CN109921847A - The localization method and system of fault branch in a kind of passive optical network - Google Patents
The localization method and system of fault branch in a kind of passive optical network Download PDFInfo
- Publication number
- CN109921847A CN109921847A CN201910299024.4A CN201910299024A CN109921847A CN 109921847 A CN109921847 A CN 109921847A CN 201910299024 A CN201910299024 A CN 201910299024A CN 109921847 A CN109921847 A CN 109921847A
- Authority
- CN
- China
- Prior art keywords
- branch
- pon
- optical
- support vector
- vector machines
- 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.)
- Granted
Links
Landscapes
- Optical Communication System (AREA)
Abstract
The present invention relates to technical field of optical fiber communication, more particularly to the localization method and system of fault branch in a kind of passive optical network, wherein method includes: before runing after PON arranges net, by being trained to obtain best support vector machines to support vector machines, then after PON starts operation, the curved line relation being lost between distance in PON branch is obtained by optical time domain reflectometer, saves as discrete data；Characteristic parameter whether capable of reacting branch trouble is read from discrete data；Characteristic parameter is imported best support vector machines to calculate；Fault branch in PON is positioned according to the calculated result of best support vector machines.The present invention is not under the premise of introducing additional devices or auxiliary device, the best support vector machines of combined training carries out effective position to the fault branch in PON, it solves in PON system, when branch identical especially in the presence of two or more pieces length can not carry out the problem of fault location.
Description
[technical field]
The present invention relates to technical field of optical fiber communication, and in particular to the localization method of fault branch in a kind of passive optical network
And system.
[background technique]
Optical time domain reflectometer (optical time domain reflector, be abbreviated as OTDR) is in optical fiber link failure
The various fields such as the construction maintenance of identification positioning, optical fiber length measuring and optical fiber cable have biggish practical value, nowadays base
In OTDR passive optical network (passive optical network, be abbreviated as PON) monitoring scheme at most widely used
Scheme.However, the PON monitoring network based on OTDR also encounters many challenges although corresponding research is carrying out always.
For PON system especially to access way point-to-multipoint in TDM-PON, the detecting optical pulses that OTDR is issued pass through optical branching
Device enters in each branch, and the Fresnel reflection signal and backward Rayleigh scattering signal of each branch generate mixed at splitter
It is folded, and it is likely to form multipath reflection, so that the test curve of OTDR be made to be difficult to differentiate between, the branch to break down can not be accurately positioned
Road.Especially when the length of branch is identical, aliasing can be more serious, and distinguishing fault branch can be more difficult.Therefore, failure branch
Road optical fiber has been identified as an important problem.
Nowadays, the leading enterprise including optical-fiber network testing field is including NTT, Fujikura and JUSD etc., it is raw
The OTDR for the monitoring of PON system physical layer is produced.Although these products use reflector all on each user Zhi Lu
Part improves OTDR to the monitoring capacity of failure, however but also they lose the ability for being accurately positioned failure.In simple terms,
When the fiber lengths of user's branch are almost the same or difference is smaller, these monitoring systems based on OTDR lack to fault branch
The ability of identification.
In consideration of it, overcoming defect present in the above-mentioned prior art is the art urgent problem to be solved.
[summary of the invention]
The technical problem to be solved in the invention is:
In PON system, the signal of each branch can generate aliasing at splitter or even form multipath reflection, make OTDR
Test curve be difficult to differentiate between, fault branch can not be accurately positioned, especially when the fiber lengths of branch are identical or difference it is smaller
When, aliasing is more serious, and traditional monitoring system distinguishes the ability that fault branch is more difficult, and shortage identifies fault branch.
The present invention reaches above-mentioned purpose by following technical solution:
In a first aspect, the present invention provides a kind of localization methods of fault branch in passive optical network, after PON arranges net
Before operation, obtain best support vector machines by being trained to support vector machines, then it is described after PON starts operation
Method includes:
The curved line relation being lost between distance in PON branch is obtained by optical time domain reflectometer, and saves as discrete data；
From the discrete data got, one or more features parameter whether capable of reacting branch trouble is read；
One or more of characteristic parameters are imported into best support vector machines, and are counted using best support vector machines
It calculates；
According to the calculated result of best support vector machines, fault branch in PON is positioned.
