JP2584478B2 - Method for processing received signal of optical fiber backscattered light - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of processing a received signal in measuring backscattered light intensity or backscattered light reflectance in the longitudinal direction of an optical fiber.

[Prior art] An optical pulse is incident on an optical fiber, and backscattered light from the optical fiber is detected.
Optical to detect external factors such as temperature and humidity
Time Domain Reflectometry (OTDR) technique is used. Various scattering and loss are originally generated in the optical fiber, and in the OTDR method, if the time change of the backscattered light with respect to the incident pulse light returning to the incident end is known, the amount of light at a certain time becomes a certain position of the optical fiber. Utilizes the ability to respond to scattered light at

When backscattered light is generated simply in proportion to the light pulse reaching a certain position and returns to the incident end, the ratio of backscattered light between two points (between two times) , The loss of the optical fiber in that section can be known. If the optical fiber is broken or distorted, backscattered light corresponding to the loss of the optical fiber is observed until the time corresponding to that position, and thereafter, the broken or distorted portion is observed. Since only weak backscattered light containing a large loss (in some cases, too weak to be detected) is detected, a large step occurs. The time until this step occurs corresponds to the centering or distortion position. Note that an extreme change in light intensity due to Fresnel reflection at the cross section is also observed at the core-cut portion.

On the other hand, a change in the temperature of the optical fiber or the like becomes an external factor that changes the scattered light, and the backscattered light intensity at that portion changes. Changes. Although the loss due to external factors such as temperature change of the optical fiber is weak, if the correlation between the external factor and the backscattered light is grasped, the magnitude and position of the external factor on the optical fiber can be known, and the distribution type Can be used as a sensor.

As described above, the OTDR method is a convenient method for obtaining various types of information along the length of an optical fiber, and is used in many fields.In particular, as the field of application as a distributed sensor expands, the position In addition, it has been demanded to improve the detection accuracy of the amount of change. For this reason, many ideas have already been devised.

Regarding the position accuracy, by making the time width of the incident pulse light as small as possible, the value of the measured backscattered light at a certain time corresponding to the change at a certain position as much as possible is advanced. Have been. If the time width of the pulsed light is large, the value at a certain time is the backscattered light due to the head of the pulsed light reaching a certain position and the backscattered light due to the continuation of the pulsed light reaching the position before that. Are overlapped, and both the position and the amount of change include an error.

On the other hand, when the time width of the incident pulse light is reduced, the amount of backscattered light at a certain position is reduced, so that the detection accuracy of the external change is reduced. In order to prevent this, it has been practiced to increase the light intensity of the incident pulse light or to increase the sensitivity of the light receiving device.

Further, the backscattered light with respect to the single pulse light includes random noise in the light emission, light reception, and amplification circuits, which lowers the accuracy. For this reason, the influence of random noise has been reduced by employing an averaging process of response signals to a large number of incident pulses. In this case, the digital processing technology of the signal is utilized, and the backscattered light that returns thereafter is sampled every certain time with the incident pulse light or the like as a time reference, A / D converted, and the time series of the signal is obtained. Making and averaging.

[Problems to be Solved by the Invention] By the above-mentioned measures for improving accuracy, the accuracy of detection and measurement by OTDR has been improved as compared with before. However, the actual accuracy (accuracy of the position and the amount of change) is still higher than the apparent accuracy (resolution) in relation to the time width or waveform of the incident pulse light and the sampling time width or the range of external change. The problem remains that it does not improve. An example is shown in FIG. In the figure, t is a time variable, x is a position transformation, p is a light intensity variable, and s is a magnitude of an external change.

It is assumed that the incident pulse light has a waveform as shown in FIG. It is also assumed that the external change has a size S in a range of a range m to m + 1 of a distance 1 from the position m as shown in FIG. 3B. If the scattering loss inherent in the optical fiber other than the external change is excluded and considered, the waveform (time change) of the obtained backscattered light is as shown in FIG. 3 (c). Here, v is the speed of light in the optical fiber. When this is sampled with a sampling time width Δt and time → position conversion, an external change that can be detected from the backscattered light is shown in FIG.
Becomes The values P 'and S of the variation are calibrated in advance.

