CN109921847B - Method and system for positioning fault branch in passive optical network - Google Patents
Method and system for positioning fault branch in passive optical network Download PDFInfo
- Publication number
- CN109921847B CN109921847B CN201910299024.4A CN201910299024A CN109921847B CN 109921847 B CN109921847 B CN 109921847B CN 201910299024 A CN201910299024 A CN 201910299024A CN 109921847 B CN109921847 B CN 109921847B
- Authority
- CN
- China
- Prior art keywords
- branch
- support vector
- vector machine
- fault
- pon
- 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.)
- Active
Links
- 230000003287 optical effect Effects 0.000 title claims abstract description 116
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000012706 support-vector machine Methods 0.000 claims abstract description 81
- 238000012549 training Methods 0.000 claims abstract description 29
- 238000004364 calculation method Methods 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims description 18
- 238000005259 measurement Methods 0.000 claims description 14
- 239000013598 vector Substances 0.000 claims description 10
- 238000006073 displacement reaction Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 2
- 238000004891 communication Methods 0.000 abstract description 2
- 239000013307 optical fiber Substances 0.000 abstract description 2
- 238000000253 optical time-domain reflectometry Methods 0.000 description 27
- 230000005540 biological transmission Effects 0.000 description 5
- 239000000835 fiber Substances 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 description 1
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Landscapes
- Optical Communication System (AREA)
Abstract
The invention relates to the technical field of optical fiber communication, in particular to a method and a system for positioning a fault branch in a passive optical network, wherein the method comprises the following steps: before the PON is operated after the network is distributed, training a support vector machine to obtain an optimal support vector machine, and after the PON starts to operate, acquiring a curve relation between loss and distance in a PON branch circuit through an optical time domain reflectometer and storing the curve relation as discrete data; reading characteristic parameters capable of reflecting whether the branch circuit is in fault or not from the discrete data; importing the characteristic parameters into an optimal support vector machine for calculation; and positioning the fault branch in the PON according to the calculation result of the optimal support vector machine. The fault branch in the PON is effectively positioned by combining the trained optimal support vector machine on the premise of not introducing additional devices or auxiliary devices, and the problem that the fault positioning cannot be carried out in a PON system, particularly when two or more branches with the same length exist, is solved.
Description
[ technical field ] A method for producing a semiconductor device
The invention relates to the technical field of optical fiber communication, in particular to a method and a system for positioning a fault branch in a passive optical network.
[ background of the invention ]
An optical time domain reflector (OTDR, abbreviated as OTDR) has a great practical value in many fields such as fiber link fault identification and location, fiber length measurement, and construction and maintenance of fiber optic cables, and a Passive Optical Network (PON) monitoring scheme based on OTDR becomes the most widely applied scheme. However, despite the ongoing research, OTDR-based PON monitoring networks also encounter many challenges. For a PON system, particularly for a point-to-multipoint access mode in a TDM-PON, detection optical pulses sent by the OTDR enter each branch through an optical splitter, Fresnel reflection signals and backward Rayleigh scattering signals of each branch are subjected to aliasing at the splitter, and multiple reflections can be formed, so that test curves of the OTDR are difficult to distinguish, and the branch with a fault cannot be accurately positioned. Aliasing is more severe and it is more difficult to distinguish a faulty branch, especially when the length of the branch is the same. Thus, identification of a faulty branch fiber becomes an important issue.
Today, OTDRs for physical layer monitoring of PON systems are produced in the field of optical network testing, including NTT, Fujikura, and jurd. Although these products employ reflective devices on each subscriber branch to improve OTDR fault monitoring, they also lose the ability to accurately locate the fault. Briefly, these OTDR based monitoring systems lack the ability to identify a faulty branch when the fiber lengths of the subscriber branches are substantially uniform or differ by a small amount.
In view of the above, it is an urgent problem in the art to overcome the above-mentioned drawbacks of the prior art.
[ summary of the invention ]
The technical problems to be solved by the invention are as follows:
in a PON system, signals of each branch may generate aliasing or even form multiple reflections at a splitter, so that a test curve of an OTDR is difficult to distinguish, and a faulty branch cannot be accurately located.