Preferably, described by being trained to obtain best support vector machines to support vector machines, it specifically includes:
It is taken multiple measurements by optical time domain reflectometer, to obtain corresponding loss under various branch trouble combinations in PON
With the curved line relation of distance, and discrete data is saved as respectively；
From the discrete data of acquisition choose one or more features parameter, and combine each branch whether failure, construction instruction
Practice sample set D；
Construct supporting vector machine model；Wherein, comprising multiple in the corresponding mathematic(al) representation of the supporting vector machine model
Undetermined parameter；
It is trained using support vector machines of the sample set D to building, determines each undetermined parameter in supporting vector machine model
Optimal solution, to obtain best support vector machines.
Preferably, there is failure and normal two kinds of situations in each branch in PON, then when in PON include t branch
When, the branch trouble combination sum in PON is 2t；
Then in the training process, one-shot measurement, measurement are carried out for every kind of branch trouble combination using optical time domain reflectometer
Total degree mutually should be 2t；Wherein, t >=2.
Preferably, the spy chosen in the characteristic parameter and training process read during the fault location after PON operation
It is consistent to levy parameter retention properties；Wherein, the characteristic parameter includes the reflection peak in the curved line relation between loss and distance
Position, the peak value of reflection peak, the full width at half maximum of reflection peak and damage curve G-bar in one or more.
Preferably, the sample set D specifically:
Wherein,InIndicate the characteristic parameter that i-th branch has j-th；yi=1 or -1, yi=1 table
Show that i-th branch does not break down, yi=-1 indicates that i-th branch breaks down；1≤i≤n, 1≤j≤m, n indicate PON
In branch number, m indicate characteristic parameter number.
Preferably, the mathematic(al) representation of the supporting vector machine model are as follows:
Wherein, r indicates " distance "；X is the column vector being made of each characteristic parameter,ω is normal vector, ω=(ω1；ω2；ω3；...；ωmn)；B is position
Transposition；| | | | indicate norm.
Preferably, in the determining supporting vector machine model each undetermined parameter optimal solution specifically: constantly correction described in
Normal vector, the value of displacement item and the order of norm in supporting vector machine model, so that each element in sample set D passes through support
The overall effect for " distance " r that vector machine obtains after calculating is best.
Preferably, the calculated result of the best support vector machines of the basis, positions fault branch in PON, specifically
Are as follows:
Y is corresponded to according to each branch that best support vector machines is calculatediValue, confirms whether each branch breaks down, into
And complete the positioning of fault branch；Wherein, if yi=1, then confirm that corresponding i-th branch is without failure；If yi=-
1, then confirm that corresponding i-th branch breaks down.
Second aspect, the present invention also provides a kind of positioning systems of fault branch in passive optical network, for realizing upper
The localization method of fault branch in passive optical network described in first aspect is stated, then the positioning system includes optical time domain reflection
Instrument and sequentially connected optical line terminal, circulator, optical splitter and at least two optical network units, the optical time domain reflection
Instrument is connect with the circulator；
Wherein, the optical line terminal is for accessing data, and converts data to after optical signal through the circulator
It is input to the optical splitter；The optical splitter is used to the optical signal received being divided at least two branches, and is separately input into
At least two optical network unit；The reflected light that the optical time domain reflectometer is used to receive in each branch is handled, and is obtained
Curved line relation between loss and distance carries out fault location will pass through the calculating of support vector machines.
Preferably, the circulator includes first port, second port and third port, the first port and the light
Line terminal connection, the second port are connect with the optical splitter, and the third port is connect with the optical time domain reflectometer,
So that first port described in the optical signals of the optical line terminal output enters the circulator, and defeated from the second port
Out to the optical splitter；Reflected light enters the circulator by the second port, and exports from the third port to described
Optical time domain reflectometer.