Comparing FIGS. 3 (b) and 3 (d), the position information has an error of ± τ · v / 2 at maximum, and the amount of external change has an error of ± S at maximum. Assuming now that τ / Δt = 10,
Assuming that the accuracy of p or s at the time of sampling is S / 10, the nominal error is Δt (Δt · v / 2 in terms of distance) and S / 10. Become.

The object of the present invention is to solve the problems of the prior art described above,
It is an object of the present invention to provide a method of processing a received signal of optical fiber backscattered light, which can reduce a measurement error of the intensity or reflectance of backscattered light at each point in an optical fiber by the OTDR technique.

Means for Solving the Problems In order to achieve the above object, the present invention obtains the time change of the backscattered light with respect to the pulse incident light incident on the optical fiber, performs sampling at Δt time intervals, In the processing of the received signal of the optical fiber backscattered light, which obtains the time series data by repeating the averaging process repeatedly and obtains the backscattered light intensity in the length direction of the optical fiber by converting the time change into the position change. , The distance of the measuring end is 0,
When the speed of light in the optical fiber is v, Δx = Δt · v / 2,
Distance x ₀ ~x ₀ + N · Δx backscattered light reflectance r (x ₀₎ of the optical fiber of the position of [Delta] x the distance between _{the, r (x 0 + Δx)} , r (x 0 +
2 · Δx),..., R (x ₀ + N · Δx)
(0), R (1), R (2),..., R (N), and the time t _{0 to} t ₀ + M when the pulse rising time of the transmission pulse light is 0
• Transmitted pulse light intensity p at the time of Δt interval between Δt
₀ (t), p ₀ (t ₀ + Δt), p ₀ (t ₀ + 2 · Δt),..., P
₀ (t ₀ + M · Δt) is P ₀ (0), P ₀ (1), P
₀ (2),..., P ₀ (M), and the time when the pulse light is transmitted is 0, t ₀ ′ = x ₀ / (v / 2), and the time t _{0 to} t ₀ ′ +
Apparent backscattered light intensity p (t ₀ ′), p (t ₀ ′ + Δ) measured at the measuring end at the time of (M + N) · Δt Δt interval
t), p (t ₀ ′ + 2Δt),..., p (t ₀ ′ + (M + N) · Δ
t) are P (0), P (1), P (2),..., P (M +
N), and the incident light pulse arrival ratio from the incident end to the distance x ₀ + n · Δx is measured from T ₁ (n) and the distance x ₀ + n · Δx of the backscattered light generated at the position of the distance x ₀ + n · Δx. T ₂ (n), T (n) = T ₁ (n) ·
T ₂ (n), and so that by solving an equation expressed by P = P ₀ × R _T, obtaining a backscattering light intensity or backscattered light reflectivity from each optical fiber point.

That is, in the present invention, the data measured at each time by the OTDR method does not correspond to the backscattered light returned from one point of the optical fiber, but is distributed according to the spread of the incident light pulse width. Focusing on the sum of the backscattered light intensity from the point, formulating this and solving it,
The true backscattered light intensity at each point in the optical fiber is determined by the OTDR technique to reduce the measurement error of the backscattered light reflectance.

[Operation] An optical pulse having an impulse waveform as shown in FIG. 2 (a) is incident from an incidence end at a distance of 0 of an optical fiber having a backscattered light reflectance distribution as shown in FIG. 2 (b). By measuring the backscattered light waveform at the light incident end at a sampling time interval Δt, the distribution of the backscattered light reflectance R (x) of the optical fiber shown in FIG. 2 (b) can be calculated as Δx = Δt · v / 2. Data as shown in FIG. 2 (c) sampled at intervals is obtained. However, the optical pulse waveform that can actually be incident on the optical fiber is a waveform that has a temporal spread due to the time constant of the drive circuit, etc. Is done.

The reception intensity P (t) of the backscattered light at time t when a light pulse having a waveform of intensity P ₀ (τ) as shown in FIG. Width d
This is the sum of the backscattered light intensities of these impulse light pulses when it is considered that many impulse light pulses of τ are collected. The light pulse of τ to τ + dτ from the pulse generation time among the incident light pulses travels in the optical fiber by a distance of (t−τ) · v during time t, where v is the speed of light in the optical fiber. The backscattered light observed at the incident end at t is given by the position (t−τ) · v / 2, when going and returning.
It can be seen that this occurred in FIG. 1 (FIG. 1 (c)).