The invention achieves the above purpose by the following technical scheme:
in a first aspect, the present invention provides a method for locating a faulty branch in a passive optical network, where before operation of a PON after deployment, an optimal support vector machine is obtained by training a support vector machine, and after the PON starts to operate, the method includes:
acquiring a curve relation between loss and distance in a PON branch circuit through an optical time domain reflectometer, and storing the curve relation as discrete data;
reading one or more characteristic parameters capable of reflecting whether the branch circuit is in fault or not from the acquired discrete data;
importing the one or more characteristic parameters into an optimal support vector machine, and calculating by using the optimal support vector machine;
and positioning the fault branch in the PON according to the calculation result of the optimal support vector machine.
Preferably, the obtaining of the optimal support vector machine by training the support vector machine specifically includes:
measuring for multiple times through an optical time domain reflectometer to obtain curve relations between corresponding loss and distance under various branch fault combinations in the PON, and storing the curve relations as discrete data;
selecting one or more characteristic parameters from the acquired discrete data, and constructing a training sample set D by combining whether each branch circuit is in fault or not;
constructing a support vector machine model; the mathematical expression corresponding to the support vector machine model comprises a plurality of undetermined parameters;
and training the constructed support vector machine by using the sample set D, and determining the optimal solution of each undetermined parameter in the support vector machine model so as to obtain the optimal support vector machine.
Preferably, each branch in the PON has both a fault condition and a normal condition, and when the PON includes t branches, the total number of the fault combinations of the branches in the PON is 2t；
In the training process, an optical time domain reflectometer is adopted to measure each branch fault combination once, and the total number of times of measurement is 2 correspondinglyt(ii) a Wherein t is more than or equal to 2.
Preferably, the characteristic parameters read in the fault positioning process after the PON operation are consistent with the characteristic parameters selected in the training process; wherein the characteristic parameters include one or more of a position of a reflection peak, a peak value of the reflection peak, a full width at half maximum of the reflection peak, and an average slope of a loss curve in a curve relationship between loss and distance.
Preferably, the sample set D specifically includes:
wherein,
in (1)
The characteristic parameter of the jth branch is represented; y isi1 or-1, y i1 means that the ith branch is not faulty, y i1 indicates that the ith branch has a fault; i is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to m, n represents the number of branches in the PON, and m represents the number of characteristic parameters.
Preferably, the mathematical expression of the support vector machine model is as follows:
wherein r represents a "distance"; x is a column vector consisting of characteristic parameters,
ω is a normal vector, ω ═ ω (ω ═ ω)1；ω2；ω3；...；ωmn) (ii) a b is a displacement term; | | · | | represents a norm.
Preferably, the determining an optimal solution of each undetermined parameter in the support vector machine model specifically includes: and continuously correcting the normal vector, the value of the displacement item and the order of the norm in the support vector machine model, so that the overall effect of the distance r obtained after each element in the sample set D is calculated by the support vector machine is optimal.
Preferably, the locating a faulty branch in the PON according to the calculation result of the optimal support vector machine specifically includes:
obtaining the corresponding y of each branch circuit by calculation according to the optimal support vector machineiDetermining whether each branch circuit has a fault, and further completing the positioning of the fault branch circuit; wherein if yiIf the number of branches is 1, confirming that the corresponding ith branch has no fault; if y isiAnd confirming that the corresponding ith branch has a fault if the branch is equal to-1.
In a second aspect, the present invention further provides a system for locating a faulty branch in a passive optical network, which is used to implement the method for locating a faulty branch in a passive optical network described in the first aspect, where the locating system includes an optical time domain reflectometer, and an optical line terminal, a circulator, an optical splitter, and at least two optical network units that are connected in sequence, where the optical time domain reflectometer is connected to the circulator;
the optical line terminal is used for accessing data, converting the data into optical signals and inputting the optical signals to the optical splitter through the circulator; the optical splitter is used for splitting the received optical signal into at least two branches, and inputting the at least two branches to the at least two optical network units respectively; the optical time domain reflectometer is used for receiving and processing the reflected light in each branch, and acquiring the curve relation between the loss and the distance so as to carry out fault positioning through calculation of the support vector machine.
Preferably, the circulator includes a first port, a second port and a third port, the first port is connected to the optical line terminal, the second port is connected to the optical splitter, and the third port is connected to the optical time domain reflectometer, so that an optical signal output by the optical line terminal enters the circulator through the first port and is output to the optical splitter through the second port; reflected light enters the circulator from the second port and is output to the optical time domain reflectometer from the third port.
Compared with the prior art, the invention has the beneficial effects that:
according to the fault branch locating method and system provided by the invention, firstly, before operation after PON distribution, an optimal support vector machine is obtained through training, then real-time data is obtained from the PON which starts to operate, parameters are extracted, and the optimal support vector machine is used for calculation, so that a fault branch is judged. The invention effectively positions the fault branch in the PON on the premise of not introducing additional devices or auxiliary devices, and solves the problem that the fault positioning cannot be carried out in the PON system, particularly two or more branches with the same length in a star network.
[ description of the drawings ]
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below. It is obvious that the drawings described below are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 is a flowchart of a method for locating a faulty branch in a passive optical network according to an embodiment of the present invention;
fig. 2 is a complete logic block diagram of a fault branch location in a passive optical network according to an embodiment of the present invention;
fig. 3 is a flowchart of a training method of an optimal support vector machine in a passive optical network according to an embodiment of the present invention;
FIG. 4 is a graph of loss versus distance measured by an OTDR according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a system for locating a faulty branch in a passive optical network according to an embodiment of the present invention.
[ detailed description ] embodiments
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In the description of the present invention, the terms "inside", "outside", "longitudinal", "lateral", "upper", "lower", "top", "bottom", "left", "right", "front", "rear", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention but do not require that the present invention must be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other. The invention will be described in detail below with reference to the figures and examples.
Example 1:
the embodiment of the invention provides a method for positioning a fault branch in a passive optical network, which can effectively position the fault branch in the PON and solve the problem that two or more branches with the same length in the PON system cannot be positioned. Referring to fig. 1, after the PON finishes networking and starts operating, a method for locating a faulty branch in the PON can be briefly summarized into four parts: acquiring data, extracting parameters, calculating by using an optimal support vector machine and positioning a fault branch, namely step 201-step 204.
In order to determine the optimal support vector machine, the support vector machine needs to be trained in advance after the PON is deployed and before the PON is actually operated, so as to obtain the optimal support vector machine. Therefore, the complete positioning method can be divided into two major steps, i.e. firstly training the support vector machine after the PON is deployed and before operation, and then positioning the fault after the PON is actually operated, as shown in fig. 2. Therefore, before introducing the actual fault location process, a description is first given to a training process of the support vector machine, as shown in fig. 3, which specifically includes the following steps:
step 101, performing multiple measurements by using an optical time domain reflectometer to obtain curve relations between corresponding losses and distances under various branch fault combinations in the PON, and storing the curve relations as discrete data.
Before the actual operation of the PON, the measurement requirements are comprehensive, so that the possible combinations of all branches in the PON with faults are measured, and further, the curve relationship between the loss and the distance under all the possible combinations is obtained. Because each branch in the PON has two situations, namely failure and normal, when the PON includes t branches, the total number of the branch failure combinations in the PON is 2t(ii) a In the training process, an optical time domain reflectometer is adopted to measure each branch fault combination once, and the total number of times of measurement is 2 correspondinglyt(ii) a Wherein t is more than or equal to 2. For example, a PON has 4 branches, each branch has two cases of failure and no failure, and the whole PON has 16 cases, and each case is measured once by the OTDR and 16 times in total, so as to obtain 16 curve relationships.
And 102, selecting one or more characteristic parameters from the acquired discrete data, and constructing a training sample set D by combining whether each branch circuit is in fault or not.
In this step, the characteristic parameter is derived from data in a curve measured by the OTDR and is capable of reflecting the relationship between the measured data and whether the branch is faulty or not. In this embodiment, the characteristic parameters may be specifically selected from: in the curve relation between the loss and the distance obtained by OTDR measurement, one or more items of the position of a reflection peak, the peak value of the reflection peak, the full width at half maximum of the reflection peak and the average slope of the loss curve; after obtaining the curve relationship through OTDR measurement, the selection of each characteristic parameter may refer to fig. 4.