Compared with prior art, the beneficial effects of the present invention are:
In the localization method and system of fault branch provided by the invention, before being runed first after PON arranges net, pass through
" training " obtains best support vector machines, real time data, extracting parameter is then obtained from the PON to start operation, and using most
Good support vector machines is calculated, and then adjudicates fault branch.The present invention is not before introducing additional devices or auxiliary device
It puts, effective position is carried out to the fault branch in PON, solves two or more pieces length in PON system, especially Star Network
The problem of fault location can not be carried out in identical branch.
[Detailed description of the invention]
In order to illustrate the technical solution of the embodiments of the present invention more clearly, will make below to required in the embodiment of the present invention
Attached drawing is briefly described.It should be evident that drawings described below is only some embodiments of the present invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is the localization method flow chart of fault branch in a kind of passive optical network provided in an embodiment of the present invention；
Fig. 2 is the complete logical block diagram of fault branch positioning in a kind of passive optical network provided in an embodiment of the present invention；
Fig. 3 is the training method process of best support vector machines in a kind of passive optical network provided in an embodiment of the present invention
Figure；
Fig. 4 is the curved line relation between a kind of loss obtained by OTDR measurement provided in an embodiment of the present invention and distance
Figure；
Fig. 5 is the structural representation of the positioning system of fault branch in a kind of passive optical network provided in an embodiment of the present invention
Figure.
[specific embodiment]
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
In the description of the present invention, term "inner", "outside", " longitudinal direction ", " transverse direction ", "upper", "lower", "top", "bottom",
The orientation or positional relationship of the instructions such as "left", "right", "front", "rear" is to be based on the orientation or positional relationship shown in the drawings, and is only
For ease of description the present invention rather than require the present invention that must be constructed and operated in a specific orientation, therefore be not construed as
Limitation of the present invention.
In addition, as long as technical characteristic involved in the various embodiments of the present invention described below is each other not
Constituting conflict can be combined with each other.Just with reference to drawings and examples, in conjunction with coming, the present invention will be described in detail below.
Embodiment 1:
The embodiment of the invention provides a kind of localization methods of fault branch in passive optical network, can be to the failure in PON
Branch carries out effective position, solves the problems, such as that the identical branch of two or more pieces length can not carry out fault location in PON system.
With reference to Fig. 1, after PON completion arranges net and starts operation, the localization method of fault branch can be briefly summarized as four portions in PON
Point: it obtains data, parameter extraction, positioned using the calculating of best support vector machines and fault branch, is i.e. step 201- step 204.
Wherein, in order to determine the best support vector machines, after PON arranges net, before actual operation, it is right in advance also to need
Support vector machines is trained, to obtain best support vector machines.Therefore, complete localization method can be divided into two big steps, first
It is the support vector machines training before being runed after PON arranges net, followed by the fault location after PON actual operation, as shown in Figure 2.
Therefore, before introducing physical fault position fixing process, the training process of support vector machines is described first, as shown in figure 3,
Specifically includes the following steps:
Step 101, it is taken multiple measurements by optical time domain reflectometer, combines lower pair to obtain various branch troubles in PON
The curved line relation of the loss and distance answered, and discrete data is saved as respectively.
Before PON actual operation, the measurement request of progress is comprehensive, thus for branches all in PON break down can
The curved line relation that will be measured, and then obtain loss and distance under all possible combinations can be combined.Due to each of PON
There is failure and normal two kinds of situations in branch, then when in PON including t branch, the branch trouble combination sum in PON is
2t；Then in the training process, it is combined using optical time domain reflectometer for every kind of branch trouble and carries out one-shot measurement, total time of measurement
Number mutually should be 2t；Wherein, t >=2.For example, share 4 branches in certain PON, every branch is faulty and not two kinds of situations of failure,
Whole to have 16 kinds of situations altogether, each case all carries out one-shot measurement with OTDR, measures 16 times altogether, obtains 16 curved line relations.
Step 102, one or more features parameter is chosen from the discrete data of acquisition, and whether former in conjunction with each branch
Barrier constructs training sample set D.
In this step, the characteristic parameter should be derived from the data in curve measured by OTDR, and can be anti-
Answer relationship whether measured data and branch trouble.In the present embodiment, the characteristic parameter can be specifically selected as: OTDR
Measure in the curved line relation between obtained loss and distance, the position of reflection peak, the peak value of reflection peak, reflection peak it is half high
One or more in the G-bar of overall with and damage curve；Wherein, after OTDR measurement obtains curved line relation, each feature
The selection of parameter can refer to Fig. 4.
The branch number in PON is indicated with n, and m indicates the number of characteristic parameter, then the sample set D specifically:
In formula,InIndicate the characteristic parameter that i-th branch has j-th；In the present embodiment, yi=1 or -1, yi
=1 i-th branch of expression does not break down, yi=-1 indicates that i-th branch breaks down；1≤i≤n, 1≤j≤m.Herein
With 1 and -1 value come whether indicating branch trouble, certainly, in the alternative, specific value can flexible choice, do not do herein
It limits.
Step 103, supporting vector machine model is constructed；Wherein, in the corresponding mathematic(al) representation of the supporting vector machine model
Include multiple undetermined parameters.In the present embodiment, mathematic(al) representation are as follows:
Wherein, r indicates " distance "；X is the column vector being made of each characteristic parameter,ω is normal vector, ω=(ω1；ω2；ω3；...；ωmn)；B is position
Transposition；| | | | indicate certain norm.
Step 104, it is trained, is determined each in supporting vector machine model using support vector machines of the sample set D to building
The optimal solution of undetermined parameter, to obtain best support vector machines.
In this step, the result of training process should be that all elements in sample set D are calculated by support vector machines
The overall effect of " distance " r obtained later is optimal.Then each undetermined parameter is most in the determining supporting vector machine model
Excellent solution specifically refers to: it constantly corrects the normal vector in the supporting vector machine model, be displaced the value of item and the order of norm, so that
The overall effect of " distance " that each element in sample set D obtains after being calculated by support vector machines is best, is desirably to obtain most
Short " distance ", to obtain best support vector machines.
Wherein, the calculated result in Fig. 2 is each y being calculated using supporting vector machine modeliValue, actual result are
For each y determined according to practical branch trouble combined situationiValue, trained result should be that expectation calculated result can be with practical knot
The fruit goodness of fit with higher.Then during tractable, after correcting parameter every time, just once counted using support vector machines
It calculates；If the goodness of fit of calculated result and actual result is higher, best support vector machines can be obtained；If calculated result and reality
As a result the goodness of fit is lower, then continuation corrected parameter is needed to be calculated, until the goodness of fit is met the requirements.Herein, the goodness of fit compared with
Height refers to that the goodness of fit reaches preset value, and the preset value can be configured according to practical application request, and general setting is higher, such as
It may be configured as 98%, then after the goodness of fit between calculated result and actual result is more than 98%, it is believed that the whole effect of " distance "
Fruit is best, obtains best support vector machines.
After 101- step 104 obtains best support vector machines through the above steps, it can complete to arrange net in PON and start to transport
After battalion, the fault branch in PON is positioned by step 201- step 204, with specific reference to Fig. 1, comprising:
Step 201, the curved line relation being lost between distance in PON branch is obtained by optical time domain reflectometer, and saved as
Discrete data.
Positioning system structure and Fig. 5 in 2 in conjunction with the embodiments, reflected light during PON actual operation, in each branch
Through being transmitted to optical time domain reflectometer OTDR after circulator, then the reflected light in the optical time domain reflectometer OTDR receiving branch is simultaneously
It is handled, obtains the curved line relation of branch loss and distance as shown in Figure 4.
Step 202, from the discrete data got, one or more features whether capable of reacting branch trouble are read
Parameter.
It should be noted that the characteristic parameter being read in the step should be kept with the characteristic parameter chosen in step 102
Property is consistent, if first chosen in step 102 characteristic parameter is the position of reflection peak, is read in step 202
The value of first characteristic parameter should also be the position of reflection peak, the i.e. spy read during fault location after PON operation
The characteristic parameter retention properties chosen in sign parameter and training process is consistent.
Step 203, one or more of characteristic parameters are imported into best support vector machines, and utilizes best supporting vector
Machine is calculated.That is, in the best support vector machines that the characteristic ginseng value steps for importing 104 read in step 202 is obtained into
Row calculates.
Step 204, according to the calculated result of best support vector machines, fault branch in PON is positioned.
It can be calculated the corresponding y of each branch in PON by best support vector machinesiValue, then it is corresponding according to each branch
YiValue, can confirm whether each branch breaks down；If yi=1, then it can confirm that corresponding i-th branch does not break down；
If yi=-1 can then confirm that corresponding i-th branch breaks down, and then the positioning of achievable fault branch.
In conclusion in the localization method of above-mentioned fault branch provided in an embodiment of the present invention, first after PON arranges net
Before operation, best support vector machines is obtained by " training ", real time data is then obtained from the PON to start operation, is extracted
Parameter, and calculated using best support vector machines, and then adjudicate fault branch；Combination supporting vector machine can be in PON
Fault branch carries out effective position, solves in PON system, there are the identical branch of two or more pieces length especially in Star Network
The problem of fault location can not be carried out when road.
Embodiment 2:
On the basis of above-described embodiment 1, the embodiment of the invention also provides fault branches in a kind of passive optical network
Positioning system can be used for realizing the localization method of fault branch described in embodiment 1, and it is long to solve two or more pieces in PON system
The problem of fault location can not be carried out by spending identical branch.
As shown in figure 5, the positioning system of fault branch includes optical time domain in passive optical network provided in an embodiment of the present invention
Reflectometer OTDR and sequentially connected optical line terminal (optical line terminal, be abbreviated as OLT), circulator,
Optical splitter and at least two optical network units (optical networkunit, be abbreviated as ONU), the optical time domain reflectometer with
The circulator connection.Specifically, whole system can be divided into local side, Optical Distribution Network (optical distribution
Network is abbreviated as ODN) and three parts of user terminal, local side connected with user terminal by Optical Distribution Network ODN, local side includes
Optical time domain reflectometer OTDR, optical line terminal OLT and circulator, Optical Distribution Network ODN include optical splitter, and user terminal includes at least
Two optical network unit ONUs.
Wherein, the optical line terminal OLT first line of a couplet upper layer network, and then can be used for accessing data, and convert data to
After the form of optical signal, the optical splitter in the Optical Distribution Network ODN is input to by the circulator；
The Optical Distribution Network ODN can receive the optical signal that the optical line terminal OLT transmits, then by user point
It is divided at least two branches with corresponding optical signal, and then by the optical signal received, and is separately input into described at least two
Optical network unit ONU；In the present embodiment, for four optical network unit ONUs are set, it is denoted as ONU1, ONU2, ONU3 respectively
And ONU4, correspondingly there are four branches, are denoted as branch 1, branch 2, branch 3 and branch 4 respectively；
The optical network unit ONU can receive the optical signal of the ODN Optical Distribution Network distribution, and to received optical signal
It carries out the processing such as demodulating, to provide voice, data and multimedia service for user；
The optical signal that the optical line terminal OLT transmits can be input in the Optical Distribution Network ODN by the circulator
Optical splitter, and the reflected light information in branch is input in the optical time domain reflectometer OTDR；
The reflected light that the optical time domain reflectometer OTDR can receive in each branch is handled, and is obtained between loss and distance
Curved line relation, and then discrete data can be obtained, extract characteristic parameter, and be calculated by support vector machines, to realize
Fault location.
Wherein, the unidirectional transmission property of circulator is utilized in the embodiment of the present invention, by the way that circulator is arranged, light point can be made
Distribution network ODN gets input optical signal, while optical time domain reflectometer can also get the reflection signal in branch, meet optical path
Transmission demand.The working principle of the circulator is specific as follows: the circulator is a kind of branch's biography with characteristics of non-reciprocity
Defeated system, signal can only be exported along one direction annular delivery and from one of port, and opposite direction is isolation, that is, works as signal
It is signal output end with the port clockwise adjacent or next port adjacent counterclockwise when being inputted by any port
Mouthful, and remaining a port is isolated port, not output signal；Specifically determine that signal is unidirectional along clockwise direction by bias field
Annular delivery or the transmission of counter clockwise direction unidirectional loop.
With reference to Fig. 5, the circulator in the embodiment of the present invention is the transmission of clockwise direction unidirectional loop, is specifically included clockwise
First port (i.e. port 1), second port (i.e. port 2) and the third port (i.e. port 3) that direction is set gradually, described first
Port is connect with the optical line terminal, and the second port is connect with the optical splitter, the third port and it is described smooth when
The connection of domain reflectometer.Therefore, the principle transmitted according to unidirectional loop, described in the optical signals of optical line terminal OLT output
It after first port enters the circulator, can be exported from the second port to the optical splitter, the third port is at this time
Isolated port；Reflected light by the second port enter the circulator after, can be exported from the third port to it is described smooth when
Domain reflectometer, the first port is isolated port at this time.
PON structure as shown in connection with fig. 5, it is assumed that a total of four branches in PON, and the length phase of branch 1 and branch 2
Together, the embodiment of the present invention further provides a kind of specific embodiment of fault branch positioning, comprising the following steps:
Step 1 disconnects the junction of branch 1 and OUN1, is lost with OTDR measurement and the curved line relation of distance,
Save as discrete data；The junction of branch 2 and OUN1 are disconnected, are lost with OTDR measurement and the curved line relation of distance, is deposited
For discrete data；The junction of branch 3 and OUN1 are disconnected, are lost with OTDR measurement and the curved line relation of distance, is saved as
Discrete data；The junction of branch 4 and OUN1 are disconnected, are lost with OTDR measurement and the curved line relation of distance, save as from
Dissipate data；The junction of branch 1 and OUN1 are disconnected, the junction of branch 2 and OUN2 are disconnected, are lost with OTDR measurement
With the curved line relation of distance, discrete data is saved as.This rule is followed, all 16 kinds of branch troubles combined situations have all been measured
At obtaining final discrete data.
Step 2, the selected characteristic parameter from the discrete data of step 1 can choose four characteristic parameters herein: And further combined with each branch whether failure, construct training sample set D.
Step 3 utilizes step 1 and the obtained data of step 2 and characteristic parameter after constructing supporting vector machine model
Support vector machines is trained, i.e., is trained using sample set D.Training process can support vector regression algorithm, trained
Best support vector machines is obtained after.
Step 4 simulates the case where completing and starting operation of arranging net: the junction of branch 1 and ONU1 being disconnected, Jin Ermo
The normal situation of quasi- 1 failure of branch, remaining branch.By OTDR experiment curv data, four characteristic parameters are therefrom read: reflection
The position at peak, the peak value of reflection peak, the full width at half maximum of reflection peak and the slope of curve of loss and distance Curve relationship.By four spies
Sign parameter, which is input in best support vector machines, to be calculated, it is possible to find the result calculated is y1=-1, y2=1, y3=1, y4
=1, i.e. branch 2,3,4 is normal, 1 failure of branch, so as to complete the identical situation of branch 1,2 length, to guilty culprit branch
The positioning on road.
In conclusion above-mentioned fault branch positioning system provided in an embodiment of the present invention, can not introduce additional devices
Or under the premise of auxiliary device, combination supporting vector machine carries out effective position to the fault branch in PON, solves PON system
In, the problem of fault location can not be carried out especially in Star Network when branch identical there are two or more pieces length.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. the localization method of fault branch in a kind of passive optical network, which is characterized in that before being runed after PON arranges net, lead to
It crosses and support vector machines is trained to obtain best support vector machines, then after PON starts operation, which comprises
The curved line relation being lost between distance in PON branch is obtained by optical time domain reflectometer, and saves as discrete data；
From the discrete data got, one or more features parameter whether capable of reacting branch trouble is read；
One or more of characteristic parameters are imported into best support vector machines, and are calculated using best support vector machines；
According to the calculated result of best support vector machines, fault branch in PON is positioned.
2. the localization method of fault branch in passive optical network according to claim 1, which is characterized in that it is described by pair
Support vector machines is trained to obtain best support vector machines, specifically includes:
Taken multiple measurements by optical time domain reflectometer, with obtain in PON under the combination of various branch troubles corresponding loss with away from
From curved line relation, and save as discrete data respectively；
From the discrete data of acquisition choose one or more features parameter, and combine each branch whether failure, construct training sample
This collection D；
Construct supporting vector machine model；Wherein, comprising multiple undetermined in the corresponding mathematic(al) representation of the supporting vector machine model
Parameter；
It is trained using support vector machines of the sample set D to building, determines that each undetermined parameter is most in supporting vector machine model
Excellent solution, to obtain best support vector machines.
3. the localization method of fault branch in passive optical network according to claim 2, which is characterized in that every in PON
There is failure and normal two kinds of situations in a branch, then when in PON including t branch, the branch trouble in PON combines sum
It is 2t；
Then in the training process, one-shot measurement is carried out for the combination of every kind of branch trouble using optical time domain reflectometer, measurement it is total
Number mutually should be 2t；Wherein, t >=2.
4. the localization method of fault branch in passive optical network according to claim 2, which is characterized in that PON runs it
The characteristic parameter retention properties chosen in the characteristic parameter and training process read during fault location afterwards is consistent, the spy
Sign parameter include the position of reflection peak in the curved line relation between loss and distance, the peak value of reflection peak, reflection peak it is half high
One or more in the G-bar of overall with and damage curve.
5. the localization method of fault branch in passive optical network according to claim 2, which is characterized in that the sample set
D specifically:
Wherein,InIndicate the characteristic parameter that i-th branch has j-th；yi=1 or -1, yi=1 indicates i-th
Branch does not break down, yi=-1 indicates that i-th branch breaks down；1≤i≤n, 1≤j≤m, n indicate the branch in PON
Road number, m indicate the number of characteristic parameter.
6. the localization method of fault branch in passive optical network according to claim 5, which is characterized in that it is described support to
The mathematic(al) representation of amount machine model are as follows:
Wherein, r indicates " distance "；X is the column vector being made of each characteristic parameter,ω is normal vector, ω=(ω1；ω2；ω3；...；ωmn)；B is position
Transposition；| | | | indicate norm.
7. the localization method of fault branch in passive optical network according to claim 6, which is characterized in that the determining branch
Hold the optimal solution of each undetermined parameter in vector machine model specifically: constantly correct normal vector in the supporting vector machine model,
Be displaced the value of item and the order of norm so that each element in sample set D calculated by support vector machines obtain later " away from
From " overall effect of r is best.
8. the localization method of fault branch in passive optical network according to claim 5, which is characterized in that the basis is most
The calculated result of good support vector machines positions fault branch in PON, specifically:
Y is corresponded to according to each branch that best support vector machines is calculatediValue, confirms whether each branch breaks down, and then completes
The positioning of fault branch；Wherein, if yi=1, then confirm that corresponding i-th branch is without failure；If yi=-1, then really
Recognize corresponding i-th branch to break down.
9. the positioning system of fault branch in a kind of passive optical network, which is characterized in that including optical time domain reflectometer, and sequentially
Optical line terminal, circulator, optical splitter and at least two optical network units of connection, the optical time domain reflectometer and the annular
Device connection；
Wherein, the optical line terminal is for accessing data, and is inputted after converting data to optical signal by the circulator
To the optical splitter；The optical splitter is used to the optical signal received being divided at least two branches, and is separately input into described
At least two optical network units；The reflected light that the optical time domain reflectometer is used to receive in each branch is handled, and loss is obtained
With the curved line relation between distance, fault location is carried out will pass through the calculating of support vector machines.
10. the positioning system of fault branch in passive optical network according to claim 9, which is characterized in that the annular
Device includes first port, second port and third port, and the first port is connect with the optical line terminal, the second end
Mouth is connect with the optical splitter, and the third port is connect with the optical time domain reflectometer, so that the optical line terminal exports
Optical signals described in first port enter the circulator, and export from the second port to the optical splitter；Each branch
Reflected light the circulator is entered by the second port, and export from the third port to the optical time domain reflectometer.
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 true CN109921847A (en) | 2019-06-21 |
| CN109921847B 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) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110492926A (en) * | 2019-07-26 | 2019-11-22 | 北京光锁科技有限公司 | A kind of optical fiber eavesdropping localization method, system and electronic equipment based on machine learning |
| CN110530551A (en) * | 2019-08-27 | 2019-12-03 | 西南交通大学 | The temperature extraction method of BOTDA based on Support Vector Machines Optimized |
| 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) |
| CN114448774A (en) * | 2021-12-16 | 2022-05-06 | 武汉光迅科技股份有限公司 | Alarm processing method, device and storage medium |
Citations (7)
| 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 |
| CN101567724A (en) * | 2009-05-23 | 2009-10-28 | 华为技术有限公司 | Network situation detection system and method |
| 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 |
-
2019
- 2019-04-15 CN CN201910299024.4A patent/CN109921847B/en active Active
Patent Citations (7)
| 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 |
| CN101567724A (en) * | 2009-05-23 | 2009-10-28 | 华为技术有限公司 | Network situation detection system and method |
| 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 |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110492926A (en) * | 2019-07-26 | 2019-11-22 | 北京光锁科技有限公司 | A kind of optical fiber eavesdropping localization method, system and electronic equipment based on machine learning |
| CN110530551A (en) * | 2019-08-27 | 2019-12-03 | 西南交通大学 | The temperature extraction method of BOTDA based on Support Vector Machines Optimized |
| 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) |
| CN114448774A (en) * | 2021-12-16 | 2022-05-06 | 武汉光迅科技股份有限公司 | Alarm processing method, device and storage medium |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109921847B (en) | 2020-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109921847A (en) | The localization method and system of fault branch in a kind of passive optical network | |
| CN101291176B (en) | Fault detection method, system and apparatus for optical distributed 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 | |
| KR101657329B1 (en) | Method and apparatus for fault discovery in a passive optical network(pon) | |
| CN101924590B (en) | The detection system of fiber fault of passive optical network and method | |
| US9240855B1 (en) | Fiber diagnosis system for point-to-point optical access networks | |
| US8948589B2 (en) | Apparatus and method for testing fibers in a PON | |
| CN110996193B (en) | Method, related device and system for identifying optical network unit connection port | |
| CN101924962B (en) | System and method thereof for detecting fiber faults | |
| CN102356571B (en) | Fault detection method and apparatus of optical distribution network and opticla network system | |
| CN101232328B (en) | Method for locating case point of branch optical fiber, optical network and network appliance | |
| CN103718478B (en) | Certification for EPON and the system for problem in the fiber arm of end and the detection of fault | |
| CN106160846B (en) | Method and apparatus for detecting PON network line fault | |
| CN103548287A (en) | Supervision of wavelength division multiplexed optical networks | |
| CN104655591B (en) | A kind of optical cable generaI investigation device and method of detection beating position | |
| CN102412902A (en) | Optical network unit photoelectric device with optical time domain reflection function | |
| CN102725975B (en) | The fault detection method of a kind of optical distribution network, device and optical network system | |
| CN110266374A (en) | It is a kind of can high precision monitor TDM-PON second level branch trouble device and method | |
| CN104935378B (en) | A kind of fiber link detecting system and method for EPON | |
| CN102104421B (en) | Branched optical fiber failure detection method and device for optical network, and optical network | |
| CN105337656B (en) | A kind of connection relation of optical network determines method, equipment and system | |
| CN101411098A (en) | Signal identifying apparatus for an optical fiber | |
| KR20110061254A (en) | Detecting apparatus and method for optical line in passive optical network system | |
| CN107070544A (en) | Optical assembly, detecting devices and decision method using the optical assembly | |
| CN108833002B (en) | Optical link detection method and system based on passive optical network |
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 |