Therefore, the arrival ratio of the incident light pulse from the incident end to the position x is T ₁ (x), and the backscattered light reflectance at the position x is R.
(X), the arrival ratio of the backscattered light generated at the position x to the light incident end (backscattered light measurement end) is T ₂ (x), and the arrival ratio of the round trip is T (x) = T ₁ (x) · If T ₂ (x), the intensity P ₀
Reception intensity of the backscattered light by the optical pulse (pulse generation time from the tau ₁ time minute time width d.tau) of (tau) is, P
₀ (τ) · T ₁ ｛(t−τ) · v / 2｝ · R ｛(t−τ) × v /
2｝ · T ₂ ｛(t−τ) ・ v / 2｝ = P ₀ (τ) ・ T ｛(t−
τ) · v / 2} is × R {(t-τ) · v / 2}, the reception intensity of the backscattered light at time t by the total pulse width tau ₁ of the optical pulse shown in FIG. 1 (a) P (t) is Becomes

When measuring the backscattered light at the sampling interval Δt, the (n + 1) th measured backscattered light intensity P
(N) Becomes

Here, the incident light pulse (FIG. 1 (a)) is divided into (M + 1) at intervals of Δτ as shown in FIG. 1 (b) and approximated by a rectangular wave. Becomes However, T (x) and R (x) cannot be defined when x <0, and in fact, when x <0, T (x) = 0, R
If (x) = 0, there is no contradiction.

This equation shows that the backscattered light intensity sampled at the n-th time is represented by Δx between the distances nΔt · v / 2 to (nΔt−mΔτ) · v / 2 by (M + 1) incident light pulses having a pulse width Δτ. = Sum of the backscattered light intensities generated at points at a distance interval of Δτ · v / 2. As described above, since the n-th sampling data is not based on the backscattered light from one point in the optical fiber, the error is large in the conventional method.

Accordingly, the present invention measures the backscattered light reflectance using the relational expression (4) of the actual backscattered light reflectance with sampling data in the measurement of the backscattered light intensity in the OTDR method. .

In the equation (4), the incident light pulse intensity P ₀ (mΔτ) and T corresponding to the round trip of the transmission loss of light in the optical fiber.
Since {(nΔt−mΔτ) · v / 2} can be measured in advance, the backscattered light reflectance R ｛(nΔt−mΔ)
t) .v / 2} can be obtained by appropriately selecting the sampling interval Δt for measuring the backscattered light velocity and the division width Δτ of the incident light pulse, and measuring the backscattered light intensity P (nΔt). Equation (4) where Δτ = Δt is given by: Where n = n ₀ , and m = m ₀ , the backscattered light generation position x ₀ , _{0
= (N 0 -m 0) Δt} · v / 2 and _{n = n 0 + 1, m} = m 0 +1 backscattered light generation position x _{1 time, 1 = {(n 0 +1} ) - (m 0 +1)}・
Since Δt · v / 2 = (n ₀ −m ₀ ) · Δt · v / 2 matches, N
When the backscattered light intensity of the sample is measured, the number of positions where the backscattered light is measured is N, and the unknown number and the measured data number match, so that the unknown backscattered light reflectance can be calculated. Assuming that Δτ = k · Δt (k is a natural number), the number of measurement data is larger than the unknown number, so that the backscattered light reflectance can be calculated in this case as well.
In the case of ≧ 2, the backscattered light reflectance that can be calculated is smaller than in the case of k = 1 (Δτ = Δt) described above.

When an optical fiber having a length of L (= (N−M) × (Δt · v / 2) N = <N ′) is measured with the setting of the number of sampling data of (N ′ + 1), after (N + 1) Are the results of measuring the backscattered light reflectance from a position (space) farther from the far end of the optical fiber, but these values are generally much smaller than the backscattered light reflectance in the optical fiber. Therefore, the measured value of the backscattered light intensity may be considered to be zero. Accordingly, N + 1 backscattered light measurement data can be obtained from the optical fiber having the length L, where N + 1 is the number of backscattered light generation positions L (Δt · v measured when Δτ = Δt). / 2) M more than +1. Therefore, M more data are measured than the number of unknowns. The relationship between the unknown (the backscattered light reflectance R (n)) and the data (the backscattered light intensity P (n) measured at the measurement end) is n = 0, 1, 2,..., M + N
Equation (5) is considered, and the matrices P, P ₀ and
With R _T , And by solving this equation (6),
R, the true backscattered light intensity from each point of the optical fiber
_T (n) (n = 0, 1, 2,..., N) can be obtained. Further, if a relational expression of R _T (n) = R (n) · T (n) is used, the following equation can be obtained. R (n) = R _T (n) / T (n); n = 1, 2,..., N to determine the backscattered light reflectance R (n) (n = 0, 1, 2,..., N) be able to.

The unknown backscattered light reflectance R (n) is (N + 1)
And the backscattered light measurement data is (M + N + 1), so if only (n + 1) rows are performed in the matrix operation of the equation (6), (N + 1) backscattered light reflectances R (n )
Can be requested. That is, the matrix (N + 1) rows consisting of arbitrary (N + 1) rows of
(N + 1) square matrix And this inverse matrix Using Consists of (N + 1) rows Using Is a matrix that is the true backscattered light intensity from each point of the optical fiber It determined each element R _T (n), it is possible to obtain the (7) using equation backscattered light reflectance R (n).

By using the above calculation method, the backscattered light reflectance of the optical fiber can be calculated without approximation calculation.However, with some approximation, from the apparent backscattered light intensity that is the measured value, First, the transmission loss of the optical fiber is corrected, and the backscattered light reflectance of the optical fiber can be obtained from the result. That is, the matrix With α as an integer Where the value of α is (n + α) where the matrix of equation (6) The basic idea is to make the center column number of a non-zero element in the n-th row.

If n <M (n = 0,..., M−1), the matrix Are non-zero elements in the n-th row from the 0-th column to the n-th column, so the central column number of the non-zero element is n / 2. Therefore, in this case, α is −n / 2.

If M ≦ n <N, matrix Are non-zero elements in the n-th row from the (n-M) th column to the n-th column, so the center column number of the non-zero element is ((n-
M) + n) / 2 = n−M / 2. Therefore, in this case, α becomes −M / 2.

If n ≧ N, matrix Are non-zero elements in the n-th row from the (n-M) th column to the N-th column, and the central column number of the non-zero element is ((n-
M) + N) / 2. Therefore, in this case, α is (N−M−
n) / 2.

T (l), T (l-1),..., T (l-M) (where T (n) = 0 when n <0 or n> N) is replaced by one value T
When approximating by (l + α), for example, when approximating by an intermediate value, when l <M, T (l + α) = T (l / 2).

Thus, l + α = 1/2. Therefore, α = −l / 2.

When M ≦ l <N, T (l + α) = T (l−M / 2).

 Therefore, α = −M / 2.

When l ≧ N, T (l + α) = T ((l + N−M) /
2). Thus, l + α = (l + N−M) / 2. Therefore, α = (N−M−1) / 2.

Defined as above, matrix operation Think Is a method of obtaining R (n) by the calculation of This is
Th measurement data P (l) is (6) P (l) is a formula _{= P 0 (M) · R} T (l-M) + P 0 (M-1) · R T (l - (M-1) ) +... + P ₀ (0) R _T (l) = P ₀ (M) R (1-M) T (1-M) + P ₀ (M-1) R (1-(M-1) )) · T (l− (M−1)) +... + P ₀ (0) · R (l) · T (l)
(13) However, when n <0, R _T (n) = 0 and R (n) =
0, T (n) = 0, while T (l) = T (l + 1) =... = T (l + M) (14)

Here, the value of T (n) seems to be constant irrespective of the value of n, but the meaning of the expression (14) is as follows: When l = l _{1 in} the expression (13), T (l ₁ ) = T Assuming that (l ₁ +1) = ... = T (l ₁ + M), T (l ₁ +1) = T ((l ₁ +1) +1) = ... = T ((l ₁ +1) + M) When l = l ₁ +2, T (l ₁ +2) = T ((l ₁ +2) +1) =... = T ((l ₁
+2) + M), the value of T (n) is not constant regardless of the value of n.

Embodiment Hereinafter, as an embodiment to which the present invention is applied, a distributed temperature sensor using a change in backscattered light reflectance of an optical fiber due to temperature will be described with reference to FIG. An optical pulse emitted from the pulse light source 1 in FIG.
7a, the light is guided to the optical fiber for sensor 8 through the optical splitter or the optical multiplexer / demultiplexer 2. The backscattered light generated in the sensor optical fiber 8 passes through the optical splitter 2 and the optical fiber 7b, is converted into an electric signal by the light receiver 3, and is input to the averaging circuit 4. The averaging circuit 4 performs averaging processing, for example, ²¹⁶ times using the synchronization signal 10 output from the pulse light source 1. As a result of the averaging process, the light reception waveform of the light receiver 3 is as shown in the graph of FIG. Since the light pulse emitted from the light source 1 and the backscattered light returning from each point of the optical fiber 8 are attenuated when propagating through the optical fiber, the observed backscattered light intensity has a slope. Further, in this embodiment, since the heating device 9 to position x ₁ of the sensor optical fiber 8 are installed, protrusion of the waveform is seen in FIG. 5 (a) the position of the distance x _1.

The backscattered light intensity R _T (n) is calculated from the waveform obtained by the averaging device 4 using equation (8), and the backscattered light reflectance R (n) is further calculated from equation (7). By passing through the signal processing circuit 5 for calculating the temperature T according to the relational expression T = F (R (n)) (11) between the backscattered light reflectance R (n) and the temperature T obtained in the experiment, along the optical fiber. Temperature distribution can be obtained.
The calculation results of the backscattered light intensity, the backscattered light reflectance, and the temperature distribution in the signal processing circuit 5 are shown in FIG.
(C) and (d) show. FIG. 5 (b), which is the result of performing the signal processing according to the equation (8), is shown in FIG. 5 (a) of the original waveform.
The change of the waveform is sharper than that of the above, and the effect of the signal processing appears.

FIG. 6 also shows the measurement results of a distributed temperature sensor using the signal processing method of the present invention. FIG. 6 (a) is the actual temperature distribution, FIG. 6 (b) is the measurement result of the temperature distribution according to the present invention,
FIG. 6 (c) shows a measurement result by the conventional method. In the conventional method, the result is considerably different from the actual temperature distribution due to the rounding of the waveform. However, in the case of using the present invention, the actual temperature distribution agrees very well.

In this embodiment, the received apparent backscattered light intensity used for calculation in the signal processing circuit 5 is the received light amount P (0), P (1), P (2),... P (M + N) itself is not used, and these received light amounts are, for example,
Although electric signals V (0), V (1), V2,..., V (M + N) which have been passed through an O / E converter such as an APD and an electric signal amplifier are used, V (l) and P ( l) (l = 0, 1, 2,..., M + N) is, for example, V (l) = kP (l), k; constant or V (l) = g {P (l)} , g (x); function (12) or V (l) = h ｛P (0), P (1), P (2),.
(M + N)｝, h (x ₀ , x ₁ , x ₂ ,..., X _{M + N} ); Since it can be expressed as a function, the data used for signal processing includes the received light amount P (0), P (1), P (2),
..., electric signals V (0), V (1), V instead of P (M + N)
Even if (2),..., V (M + N) are used, the signal obtained by performing the inverse operation of the equation (12) from the backscattered light intensity and the backscattered light reflectance at each point of the optical fiber obtained by using these, The received light amounts P (0), P (1), P (2),.
It is possible to obtain almost the same result as that when (M + N) is used (although some errors may be included). The data used for signal processing is the received light amount P (0), P
(1), P (2),..., P (M + N), and these are derived from the electric signals V (0), V (1), V (2),.
2) It may be obtained by performing an inverse operation using the relationship of the expression.

Also, regarding the incident light pulse intensity P ₀ (m) in the equation (6), even if these are not the absolute value of the incident light pulse intensity, a function such as a proportional relationship between the incident light pulse intensity and the absolute value of the incident light pulse intensity is obtained. If there is a relationship that can be displayed, the absolute value of the backscattered light intensity and the backscattered light reflectance from each point of the optical fiber, or an amount proportional thereto can be obtained.
The incident light pulse intensity P ₀ (m) used in the equation need not be an absolute value.

On the other hand, the signal processing procedure in the signal processing circuit 5 is not limited to the above-described procedure. First, the optical fiber transmission loss T (l) is corrected, and then the backscattered light reflection is performed using the equation (10). The rate may be calculated, and finally the temperature may be obtained using, for example, equation (11). Alternatively, first, the optical fiber transmission loss T (l) is corrected (P → P _{1 / T} ), and Alternatively, the temperature may be converted into a temperature (P1 _{/ T} → T1 _{/ T} ), and finally, an accurate temperature distribution may be obtained by conversion in which P1 _{/ T in} Expression (10) is replaced with T1 _{/ T.} In this manner, the conversion from the electric signal to the optical signal performed by the signal processing circuit 5, the calculation for obtaining the accurate distribution as shown by the equations (8) and (10), the optical transmission loss T
The signal processing such as the correction of (l) and conversion into a temperature signal may be performed in any order.

In addition, T used for correcting the optical transmission loss of the optical fiber
(L) ie of the incident light pulse from the light incident end to the back-scattered light generation position l arriving ratio T ₁ (l) and the arrival from the back-scattered light generation position l of the backscattered light to the measurement end ratio T ₂ (l)
To be precise, the arrival ratio of the incident light pulse from the light source to the backscattered light generation position and the arrival ratio of the backscattered light from the backscattered light generation position to the light receiver, that is, the light emitted from the pulse light source 1 in FIG. The arrival ratio of the light pulse to the backscattered light generation position and the arrival ratio of the backscattered light from the backscattered light generation position to the light receiving surface of the photodetector 3 are shown in FIG. And the arrival ratio of the backscattered light from the backscattered light generation position to the connection end of the sensor optical fiber 8 with the optical branching device 2 from the backscattered light generation position to the backscattered light generation position. From the position of the optical waveguide on the light source side to the position of the optical fiber on the light source side, or from the position of the optical waveguide on the light receiver side, from the position of the optical fiber with the information to be known such as the backscattered light reflectance and the temperature,
Since these values are common to all data,
_It can be used as ₁ (l) and T ₂ (l).

The backscattered light used as a distributed temperature sensor includes a method using Rayleigh scattering having the same wavelength as the incident light pulse and Raman scattering having a different wavelength. Is to perform the same measurement by the method described above by using the optical splitter 2 of FIG. 4 as an optical splitter including an optical filter, or an optical splitter or an optical multiplexer / demultiplexer. Can be.

As a distributed temperature sensor using backscattered light from an optical fiber, it measures the backscattered light of anti-Stokes light and Stokes light, which are two components of Raman scattered light, and measures the backscattered light intensity, backscattered light coefficient, Alternatively, there is a method of obtaining a temperature distribution by taking a ratio of an amount proportional to these, but the present invention can be applied to this method.

That is, using a measuring system as shown in FIG. 7, the backscattered light waveform of the anti-Stokes light having the wavelength λa of the backscattered light is measured by the light receiver 3a and the averaging circuit 4a, and the wavelength λ
The backscattered light waveform of the Stokes light of s is measured by the photodetector 3b and the averaging circuit 4b, and the electric signal is
Conversion of optical signal, calculation for obtaining accurate distribution as shown in equations (8) and (10), correction of optical transmission loss T (l), calculation of ratio of anti-Stokes light to Stokes light, temperature signal The temperature distribution is displayed on the temperature distribution display 6. Also in this case, the order of the signal processing performed by the signal processing circuit 5 in the same manner as in the above-described example may be in any order.

As a distributed temperature sensor using backscattered light from an optical fiber, as shown in FIG. 8, pulsed light from pulsed light sources 1a and 1b enters from both ends of the optical fiber 8 and is emitted from both ends of the optical fiber 8. A method of calculating the temperature distribution by measuring the backscattered light using the light receivers 3c and 3d, respectively, and calculating the measurement results is also known, but the present invention is also applicable to this method. Yes, the measurement error can be reduced. In this case, the signal processing circuit 5 converts the electric signal to the optical signal, equation (8), (1)
0) Calculation for obtaining an accurate distribution as shown by the equation, correction of optical transmission loss T (l), processing of combining data measured at both ends of the optical fiber into data at each point of the optical fiber, temperature signal Conversion to etc. is performed. Also in this case, the order of the signal processing performed by the signal processing circuit 5 and the like may be any order.

[Effects of the Invention] In summary, according to the present invention, it is possible to improve the measurement accuracy of backscattered light even if basically the same light emitting / receiving element, driving circuit, and sampling and averaging circuit are used. This makes it possible to easily and economically detect a more accurate distribution of external factors such as loss of the optical fiber and temperature, humidity, and pressure.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram for explaining a signal processing method according to the present invention, FIG. 2 is a diagram showing a measurement result when impulse light is used as incident pulse light, and FIG. FIG. 4 is a diagram showing a conventional example of a waveform of a backscattered light returning with respect to a pulsed light and a reception signal obtained by sampling the waveform. FIG. 4 is a configuration diagram showing an example in which the method of the present invention is applied to a distributed temperature sensor. 6 and 7 are waveform diagrams showing measurement results obtained by the sensor, and FIGS. 7 and 8 are configuration diagrams showing another example of a distributed temperature sensor to which the method of the present invention is applied. In the figure, 1, 1a and 1b are pulse light sources, 2, 2a and 2b are optical splitters (or optical multiplexer / demultiplexers), 3, 3a to 3d are light receivers, and 4, 4a to 4d are averaging circuits, Is a signal processing circuit, 6 is a temperature distribution display, 7a
7d is an optical fiber, 8 is an optical fiber for a sensor, 9 is a heating device, and 10 is a synchronization signal of a pulse light source.

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Satoshi Yamamoto 5-1-1 Hidaka-cho, Hitachi City, Ibaraki Prefecture Inside the Wire Research Laboratory, Hitachi Cable Ltd. (72) Inventor Hisaichi Sasahara Hidaka-cho, Hitachi City, Ibaraki Prefecture 5-1-1, Nippon Electric Wire & Cable Co., Ltd. (72) Inventor Teruaki Tsutsui 5-1-1, Hidaka-cho, Hitachi City, Ibaraki Pref.

Claims (1)

(57) [Claims] 1. A time change of backscattered light with respect to a pulse incident light incident on an optical fiber is obtained, sampling is performed at Δt time intervals, this is repeated many times and an averaging process is performed to obtain time-series data. In the processing of the received signal of the optical fiber backscattered light, which calculates the backscattered light intensity in the length direction of the optical fiber by converting the change into the position change, the distance between the measurement ends is 0, the speed of light in the optical fiber is v, Δx =
When Δt · v / 2, Δ between distances x _{0 to} x ₀ + N · Δx
Backscattered light reflectance r of optical fiber at x-spacing position
(X ₀ ), r (x ₀ + Δx), r (x ₀ + 2 · Δx), ..., r (x ₀
+ N · Δx) are R (0), R (1), R (2),
..., R (N), and the pulse rise time of the transmitted pulse light is 0
And the time t ₀ ~t ₀ + M · sending pulse light intensity over time of Delta] t interval between Delta] t p ₀ when _{(t 0), p 0 (} t 0 + Δt), p 0 (t 0
+ 2 · Δt),..., P ₀ (t ₀ + M · Δt)
₀ (0), P ₀ (1), P ₀ (2),..., P ₀ (M), and the time when the pulse light is transmitted is 0, t ₀ ′ = x ₀ / (v / 2). time _{t 0 '~t 0' + (} M + N) · Δt of Delta] t backscattered light intensity of apparent measured by the measuring portion of the time interval p when the
(T ₀ '), p (t ₀ ' + Δt), p (t ₀ '+ 2Δt), ..., p
(T ₀ ′ + (M + N) · Δt) are P (0), P
(1), P (2), ..., P (M + N), and distance from the incident end
The incident light pulse arrival ratio up to x ₀ + n · Δx is represented by T ₁ (n),
The distance x of the backscattered light generated at the position of the distance x ₀ + n · Δx
The arrival ratio of backscattered light from + n · Δx to the measurement end is represented by T ₂
(N), T (n) = T ₁ (n) · T ₂ (n), and the following equation A method for processing a received signal of backscattered light from an optical fiber, comprising calculating the backscattered light intensity or the backscattered light reflectance from each point of the optical fiber by solving the following equation.