N represents the number of branches in the PON, and m represents the number of characteristic parameters, so that the sample set D specifically includes:
103, constructing a support vector machine model; and the mathematical expression corresponding to the support vector machine model comprises a plurality of undetermined parameters. In this embodiment, the mathematical expression is:
wherein r represents a "distance"; x is a column vector consisting of characteristic parameters,
ω is a normal vector, ω ═ ω (ω ═ ω)1；ω2；ω3；...；ωmn) (ii) a b is a displacement term; i | · | | represents a certain norm.
And 104, training the constructed support vector machine by using the sample set D, and determining the optimal solution of each undetermined parameter in the support vector machine model so as to obtain the optimal support vector machine.
In this step, the result of the training process should be that the overall effect of the "distance" r obtained after calculation by the support vector machine for all elements in the sample set D is optimal. The determining the optimal solution of each undetermined parameter in the support vector machine model specifically includes: and continuously correcting the values of normal vectors and displacement terms and the orders of norms in the support vector machine model, so that the overall effect of the distance obtained after each element in the sample set D is calculated by the support vector machine is optimal, and the shortest distance is obtained, thereby obtaining the optimal support vector machine.
Wherein, the calculation result in fig. 2 is each y calculated by using the support vector machine modeliThe value and the actual result are all y determined according to the actual branch fault combination conditioniThe result of the training should be that the expected calculated result can have a high degree of agreement with the actual result. In the process of domestication, after correcting the parameters each time, performing calculation once by using a support vector machine; if the coincidence degree of the calculation result and the actual result is higher, an optimal support vector machine can be obtained; if the goodness of fit of the calculation result and the actual result is low, parameters need to be corrected continuously for calculation until the goodness of fit meets the requirements. Here, the high goodness of fit means that the goodness of fit reaches a preset value, the preset value can be set according to actual application requirements, and is generally set to be high, for example, 98%, and then when the goodness of fit between a calculation result and an actual result exceeds 98%, the overall effect of "distance" is considered to be the best, and the best support vector machine is obtained.
After the optimal support vector machine is obtained through the steps 101 to 104, that is, after the PON finishes network deployment and starts operation, the fault branch in the PON is located through the steps 201 to 204, which specifically refers to fig. 1, and includes:
step 201, obtaining a curve relation between loss and distance in the PON branch by using an optical time domain reflectometer, and storing the curve relation as discrete data.
With reference to the positioning system structure in embodiment 2 and fig. 5, in an actual operation process of the PON, reflected light in each branch is transmitted to the optical time domain reflectometer OTDR after passing through the circulator, and the optical time domain reflectometer OTDR receives and processes the reflected light in the branch to obtain a curve relationship between the branch loss and the distance as shown in fig. 4.
It should be noted that the characteristic parameters read in this step should be consistent with the characteristic parameters selected in step 102, and if the first characteristic parameter selected in step 102 is the position of the reflection peak, the value of the first characteristic parameter read in step 202 should also be the position of the reflection peak, that is, the characteristic parameters read in the fault location process after the PON operation should be consistent with the characteristic parameters selected in the training process.
And step 204, positioning the fault branch in the PON according to the calculation result of the optimal support vector machine.
Y corresponding to each branch in the PON can be obtained through calculation of the optimal support vector machineiValue, then according to y corresponding to each branchiThe value can confirm whether each branch circuit has a fault; if y isiIf the number of branches is 1, the corresponding ith branch is determined to be not failed; if y isiAnd (4) determining that the ith branch fails, and further completing the positioning of the failed branch.
In summary, in the method for locating a faulty branch according to the embodiments of the present invention, before operation after PON deployment, an optimal support vector machine is obtained through "training", then real-time data is obtained from a PON that starts to operate, parameters are extracted, and the optimal support vector machine is used for performing calculation, so as to determine a faulty branch; the fault branch in the PON can be effectively positioned by combining a support vector machine, and the problem that the fault positioning cannot be carried out in a PON system, particularly when two or more branches with the same length exist in a star network is solved.
Example 2:
on the basis of the foregoing embodiment 1, an embodiment of the present invention further provides a system for locating a faulty branch in a passive optical network, which can be used to implement the method for locating a faulty branch described in embodiment 1, and solve a problem that two or more branches with the same length in a PON system cannot be located.
As shown in fig. 5, a system for locating a faulty branch in a passive optical network according to an embodiment of the present invention includes an optical time domain reflectometer OTDR, and an Optical Line Terminal (OLT), a circulator, an optical splitter, and at least two Optical Network Units (ONUs) that are connected in sequence, where the optical time domain reflectometer is connected to the circulator. Specifically, the whole system can be divided into three parts, namely a local side, an Optical Distribution Network (ODN), and a user side, wherein the local side and the user side are connected through the ODN, the local side includes an OTDR, an OLT and a circulator, the ODN includes an optical splitter, and the user side includes at least two ONUs.
The optical line terminal OLT is connected with an upper network and then can be used for accessing data, converting the data into an optical signal form and inputting the optical signal form to an optical splitter in the optical distribution network ODN through the circulator;
the optical distribution network ODN can receive optical signals transmitted by the optical line terminal OLT, then distributes corresponding optical signals according to users, further divides the received optical signals into at least two branches, and inputs the at least two branches to the at least two optical network units ONU respectively; in this embodiment, taking the example of setting four optical network units ONU, which are respectively denoted as ONU1, ONU2, ONU3, and ONU4, there are four branches, which are respectively denoted as branch 1, branch 2, branch 3, and branch 4;
the optical network unit ONU can receive the optical signal distributed by the ODN optical distribution network and demodulate and process the received optical signal, thereby providing voice, data and multimedia services for users;
the circulator can input an optical signal transmitted by the optical line terminal OLT into an optical splitter in the optical distribution network ODN and input reflected light information in a branch into the optical time domain reflectometer OTDR;
the OTDR can receive and process the reflected light in each branch, obtain the curve relation between loss and distance, further obtain discrete data, extract characteristic parameters, and calculate through a support vector machine, thereby realizing fault positioning.
In the embodiment of the invention, the unidirectional transmission characteristic of the circulator is utilized, the optical distribution network ODN can acquire the input optical signal by arranging the circulator, and meanwhile, the optical time domain reflectometer can also acquire the reflected signal in the branch, so that the transmission requirement of an optical path is met. The working principle of the circulator is as follows: the circulator is a branch transmission system with nonreciprocal characteristic, signals can only be transmitted in a single-direction ring and output from one port, and the opposite direction is isolated, namely when the signals are input from any one port, the next port which is adjacent to the port clockwise or anticlockwise is a signal output port, and the rest one port is an isolated port and does not output the signals; specifically, the bias magnetic field determines whether the signal is transmitted in a clockwise unidirectional ring or a counterclockwise unidirectional ring.
Referring to fig. 5, the circulator in the embodiment of the present invention is configured to perform unidirectional ring transmission in a clockwise direction, and specifically includes a first port (i.e., port 1), a second port (i.e., port 2), and a third port (i.e., port 3) that are sequentially arranged in the clockwise direction, where the first port is connected to the optical line terminal, the second port is connected to the optical splitter, and the third port is connected to the optical time domain reflectometer. Therefore, according to the principle of unidirectional ring transmission, after an optical signal output by the optical line terminal OLT enters the circulator from the first port, the optical signal may be output to the optical splitter from the second port, where the third port is an isolated port; and after the reflected light enters the circulator from the second port, the reflected light can be output to the optical time domain reflectometer from the third port, and at the moment, the first port is an isolation port.
In combination with the PON structure shown in fig. 5, assuming that there are four branches in the PON and that the length of branch 1 is the same as that of branch 2, the embodiment of the present invention further provides a specific implementation manner of fault branch location, including the following steps:
disconnecting the connection part of the branch 1 and the OUN1, obtaining a curve relation between loss and distance by OTDR measurement, and storing the curve relation as discrete data; disconnecting the connection part of the branch 2 and the OUN1, obtaining the curve relation of loss and distance by OTDR measurement, and storing the curve relation as discrete data; disconnecting the connection part of the branch 3 and the OUN1, obtaining the curve relation of loss and distance by OTDR measurement, and storing the curve relation as discrete data; disconnecting the connection part of the branch 4 and the OUN1, obtaining the curve relation of loss and distance by OTDR measurement, and storing the curve relation as discrete data; and disconnecting the connection part of the branch 1 and the OUN1, disconnecting the connection part of the branch 2 and the OUN2, and obtaining the curve relation of loss and distance by OTDR measurement, wherein the curve relation is stored as discrete data. And (4) following the rule, completely measuring all 16 branch fault combination conditions to obtain final discrete data.
Step two, selecting characteristic parameters from the discrete data in the step one, wherein four characteristic parameters can be selected:
and further combining whether each branch circuit is in fault or not, and constructing a training sample set D.
And step three, after a support vector machine model is constructed, training the support vector machine by using the data and the characteristic parameters obtained in the step one and the step two, namely training by using a sample set D. The training process can support a vector regression algorithm, and the optimal support vector machine is obtained after the training is finished.
Step four, simulating the situation that the network distribution is completed and the operation is started: the connection between the branch 1 and the ONU1 is disconnected, so that the conditions that the branch 1 is in failure and the rest branches are normal are simulated. Through OTDR measurement curve data, four characteristic parameters are read from the data: the position of the reflection peak, the peak value of the reflection peak, the full width at half maximum of the reflection peak and the slope of the loss versus distance curve. Inputting the four characteristic parameters into an optimal support vector machine for calculation, and finding out that the calculated result is y1＝-1,y2＝1,y3＝1,y4When the branch 1 fails, the length of the branches 1 and 2 is the same, and the branch where the fault is located.
In summary, the fault branch positioning system provided in the embodiment of the present invention can effectively position a fault branch in a PON by combining a support vector machine without introducing an additional device or an auxiliary device, so as to solve a problem that fault positioning cannot be performed in the PON system, especially when two or more branches having the same length exist in a star network.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. A method for positioning a fault branch in a passive optical network is characterized in that before PON is arranged and operated, an optimal support vector machine is obtained by training a support vector machine, and after the PON starts to operate, the method comprises the following steps:
acquiring a curve relation between loss and distance in a PON branch circuit through an optical time domain reflectometer, and storing the curve relation as discrete data;
reading one or more characteristic parameters capable of reflecting whether the branch circuit is in fault or not from the acquired discrete data;
importing the one or more characteristic parameters into an optimal support vector machine, and calculating by using the optimal support vector machine;
according to the calculation result of the optimal support vector machine, positioning a fault branch in the PON;
wherein the characteristic parameters include one or more of a position of a reflection peak, a peak value of the reflection peak, a full width at half maximum of the reflection peak, and an average slope of a loss curve in a curve relationship between loss and distance.
2. The method according to claim 1, wherein the training of the support vector machine to obtain the optimal support vector machine specifically comprises:
measuring for multiple times through an optical time domain reflectometer to obtain curve relations between corresponding loss and distance under various branch fault combinations in the PON, and storing the curve relations as discrete data;
selecting one or more characteristic parameters from the acquired discrete data, and constructing a training sample set D by combining whether each branch circuit is in fault or not;
constructing a support vector machine model; the mathematical expression corresponding to the support vector machine model comprises a plurality of undetermined parameters;
and training the constructed support vector machine by using the sample set D, and determining the optimal solution of each undetermined parameter in the support vector machine model so as to obtain the optimal support vector machine.
3. The method according to claim 2, wherein each branch in the PON has both fault and normal conditions, and when t branches are included in the PON, the total number of branch fault combinations in the PON is 2t；
In the training process, an optical time domain reflectometer is adopted to measure each branch fault combination once, and the total number of times of measurement is 2 correspondinglyt(ii) a Wherein t is more than or equal to 2.
4. The method according to claim 2, wherein the characteristic parameters read in the fault location process after PON operation are consistent with the characteristic parameters selected in the training process.
5. A method for locating a faulty branch in a passive optical network as claimed in claim 2, wherein the sample set D is specifically:
wherein,
in (1)
The characteristic parameter of the jth branch is represented; y isi1 or-1, yi1 means that the ith branch is not faulty, yi1 indicates that the ith branch has a fault; i is more than or equal to 1 and less than or equal to n, j is more than or equal to 1 and less than or equal to m, n represents the number of branches in the PON, and m represents the number of characteristic parameters.
6. The method according to claim 5, wherein the mathematical expression of the support vector machine model is as follows:
wherein r represents a "distance"; x is a column vector consisting of characteristic parameters,
ω is a normal vector, ω ═ ω (ω ═ ω)1；ω2；ω3；...；ωmn) (ii) a b is a displacement term; and | | represents a norm.
7. The method according to claim 6, wherein the determining the optimal solution of each to-be-determined parameter in the support vector machine model specifically comprises: and continuously correcting the normal vector, the value of the displacement item and the order of the norm in the support vector machine model, so that the overall effect of the distance r obtained after each element in the sample set D is calculated by the support vector machine is optimal.
8. The method according to claim 5, wherein the method for locating the faulty branch in the PON according to the calculation result of the optimal support vector machine specifically comprises:
obtaining the corresponding y of each branch circuit by calculation according to the optimal support vector machineiDetermining whether each branch circuit has a fault, and further completing the positioning of the fault branch circuit; wherein if yiIf the number of branches is 1, confirming that the corresponding ith branch has no fault; if y isiAnd confirming that the corresponding ith branch has a fault if the branch is equal to-1.
9. A system for locating a faulty branch in a passive optical network, for implementing the method for locating a faulty branch in a passive optical network according to any one of claims 1 to 8, comprising an optical time domain reflectometer, and an optical line terminal, a circulator, an optical splitter and at least two optical network units connected in sequence, wherein the optical time domain reflectometer is connected to the circulator;
the optical line terminal is used for accessing data, converting the data into optical signals and inputting the optical signals to the optical splitter through the circulator; the optical splitter is used for splitting the received optical signal into at least two branches, and inputting the at least two branches to the at least two optical network units respectively; the optical time domain reflectometer is used for receiving and processing the reflected light in each branch, and acquiring the curve relation between the loss and the distance so as to carry out fault positioning through calculation of the support vector machine.
10. A system for locating a faulty branch in a passive optical network as claimed in claim 9, wherein the circulator includes a first port, a second port and a third port, the first port is connected to the optical line terminal, the second port is connected to the optical splitter, and the third port is connected to the optical time domain reflectometer, so that an optical signal output from the optical line terminal enters the circulator through the first port and is output from the second port to the optical splitter; the reflected light of each branch enters the circulator from the second port and is output to the optical time domain reflectometer from the third port.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910299024.4A CN109921847B (en) | 2019-04-15 | 2019-04-15 | Method and system for positioning fault branch in passive optical network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910299024.4A CN109921847B (en) | 2019-04-15 | 2019-04-15 | Method and system for positioning fault branch in passive optical network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109921847A CN109921847A (en) | 2019-06-21 |
| CN109921847B true CN109921847B (en) | 2020-09-22 |
Family
ID=66969838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910299024.4A Active CN109921847B (en) | 2019-04-15 | 2019-04-15 | Method and system for positioning fault branch in passive optical network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109921847B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110492926B (en) * | 2019-07-26 | 2021-01-05 | 北京光锁科技有限公司 | Optical fiber eavesdropping positioning method and system based on machine learning and electronic equipment |
| CN110530551B (en) * | 2019-08-27 | 2021-07-23 | 西南交通大学 | BOTDA temperature extraction method based on optimized support vector machine |
| CN111130634A (en) * | 2019-12-13 | 2020-05-08 | 国网西藏电力有限公司 | Method and system for identifying loss event in OPGW (optical fiber composite overhead ground wire) |
| CN114448774B (en) * | 2021-12-16 | 2023-12-05 | 武汉光迅科技股份有限公司 | Alarm processing method, device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102170309A (en) * | 2011-03-24 | 2011-08-31 | 索尔思光电（成都）有限公司 | Optical line terminal (OLT) optical module integrating optical time domain reflectometer (OTDR) monitoring function |
| CN103166699A (en) * | 2011-12-16 | 2013-06-19 | 中国电信股份有限公司 | Method and system for positioning fault of optical fiber behind optical splitter in passive optical network (PON) |
| EP2670067A1 (en) * | 2012-05-30 | 2013-12-04 | Broadcom Corporation | Passive optical fiber plant analysis |
| CN108155934A (en) * | 2017-12-22 | 2018-06-12 | 武汉光迅科技股份有限公司 | A kind of OLT ends optical transceiver module and method with time domain reflection function |
| CN108229553A (en) * | 2017-12-29 | 2018-06-29 | 国网吉林省电力有限公司信息通信公司 | A kind of OTDR curve datas analysis method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0723002A (en) * | 1993-06-22 | 1995-01-24 | Nippon Telegr & Teleph Corp <Ntt> | Identifying method for branch optical fiber |
| CN101567724B (en) * | 2009-05-23 | 2012-08-22 | 华为技术有限公司 | Network situation detection system and method |
-
2019
- 2019-04-15 CN CN201910299024.4A patent/CN109921847B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102170309A (en) * | 2011-03-24 | 2011-08-31 | 索尔思光电（成都）有限公司 | Optical line terminal (OLT) optical module integrating optical time domain reflectometer (OTDR) monitoring function |
| CN103166699A (en) * | 2011-12-16 | 2013-06-19 | 中国电信股份有限公司 | Method and system for positioning fault of optical fiber behind optical splitter in passive optical network (PON) |
| EP2670067A1 (en) * | 2012-05-30 | 2013-12-04 | Broadcom Corporation | Passive optical fiber plant analysis |
| CN108155934A (en) * | 2017-12-22 | 2018-06-12 | 武汉光迅科技股份有限公司 | A kind of OLT ends optical transceiver module and method with time domain reflection function |
| CN108229553A (en) * | 2017-12-29 | 2018-06-29 | 国网吉林省电力有限公司信息通信公司 | A kind of OTDR curve datas analysis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109921847A (en) | 2019-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109921847B (en) | Method and system for positioning fault branch in passive optical network | |
| CN102187603B (en) | Method and apparatus for deriving parameters of optical paths in optical networks using a two-wavelength otdr and a wavelength-dependent reflective element | |
| CN110996193B (en) | Method, related device and system for identifying optical network unit connection port | |
| US9225607B2 (en) | Network system and topology map generating method | |
| US7894362B2 (en) | Passive optical network topology estimation | |
| EP3029853B1 (en) | Method, device, and system for optical fiber link identification | |
| US8805184B2 (en) | Technology for fault allocation in passive optical networks (PON) | |
| CN102118658B (en) | Method and device for port identification of light splitter | |
| WO2013014531A2 (en) | Techniques for detecting optical faults in passive optical networks | |
| WO2013002692A1 (en) | Otdr trace analysis in pon systems | |
| CN111669221B (en) | Method, device and system for fault location | |
| CN101232328A (en) | Method for locating case point of branch optical fiber, optical network and network appliance | |
| CN104202084A (en) | Device and method for monitoring failures of time division multiplexing optical network link | |
| CN110266374A (en) | It is a kind of can high precision monitor TDM-PON second level branch trouble device and method | |
| WO2014002741A1 (en) | Optical path monitoring method and optical path monitoring system | |
| CN108833002B (en) | Optical link detection method and system based on passive optical network | |
| CN102684785B (en) | Based on optical network fault checkout gear and the detection method thereof of noise signal | |
| KR20110061254A (en) | Detecting apparatus and method for optical line in passive optical network system | |
| CN110176957A (en) | A kind of device and method of high-precision, Larger Dynamic range monitoring WDM-PON failure | |
| RU2628767C2 (en) | Method and device to configure testing parameters set by using the optic time domain reflectometer (otdr) | |
| CN110289905B (en) | Device and method for accurately monitoring TWDM-PON fault by using FP laser | |
| CN106415195B (en) | Optical fibre measuring method, device and system | |
| CN115967445B (en) | Method, equipment and system for generating optical network topology | |
| Laferriere et al. | Original method for analyzing multipaths networks by OTDR measurement | |
| CN114430512A (en) | Port identification method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |