# US 11,892,345 B2 · Method and system for detecting and identifying vibration on basis of optical fiber signal feature to determine time-space

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 11,892,345 B2
- Document ID: US11892345B2
- Status: Granted
- Kind: B2
- Application number: 17/418,459
- Filed: 2019-12-25
- Published: 2021-12-23
- Granted: 2024-02-06
- Earliest claimed: 2018-12-29
- Assignees: Neubrex Co., Ltd.
- Applicants: Neubrex Co., Ltd.
- Inventors: Feng Xie; Hongliang Cong; Guidong Li; Xiaohui Hu; Longhai Xu
- Web page: https://patentgrep.com/patent/US11892345B2

## CPC

- G01H 9/004: {using fibre optic sensors (light guides per se G02B6/00, acousto-optical devices specially adapted for gating or modulating in optical wave guides G02F1/125)}
- G06F 18/2135: based on approximation criteria, e.g. principal component analysis
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/24137: {Distances to cluster centroïds}
- G06F 18/24323: {Tree-organised classifiers}
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06F 2218/12: Classification; Matching
- G01H: MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A method for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location of the present invention includes: Step 1 of acquiring a feature-expanded function vector and C-number of vibration categories by expanding a feature of initial data of a vibration signal from a distributed fiber-optic sensor; Step 2 of calculating a dimensionality reduction matrix based on the feature-expanded function vector; Step 3 of acquiring a dimensionality-reduced feature function by operating the dimensionality reduction matrix to the initial data and the feature-expanded function vector; Step 4 of acquiring a primary classification result of the vibration signal by performing a classification with reference to primary classification parameter acquired from a parameter database; and Step 5 of acquiring and outputting a secondary classification result of the vibration signal by performing removal of a wrong detection result and correction of a wrong classification result of the primary classification result.

## Claims

### Claim 1

A method for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location, comprising:
Step 1 of acquiring a feature-expanded function vector and C-number of vibration categories by expanding a feature of initial data of a vibration signal from a distributed fiber-optic sensor;
Step 2 of calculating a dimensionality reduction matrix based on the feature-expanded function vector;
Step 3 of acquiring a dimensionality-reduced feature function by operating the dimensionality reduction matrix to the initial data and the feature-expanded function vector;
Step 4 of acquiring a primary classification result of the vibration signal by performing a classification with reference to a primary classification parameter acquired from a parameter database; and
Step 5 of acquiring and outputting a secondary classification result of the vibration signal by performing removal of a wrong detection result and correction of a wrong classification result of the primary classification result, wherein
Step 2 includes:
Sub-step 21 of collecting a training sample set represented by the following Equation 1 for vibration category c;
Sub-step 22 of calculating, for each vibration category, a center of a cluster represented by the following Equation 2, a variance within the cluster represented by the following Equation 3, a sum of variances within clusters represented by the following Equation 4, a center of all the data represented by the following Equation 5, and a variance between clusters represented by the following Equation 6; and
Sub-step 23 of calculating a feature value of a matrix represented by the following Equation 7, decomposing the matrix, acquiring Q-number of maximum feature values and corresponding feature vectors, and composing the dimensionality reduction matrix by using columns of the feature vectors,
Training
⁢
sample
⁢
set
:
{
g
1
(
c
)
,
g
2
(
c
)
,
…
,
g
K
c
(
c
)
}
,
(
Equation
⁢
1
)
where, in Equation 1, gk(c) denotes a k-th training sample in vibration category c,
Center
⁢
of
⁢
a
⁢
cluster
:
g
¯
(
c
)
=
1
K
c
⁢
∑
k
=
1
K
c
⁢
g
K
(
c
)
,
(
Equation
⁢
2
)
where, in Equation 2, Kc denotes the number of training samples in the category c,
Variance
⁢
within
⁢
the
⁢
cluster
:
(
Equation
⁢
3
)
S
(
c
)
=
∑
k
=
1
K
c
⁢
(
g
k
(
c
)
-
g
¯
(
c
)
)
⁢
(
g
k
(
c
)
-
g
¯
(
c
)
)
T
,
Sum
⁢
of
⁢
variances
⁢
within
⁢
clusters
:
S
∑
=
∑
c
=
1
C
⁢
S
(
c
)
,
(
Equation
⁢
4
)
where, in Equation 4, C denotes the total number of vibration categories,
Center
⁢
of
⁢
all
⁢
of
⁢
the
⁢
data
:
g
¯
=
1
∑
c
=
1
C
⁢
K
c
⁢
∑
c
=
1
C
⁢
∑
k
=
1
K
c
⁢
g
K
(
c
)
,
(
Equation
⁢
5
)
Variance
⁢
between
⁢
clusters
:
S
=
∑
c
=
1
C
⁢
K
c
(
g
_
-
g
_
(
c
)
)
⁢
(
g
_
-
g
_
(
c
)
)
T
,
(
Equation
⁢
6
)
Matrix
:
S
∑
-
1
⁢
S
.
(
Equation
⁢
7
)

### Claim 2 (depends on 1)

The method for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location according to claim 1, wherein
in Step 2, the dimensionality reduction matrix is acquired by Principal Component Analysis or Independent Component Analysis.

### Claim 3

A method for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location comprising:
Step 1 of acquiring a feature-expanded function vector and C-number of vibration categories by expanding a feature of initial data of a vibration signal from a distributed fiber-optic sensor;
Step 2 of calculating a dimensionality reduction matrix based on the feature-expanded function vector;
Step 3 of acquiring a dimensionality-reduced feature function by operating the dimensionality reduction matrix to the initial data and the feature-expanded function vector;
Step 4 of acquiring a primary classification result of the vibration signal by performing a classification with reference to a primary classification parameter acquired from a parameter database; and
Step 5 of acquiring and outputting a secondary classification result of the vibration signal by performing removal of a wrong detection result and correction of a wrong classification result of the primary classification result, wherein
Step 5 includes:
Sub-step 51 of converting the primary classification result represented by the following Equation 8 to a logical variable represented by the following Equation 9;
Sub-step 52 of calculating, for each vibration category, a score of B-number of whole logical equations {J1(L), J2(L), . . . , JB(L)} in a logical rule base on the basis of a calculation equation represented by the following Equation 10; and
Sub-step 53 of outputting a vibration category corresponding to a highest score represented by the following Equation 11 as an output of a secondary classifier,
Primary
⁢
Classification
⁢
Result
:
e
⁡
(
t
,
d
)
,
(
Equation
⁢
8
)
Logical
⁢
Variable
:
L
m
,
n
(
t
)
=
{
1
if
⁢
e
⁡
(
t
-
m
,
d
-
n
)
==
i
0
otherwise
,
(
Equation
⁢
9
)
where, in Equation 9, Lm,n(i) denotes a logical variable when a fiber-optic signal at a time (t−m) and a location (d−n) is determined to be a vibration of category i in Step 4,
Calculation
⁢
Equation
:
s
(
c
)
=
∑
b
=
1
B
⁢
J
b
(
L
)
⁢
s
b
(
c
)
,
(
Equation
⁢
10
)
where, in Equation 10, s(c) denotes a score on the basis of the whole logical equations in the logical rule base when the current vibration belongs to category c, the input L in Jb(L) configures all the logical variables Lm,n(i), the subscript b denotes the b-th logical equation, and sb(c) denotes a contribution degree to a score in each vibration category when the logical equation Jb (L) is established,
Vibration
⁢
category
⁢
corresponding
⁢
to
⁢
a
⁢
highest
⁢
score
:
(
Equation
⁢
11
)
e
*
(
t
,
d
)
=
arg
max
c
s
(
c
)
.

### Claim 4 (depends on 3)

The method for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location according to claim 3, wherein
in Step 2, the dimensionality reduction matrix is acquired by Principal Component Analysis or Independent Component Analysis.

### Claim 5

A system for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location, comprising:
a feature expansion unit for acquiring a feature-expanded function vector and C-number of vibration categories by expanding a feature of initial data of a vibration signal from a distributed fiber-optic sensor;
a dimensionality reduction matrix calculation unit for calculating a dimensionality reduction matrix based on the feature-expanded function vector;
a linear dimensionality reduction unit for acquiring a dimensionality-reduced feature function by operating the dimensionality reduction matrix to the initial data and the feature-expanded function vector;
a primary classification unit for acquiring a primary classification result of the vibration signal by performing a classification with reference to a primary classification parameter acquired from a parameter database; and
a secondary classification unit for acquiring and outputting a secondary classification result of the vibration signal by performing removal of a wrong detection result and correction of a wrong classification result of the primary classification result, wherein
the dimensionality reduction matrix calculation unit
collects a training sample set represented by the following Equation 1 for vibration category c;
calculates, for each vibration category, a center of a cluster represented by the following Equation 2, a variance within the cluster represented by the following Equation 3, a sum of variances within clusters represented by the following Equation 4, a center of all the data represented by the following Equation 5, and a variance between clusters represented by the following Equation 6; and
calculates a feature value of a matrix represented by the following Equation 7, decomposing the matrix, acquiring Q-number of maximum feature values and corresponding feature vectors, and composing the dimensionality reduction matrix by using columns of the feature vectors,
Training
⁢
sample
⁢
set
:
{
g
1
(
c
)
,
g
2
(
c
)
,
…
,
g
K
c
(
c
)
}
,
(
Equation
⁢
1
)
where, in Equation 1, gk(c) denotes a k-th training sample in vibration category c,
Center
⁢
of
⁢
a
⁢
cluster
:
g
¯
(
c
)
=
1
K
c
⁢
∑
k
=
1
K
c
⁢
g
K
(
c
)
,
(
Equation
⁢
2
)
where, in Equation 2, Kc denotes the number of training samples in the category c,
Variance
⁢
within
⁢
the
⁢
cluster
:
(
Equation
⁢
3
)
S
(
c
)
=
∑
k
=
1
K
c
⁢
(
g
k
(
c
)
-
g
¯
(
c
)
)
⁢
(
g
k
(
c
)
-
g
¯
(
c
)
)
T
,
Sum
⁢
of
⁢
variances
⁢
within
⁢
clusters
:
S
∑
=
∑
c
=
1
C
⁢
S
(
c
)
,
(
Equation
⁢
4
)
where, in Equation 4, C denotes the total number of vibration categories,
Center
⁢
of
⁢
all
⁢
of
⁢
the
⁢
data
:
g
¯
=
1
∑
c
=
1
C
⁢
K
c
⁢
∑
c
=
1
C
⁢
∑
k
=
1
K
c
⁢
g
K
(
c
)
,
(
Equation
⁢
5
)
Variance
⁢
between
⁢
clusters
:
S
=
∑
c
=
1
C
⁢
K
c
(
g
_
-
g
_
(
c
)
)
⁢
(
g
_
-
g
_
(
c
)
)
T
,
(
Equation
⁢
6
)
Matrix
:
S
∑
-
1
⁢
S
.
(
Equation
⁢
7
)

### Claim 6 (depends on 5)

The system for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location according to claim 5, wherein
the parameter database is used to accumulate at least one of the dimensionality reduction matrix, a classifier parameter, and a logical rule base.

### Claim 7 (depends on 5)

The system for detecting and specifying a vibration on the basis of a feature of a fiber-optic signal to determine a time and a spatial location according to claim 5, wherein
the secondary classification unit
converts the primary classification result represented by the following Equation 8 to a logical variable represented by the following Equation 9;
calculates, for each vibration category, a score of B-number of whole logical equations {J1(L), J2(L), . . . , JB(L)} in a logical rule base on the basis of a calculation equation represented by the following Equation 10; and
outputs a vibration category corresponding to a highest score represented by the following Equation 11 as an output of a secondary classifier,
Primary
⁢
Classification
⁢
Result
:
e
⁡
(
t
,
d
)
,
(
Equation
⁢
8
)
Logical
⁢
Variable
:
L
m
,
n
(
t
)
=
{
1
if
⁢
e
⁡
(
t
-
m
,
d
-
n
)
==
i
0
otherwise
,
(
Equation
⁢
9
)
where, in Equation 9, Lm,n(i) denotes a logical variable when a fiber-optic signal at a time (t−m) and a location (d−n) is determined to be a vibration of category i in Step 4,
Calculation
⁢
Equation
:
s
(
c
)
=
∑
b
=
1
B
⁢
J
b
(
L
)
⁢
s
b
(
c
)
,
(
Equation
⁢
10
)
where, in Equation 10, s(c) denotes a score on the basis of the whole logical equations in the logical rule base when the current vibration belongs to category c, the input L in Jb(L) configures all the logical variables Lm,n(i), the subscript b denotes the b-th logical equation, and sb(c) denotes a contribution degree to a score in each vibration category when the logical equation Jb (L) is established,
Vibration
⁢
category
⁢
corresponding
⁢
to
⁢
a
⁢
highest
⁢
score
:
(
Equation
⁢
11
)
e
*
(
t
,
d
)
=
arg
max
c
s
(
c
)
.