of the reconstructed signal. Based on this, the present embodiment redefines a threshold judgment function between hard and soft threshold functions, and reduces the deviation while ensuring the continuity. Based on the improved threshold judgment function, the updated wavelet coefficient d' of each layer is obtained by comparing the wavelet coefficient d of each layer with the optimal threshold rho of the wavelet coefficient of each layer, and the expression is as follows:
step 5: extracting an intermediate data segment of the denoised OTDR response curve, searching a suspected reflection event and a non-suspected reflection event in the intermediate data segment, removing misjudgment points caused by noise in the suspected event according to the event interval distance and the event duration distance, and finally determining reflection event position information, non-reflection event position information and attenuation loss of each event.
The method for extracting the intermediate data segment of the denoised OTDR response curve comprises the following sub-steps:
step 5.1: performing multi-section least square fitting on a start end data section in the extracted denoised OTDR response curve to locate a start end blind area range of a to-be-measured communication optical fiber measurement section, wherein the method comprises the following steps:
obtaining initial data in the denoised OTDR response curve, setting initial data points of optical fibers as j, setting total data length as L, setting each w data points to perform least square fitting, and assuming that w=50, the fitting data segments corresponding to the data points are (j, j+49), so as to obtain a total of L-49 data segments, wherein the fitting data segments comprise: (1, 50), (2,51), (3, 52) … …. A least squares fit is performed on each data segment, the basic principle of the fitting method being to minimize the variance. Taking the first fit data segment as an example, the data segment contains 50 data { (t) 1 ,s 1 ),(t 2 ,s 2 ),…,(t 50 ,s 50 ) -where t and s represent the measured distance and the relative power strength of the measured signal, respectively, assuming the OTDR measured segment signal as a linear function form s=k 2 t+b), then the least squares fit solves the optimization problem:
the slope k of the data segment to be fitted is needed to locate the dead zone range of the beginning end of the optical fiber 2 The formula (3) is used for the k 2 Deriving to obtain slope k of each fitting data segment 2 The calculation formula of (2) is as follows:
wherein t is j Representing the measured distance, s, of the jth data point in a fitted data segment j Representing a simulationThe relative power intensity of the j-th data point in the data segment is combined.
Determining the slope k of all the fitted data segments according to equation (4) 2 Considering that the initial dead zone comprises a rising edge, the method for positioning the end point of the initial dead zone comprises the following steps of: and if the slope of the last fitted data segment and the slope of the next fitted data segment of the current fitted data segment are smaller than zero, and the difference between the slope of the next fitted data segment and the theoretical fiber attenuation rate (-0.21) is not more than a preset value (such as 0.02), selecting the abscissa of the tail end data point of the current fitted data segment as the end point of the initial end blind area of the to-be-measured communication fiber measurement segment.
Step 5.2: and extracting an end data segment in the denoised OTDR response curve, setting a first threshold according to a mode maximum value corresponding to the end data segment, and considering a first data point abscissa exceeding the first threshold in the end data segment as the end position of the to-be-measured communication optical fiber measurement segment.
Specifically, after the range of the dead zone at the beginning end of the optical fiber measuring section is positioned, the position of the tail end of the optical fiber measuring section is positioned, a db1 wavelet basis function in Daubechies, namely haar wavelet, is adopted to decompose a layer of denoised signal, and a high-frequency coefficient is taken out for direct reconstruction, so that only the information of an event point is reserved, and the normal attenuation which does not need to be concerned is ignored. In the field of signal processing, singular points and abrupt points of a signal represent discontinuities or turns of the signal at the points, and in general, the singular points and abrupt points of the signal carry a great amount of important information of the signal and present local characteristics of the signal, and in a wavelet transformation algorithm, a mode maximum value point corresponds to the singular points and abrupt points of the signal, so that the mode maximum value can be used as a means for detecting signal singularities. The mode maxima point can be described by the following principle: for a value belonging to t 0 If it satisfies Wf (t, s) |less than |wf (t) 0 S) |, and at t 0 Exists in the left and right adjacent domains of (t, s) |wf<|Wf(t 0 S) |, then it is called (t) 0 S) is the maximum point at the scale s, |wf (t) 0 S) is at point (t 0 S) corresponding modulus maxima at the same time, the algorithm code is shown in fig. 5. Analyzing the reconstructed signal by using a mode maximum value method, and transmitting the signalThe number is separated from noise by signal singularities, preserving the modal maxima of the useful signal, as shown in fig. 6. Then intercepting the end 3 km range of the optical fiber measurement section signal (namely, the end data section in the denoised OTDR response curve) as an end space signal, selecting 1/2 of the maximum value of the mode maximum value corresponding to the end space signal as a first threshold, and considering the abscissa of the first data point exceeding the first threshold as the end position of the optical fiber measurement section to be measured.
Step 5.3: after the initial dead zone and the end position of the optical fiber are determined, a data segment between the initial dead zone and the end position is intercepted to be used as an intermediate data segment of the denoised OTDR response curve for positioning event position information and calculating event loss.
The method comprises the following steps of:
step 5.4: and setting a second threshold according to the mode maximum value corresponding to the intermediate data segment, and initially setting the abscissa of the data point exceeding the second threshold in the intermediate data segment as the start position of the suspected reflection event. Optionally, the second threshold set in this embodiment is 1/3 of the maximum value of the modulus maximum value corresponding to the intermediate data segment.
Step 5.5: sequentially judging the relation between the distance between the starting positions of two adjacent suspected reflecting events and the set event interval distance by taking the starting position of the first suspected reflecting event as a starting point, if the relation is larger than the event interval distance, considering the two adjacent suspected reflecting events as different suspected reflecting events, and reserving the starting positions of the two suspected reflecting events; otherwise, the starting position of the first suspected reflection event in the two adjacent suspected reflection events is taken as the starting position of the same suspected reflection event.
Step 5.6: and after the screening, sequentially judging the relation between the duration distance of each suspected reflection event and the set event duration distance by taking the starting position of the first suspected reflection event as a starting point, and if the relation is smaller than the event duration distance, considering the starting position of the suspected reflection event as a misjudgment point and removing the misjudgment point. The algorithm codes of step 5.5 and step 5.6 are shown in fig. 7. Optionally, the event interval distance and the event duration distance set in this embodiment are set to a constant value according to the resolution space, where the event interval distance is 1.2 times the resolution space, and the event duration distance is 0.8 times the resolution space.
Step 5.7: and obtaining the starting position of each reflection event after the secondary screening, determining the ending position of each reflection event according to the starting position of each reflection event, and calculating the attenuation loss of each event according to the relative power intensity corresponding to the starting position and the relative power intensity corresponding to the ending position.
The method comprises the following steps of finding out suspected non-reflection events in an intermediate data segment, removing misjudgment points caused by noise in the suspected events according to event interval distances and event duration distances, and finally determining non-reflection event position information:
step 5.8: the data in the intermediate data segment is subjected to multi-segment least squares fitting to obtain the slopes of all the fitted data segments, as shown in fig. 8, and the implementation steps of the multi-segment least squares fitting are the same as those of step 5.1, and are not described in detail herein. And when the slope of the fitted data segment meets the condition that the slope is smaller than the third threshold and larger than the fourth threshold, the abscissa of the end data of the fitted data segment is initially defined as the start position of the suspected non-reflection event. Wherein the third threshold is set to be the average slope k of all the fitted data segments 0 The fourth threshold is set to a theoretical fiber attenuation in resolution space, 1.15 times.
Similarly, as in the two screening steps for determining the reflection event start position, the non-reflection event start position is determined based on the event interval distance and the event duration distance. Please refer to step 5.5 and step 5.6 for details, and details are not described here.
Step 5.9: and obtaining the starting position of each non-reflection event after the secondary screening, determining the ending position of each non-reflection event according to the starting position of each non-reflection event, and calculating the attenuation loss of each event according to the relative power intensity corresponding to the starting position and the relative power intensity corresponding to the ending position.
In the above steps 5.7 and 5.9, the method for determining the end position of the reflected event or the non-reflected event is the same, and taking an event as an example, the method specifically includes the following steps:
defining the starting position of the event plus the position of the event after a resolution space as the ending position of the initial event. Starting from the end position of the initial event, performing multi-section least square fitting on the data points to obtain the slope of each fitted data section, namely performing least square fitting once on every 50 data points, wherein the fitting method is the same as that of step 5.1, and the description is omitted here. When the slope does not differ from the theoretical fiber decay rate (-0.21) by more than a preset value (e.g., 0.02), then the end data point abscissa of the fitted data segment is selected as the end position of the event.
Optionally, the dynamic range of the OTDR system (i.e. the total attenuation of the optical fiber) is calculated according to the relative power intensity corresponding to the dead zone at the beginning of the optical fiber and the relative power intensity corresponding to the position of the end of the optical fiber.
Optionally, the detection method further includes step 6: and finally, combining the event misjudgment and outputting a document, wherein the dynamic range, the initial dead zone range, the tail end position of the optical fiber, the event positions and the attenuation loss of the OTDR system are automatically recorded in the document, as shown in fig. 9.
After the document is generated, researchers can directly download the document, so that the workload of manual discrimination and recording is reduced; the missing judgment or multiple judgment events generated in the prior art are reduced, namely the event detection method provided by the embodiment is more efficient and intelligent.
What has been described above is only a preferred embodiment of the present application, and the present invention is not limited to the above examples. It is to be understood that other modifications and variations which may be directly derived or contemplated by those skilled in the art without departing from the spirit and concepts of the present invention are deemed to be included within the scope of the present invention.
Claims (10)
1. An OTDR event detection method, the method comprising:
acquiring an OTDR response curve of a communication optical fiber to be tested, and performing wavelet denoising treatment to obtain wavelet coefficients decomposed into layers; wherein the abscissa of the OTDR response curve is the measurement distance, and the ordinate is the relative power intensity;
obtaining an optimal threshold value of each layer of wavelet coefficient by adopting an improved simulated annealing algorithm;
performing threshold processing on the wavelet coefficients of each layer according to the optimal threshold, and using the wavelet coefficients subjected to the threshold processing for reconstructing signals to obtain a denoised OTDR response curve;
extracting an intermediate data segment of the denoised OTDR response curve, searching a suspected reflection event and a non-suspected reflection event in the intermediate data segment, removing misjudgment points caused by noise in the suspected event according to the event interval distance and the event duration distance, and finally determining reflection event position information, non-reflection event position information and attenuation loss of each event;
the event interval distance and the event duration distance are fixed values set according to a resolution space, and the resolution space is a propagation distance corresponding to a pulse width.
2. The OTDR event detection method according to claim 1, wherein the method of obtaining an OTDR response curve of the communication fiber to be measured comprises:
detecting the communication optical fiber to be detected to obtain original received data, wherein the original received data comprises acquisition time and corresponding acquisition voltage;
extracting a data segment with the acquisition time less than zero, and carrying out average processing on the segment data to obtain bias voltage;
extracting a data segment with the acquisition time longer than zero, converting the acquisition time into a measurement distance, converting the acquisition voltage into relative power intensity, and expressing as:
wherein S represents the relative power intensity, V represents the acquisition voltage, V 0 Representing the bias voltage, V 1 Representing the first measurementAnd (3) collecting voltage of the signals.
3. The OTDR event detection method according to claim 1, wherein the method of obtaining the optimal threshold value of each layer of wavelet coefficients using the modified simulated annealing algorithm comprises, for each layer of wavelet coefficients:
giving a certain initial value and a corresponding objective function to the wavelet coefficient threshold of the layer;
setting a proper threshold generating function, generating a new threshold value by disturbance in a threshold initial value neighborhood range, and calculating an objective function of the new threshold value;
comparing the increment of the front and rear objective functions to determine whether to accept the new threshold;
if the circulation times at the current temperature T reach the set circulation times, annealing operation is carried out, otherwise, the operation of generating a new threshold value by disturbance in the threshold value neighborhood range is carried out again;
outputting the optimal threshold value of the wavelet coefficient of the layer when the current temperature T reaches the set termination temperature after annealing for a certain number of times, otherwise, carrying out annealing operation again;
calculating the change rate of the objective function before each annealing operation is performed to select an appropriate annealing function; after each annealing operation, the upper and lower limits of the threshold neighborhood range are reduced according to a certain proportion; wherein the change rate of the objective function is the ratio of the increment of the objective function of the front and back times and the previous objective function.
4. An OTDR event detection method according to claim 3, characterized in that the method of selecting an appropriate annealing function according to the rate of change of the objective function comprises:
defining an annealing function as a linear function: t' =k 1 T, where the slope k 1 Adjusting according to the change rate of the objective function; when the change rate is larger than the set value, the slope k is adjusted 1 The value of (2) is larger than 1 so as to realize temperature rise; otherwise, the slope k is adjusted 1 The value of (2) is smaller than 1 so as to realize cooling.
5. The OTDR event detection method according to claim 1, wherein the method of thresholding the wavelet coefficients of each layer according to an optimal threshold comprises:
based on the improved threshold judgment function, the updated wavelet coefficient d' of each layer is obtained by comparing the wavelet coefficient d of each layer with the optimal threshold rho of the wavelet coefficient of each layer, and the expression is as follows:
6. the OTDR event detection method according to claim 1, wherein the method of extracting the intermediate data segment of the denoised OTDR response curve comprises:
performing multi-section least square fitting on the initial end data segment in the extracted denoised OTDR response curve, and positioning the initial end blind area range of the to-be-measured communication optical fiber measurement segment;
extracting an end data segment in the denoised OTDR response curve, setting a first threshold according to a mode maximum value corresponding to the end data segment, and considering a first data point abscissa exceeding the first threshold in the end data segment as the end position of a to-be-measured communication optical fiber measurement segment;
and intercepting a data segment between the initial dead zone and the tail end position as an intermediate data segment of the denoised OTDR response curve.
7. The method for detecting an OTDR event according to claim 1, wherein the method for finding a suspected reflection event in the intermediate data segment, removing a misjudgment point caused by noise in the suspected event according to the event interval distance and the event duration distance, and finally determining the position information of the reflection event comprises:
setting a second threshold according to a mode maximum value corresponding to the intermediate data segment, and initially determining the abscissa of a data point exceeding the second threshold in the intermediate data segment as a suspected reflection event starting position;
sequentially judging the relation between the distance between the starting positions of two adjacent suspected reflecting events and the set event interval distance by taking the starting position of the first suspected reflecting event as a starting point, and if the relation is larger than the event interval distance, considering the two adjacent suspected reflecting events as different suspected reflecting events; otherwise, taking the beginning position of the first suspected reflection event in the two adjacent suspected reflection events as the beginning position of the same suspected reflection event;
after one-time screening, sequentially judging the relation between the continuous distance of each suspected reflecting event and the set event continuous distance by taking the starting position of the first suspected reflecting event as a starting point, and if the relation is smaller than the event continuous distance, considering the starting position of the suspected reflecting event as a misjudgment point and removing the misjudgment point;
and obtaining the starting position of each reflection event after the secondary screening, and determining the ending position of each reflection event according to the starting position of the event.
8. The method for detecting an OTDR event according to claim 1, wherein the method for finding a suspected non-reflection event in the intermediate data segment, removing a misjudgment point caused by noise in the suspected event according to the event interval distance and the event duration distance, and finally determining the non-reflection event position information comprises:
performing multi-section least square fitting on the data in the middle data section, and when the slope of the fitted data section is smaller than a third threshold and larger than a fourth threshold, initially determining the abscissa of the tail end data of the fitted data section as the start position of the suspected non-reflection event; setting the third threshold according to the average slope of all the fitting data segments, and setting the theoretical fiber attenuation in a resolution space as the fourth threshold;
sequentially judging the relation between the distance between the starting positions of two adjacent suspected non-reflection events and the set event interval distance by taking the starting position of the first suspected non-reflection event as a starting point, and if the relation is larger than the event interval distance, considering the two adjacent suspected non-reflection events as different suspected non-reflection events; otherwise, taking the start position of the suspected non-reflection event which appears first in the two adjacent suspected non-reflection events as the start position of the same suspected non-reflection event;
after one-time screening, sequentially judging the relation between the continuous distance of each suspected non-reflection event and the set event continuous distance by taking the starting position of the first suspected non-reflection event as a starting point, and if the relation is smaller than the event continuous distance, considering the starting position of the suspected non-reflection event as a misjudgment point and removing the misjudgment point;
and obtaining the starting position of each non-reflection event after the secondary screening, and determining the ending position of each non-reflection event according to the starting position of the event.
9. The OTDR event detecting method of claim 6, wherein the method for performing a multi-segment least squares fitting on the extracted starting data segment in the denoised OTDR response curve to locate a starting blind area range of the measured segment of the communication optical fiber to be detected includes:
acquiring initial end data in the denoised OTDR response curve, setting each w data points to perform least square fitting once to obtain the slope k of each fitted data segment 2 The expression is:
wherein t is j Representing the measured distance, s, of the jth data point in a fitted data segment j Representing the relative power intensity of the jth data point in a fitted data segment;
and if the slope of the last fitted data segment and the slope of the next fitted data segment of the current fitted data segment are smaller than zero and the difference between the slope of the next fitted data segment and the theoretical fiber attenuation rate is not greater than a preset value, selecting the abscissa of the tail data point of the current fitted data segment as the end point of the initial end blind area of the measured optical fiber segment of the communication to be measured.
10. An OTDR event detection method as claimed in claim 7 or 8, characterized in that the method of determining the end position of the event comprises, for each event:
defining the starting position of the event plus a position of the resolved space as the ending position of the initial event; starting from the end position of the initial event, performing multi-section least square fitting on the data points to obtain the slope of each fitted data section; and when the difference between the slope and the theoretical fiber attenuation rate is not more than a preset value, selecting the abscissa of the tail end data point of the fitting data segment as the ending position of the event.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310111028.1A CN116155371A (en) | 2023-02-13 | 2023-02-13 | OTDR event detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310111028.1A CN116155371A (en) | 2023-02-13 | 2023-02-13 | OTDR event detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116155371A true CN116155371A (en) | 2023-05-23 |
Family
ID=86355771
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310111028.1A Pending CN116155371A (en) | 2023-02-13 | 2023-02-13 | OTDR event detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116155371A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117118506A (en) * | 2023-10-24 | 2023-11-24 | 北京瑞祺皓迪技术股份有限公司 | OTDR data analysis diagnosis method and device, electronic equipment and storage medium |
| CN117196446A (en) * | 2023-11-06 | 2023-12-08 | 北京中海通科技有限公司 | Product risk real-time monitoring platform based on big data |
-
2023
- 2023-02-13 CN CN202310111028.1A patent/CN116155371A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117118506A (en) * | 2023-10-24 | 2023-11-24 | 北京瑞祺皓迪技术股份有限公司 | OTDR data analysis diagnosis method and device, electronic equipment and storage medium |
| CN117118506B (en) * | 2023-10-24 | 2024-02-02 | 北京瑞祺皓迪技术股份有限公司 | OTDR data analysis diagnosis method and device, electronic equipment and storage medium |
| CN117196446A (en) * | 2023-11-06 | 2023-12-08 | 北京中海通科技有限公司 | Product risk real-time monitoring platform based on big data |
| CN117196446B (en) * | 2023-11-06 | 2024-01-19 | 北京中海通科技有限公司 | Product risk real-time monitoring platform based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116155371A (en) | OTDR event detection method | |
| CN110793616A (en) | All-fiber distributed cable safety and reliability monitoring system | |
| JP6822756B2 (en) | Edge Propagation Optical Time Domain Reflectometry and How to Use It | |
| US10209138B2 (en) | Filtering distributed sensing data | |
| CN116073896B (en) | Method, device, equipment and storage medium for determining test parameters | |
| CN113489534A (en) | Optical cable abnormity detection method and device | |
| KR101677008B1 (en) | Method for Data Forecasting based on Time Series Decomposition | |
| US20110149270A1 (en) | Enhanced OTDR Sensitivity by Utilizing Mode-Field Diameter Measurements | |
| CN110297218B (en) | Method for detecting unknown modulation mode of radar signal based on generation countermeasure network | |
| CN103675791A (en) | Method for recognizing cloud based on mie-scattering laser radar with equalized value distribution | |
| JP2023103994A (en) | Test and measurement system and neural network utilization method | |
| CN114696893B (en) | Event positioning method, system, device and storage medium based on OTDR | |
| CN117147107B (en) | Communication module detection system and detection method | |
| CN111522016B (en) | Method and device for constructing weak signal identification model for space target laser ranging | |
| CN117269333A (en) | OLED device defect prediction method | |
| CN107194377B (en) | Method and device for smoothing COTDR curve and detecting event | |
| CN109444895B (en) | Vibration information positioning method for eliminating interference fading of distributed vibration sensor | |
| CN116388858A (en) | Fiber breakage detection method and device based on phase demodulation | |
| CN114660406A (en) | Method and system for adaptive optimization of cable fault detection waveform | |
| CN114114400A (en) | Microseism event effective signal pickup method | |
| CN114660378A (en) | Multi-source detection parameter-based contact network comprehensive diagnosis method | |
| CN113986700A (en) | Data acquisition frequency optimization method, system, device and storage medium | |
| CN102696042B (en) | For measuring electromagnetic apparatus and method | |
| CN115001571B (en) | Adaptive digital filtering method of optical time domain reflectometer and optical time domain reflectometer | |
| CN116915321B (en) | Rapid test method and system for optical fiber bus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |