JPH04132932A - Auto-correlation type optical fiber failure point position inspector - Google Patents

Abstract

PURPOSE:To enable improvement ratio of dynamic range to be expanded and distance resolution to be improved by adding output of a first correlator and that of a second correlator for forming a reflection light of a transmission light pulse at a single optical pulse equivalently. CONSTITUTION:Four types of wavelengths lambda1-lambda4 are modulated by four electrochemical modulators 5-8 by four types of codes which are obtained by offsetting two types of complimentary codes, namely A code and B code, for obtaining one optical signal by a synthesizer 9. Reflection light from an optical fiber to be measured 12 is branched into four kinds of wavelengths lambda1-lambda4 by a branching filter 13 and they are converted to electrical signal by photoelectric converters 14-17. Then, output of the first and second photoelectric converters 14 and 15 is subtracted by a subtractor A18, auto-correlation is performed by a correlator A20, output of the third and fourth photoelectric converters 16 and 17 is subtracted by a subtractor B19, and then auto-correlation operation is performed by a correlator B21. Then, output of the correlator 20 is added to that of the correlator 22, thus enabling a reflection light of a transmission optical pulse to be formed as a single pulse with a large amplitude equivalently.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to an optical fiber fault locator (OTDR) that detects the location of fault points occurring in an optical fiber, and in particular improves the dynamic range and distance resolution. This invention relates to an autocorrelation type optical fiber fault position inspection machine.

(Prior Art) OTDR is an inspection machine used for locating break points, discontinuity points, etc. of optical fibers. Conventionally, even in coaxial cables, failure points have been searched for using the pulse method. When a pulse wave is propagated from one end of a coaxial cable, if there is a discontinuity point, reflection occurs due to impedance mismatch, and the reflected pulse wave returns to the input end. In the case of complete rupture, approximately 100% reflection occurs, and the position of the rupture point can be measured from the time difference between the incident pulse and the reflected pulse. In the case of optical fibers, the pulse method is similarly used to detect the break point, but the reflectance at the break point is only about 4% even in the ideal break state (mirror-like break perpendicular to the fiber axis). . Depending on the state of the break, there may be cases where almost no reflected waves are generated, and in this case, it is possible to locate the fault in the optical fiber by observing the backscattered light. However, in the case of an optical fiber with low loss such as a single-mode optical fiber, the backscattered light is weak, and it is necessary to send a high-energy optical pulse to detect the fault location. In order to send a high-energy optical pulse, it is possible to widen the pulse width of the optical pulse, but this reduces the temporal resolution. Therefore, it is sufficient to increase the energy of the optical pulse by increasing the amplitude. However, in order to increase the amplitude, there is a limit to the power of the light generated by the laser, and it is difficult to increase the amplitude sufficiently. 0TDR using an autocorrelation function is a method for effectively increasing the amplitude without widening the pulse width. A method of determining an autocorrelation function by autocorrelation will be explained with reference to FIG. (A) is the waveform of code A; 1. -1. 1. 1
．． It consists of a 4-bit code. (b) is the shifted code A, which is assumed to be A1. Find the correlation between (a) and (b).

Note that this also involves finding the sum of the products of codes A and A1 at each time. AXAI is 0,0,0,1°0.0, and the sum is 1. In (c), when we calculate the correlation between A2 and A by shifting A1 by one, AXA2 is 0.0. 1
．． -1.0.0, and the sum is O. In (d), when A2 is further shifted by one and the correlation between A and A3 is calculated, AXA is 0, -1. -1.1, and the sum is -1
It is.

In (e), when the correlation between A and A4 is found in the same way, AXA4 is 1. 1. 1. 1, and the sum is 4. Similarly, when the code A is shifted and the correlation is determined and synthesized, a signal having a waveform as shown in FIG. 3 is obtained. That is, the amplitude is 1. By determining the autocorrelation of a pulse with a pulse width (number of bits) of 4, a pulse with an amplitude of 4 can be obtained.

In the waveform shown in Fig. 3, there are many side lobes outside the main pulse, so in order to obtain a waveform in which the main pulse is in phase and the side lobes are in opposite phase, as shown in Fig.
Correlation between code B and shifted code B7 is determined using a method similar to that shown in the figure. The resulting waveform is the waveform shown in FIG. Here, when the waveforms of FIG. 3 and FIG. 4 are combined, the waveform of FIG. 5 is obtained. In this case, the waveform in FIG. 3 and the waveform in FIG. 4 are complementary and are called complementary codes. Also, when correlation is taken in this way, the codes that yield a peak in the center are M, J, E, Gola.
The Golay code introduced by y is used and the second
Since the side lobes are canceled by the waveforms shown in Fig. 3 and Fig. 3, it is called a complementary Golay code.

(Problem to be Solved by the Invention) By the way, the Golay code shown in FIG. 2 for obtaining the pulse shown in FIG. 3 is called code A, and the code for obtaining the pulse shown in FIG. 4 is called code B (not shown). do. Regarding code A, the amplitude of this code exists between -1 and 1, but since there is no negative pulse in the optical signal, it is necessary to offset it with a positive signal. For this reason, a conversion as shown in FIG. 6 is performed, which is converted into an electrical signal and then synthesized again to restore the original signal. In Figure 6, (a) is the original Golay code, (b) is code A+ which is code A plus 1 and divided by 2, and (c) is code which is code A plus 1 and divided by 2. It is A-. When code A- is subtracted from code A+, the original code A is obtained as shown in (d). The same applies to code B. In this way, 0TDR using autocorrelation has an advantage over normal 0TDR in terms of dynamic range because the peak value of the composite waveform becomes large, but in order to obtain one data, A+, i
, B, and B-, the improvement rate of the dynamic range is as follows.

In addition, the measured values corresponding to the four codes mentioned above will be taken into memory after AD conversion and will be digitally calculated, but the irradiation of the optical fiber with the pulsed light of the four codes is done serially in time. Therefore, memory is required before and after arithmetic processing, making the circuit configuration complicated. Further, since digital calculations are performed, signal processing is complicated. Furthermore, 0TD
The distance resolution of R is determined by the length of 1 bit of the code, so by reducing the length of 1 bit and increasing the number of irradiations, the SN
High-speed processing is essential to prevent a decrease in the ratio, but there is a limit to how high the processing speed can be increased.

The present invention has been made in view of the above points, and its purpose is to expand the improvement rate of the dynamic range and realize autocorrelation type 0TDR with improved distance resolution.

(Means for Solving the Problems) The present invention solves the above problems by transmitting optical pulses having four different pulse codes created by offsetting two types of complementary pulse code signals to an optical fiber under test. In an autocorrelation type optical fiber fault position inspection machine, the reflected light from the fault point position is returned to the original two types of codes, and then autocorrelation is performed and summed for each of the four different pulse codes. The first modulates each of four different wavelengths of light using a signal. Second. Third. a fourth electro-optic modulator; Second. Third. A multiplexer that combines the lights of four different wavelengths from the fourth electro-optic modulator into one optical signal and guides it to the optical fiber to be measured, and backscattered light returning from the optical fiber to be measured. a splitter that separates the Fresnel reflected light into the four different wavelengths; and a first splitter that converts the optical signal separated by the splitter into an electrical signal.
2nd, 3rd. a fourth photoelectric converter; a first method for determining the difference between the output signals of the first photoelectric converter and the second photoelectric converter;
a second subtracter for determining the difference between the output signals of the third photoelectric converter and the fourth photoelectric converter;
a first correlator that calculates the autocorrelation of the output signal of the subtracter; a second correlator that calculates the autocorrelation of the output signal of the second subtractor; and an output signal of the first correlator. and an adder that adds the output signals of the second correlator and equivalently forms the reflected light of the transmitted light pulse into a single light pulse.

(Operation) Using four types of codes obtained by offsetting two types of complementary codes, light of four types of wavelengths is modulated by four electro-optic modulators, and a multiplexer generates 1' optical signals. shall be. After the reflected light from the optical fiber to be measured is divided into four types of wavelength light by a demultiplexer and converted into electric signals by first to fourth photoelectric converters, the first to fourth photoelectric converters convert the reflected light into electrical signals. The second photoelectric converter output is subtracted by the first subtracter, the first correlator performs an autocorrelation calculation, and the third. The output of the fourth photoelectric converter is subtracted by the second subtracter, the second correlator performs an autocorrelation operation, and the output of the first and second correlators is added to equivalently calculate the output optical pulse. Forms the reflected light into a single pulse with large amplitude.

(Example) Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a block diagram of one embodiment of the present invention.

In the figure, reference numeral 1 denotes an A code generator that generates a code A waveform signal using the Golay code shown in FIG. 2, for example;
2 is a B28 generator that generates a code B waveform signal which is a complementary signal to the code A signal. 3 is an offset shifter A which offsets the output code A of the A go 1 generator 1 and outputs a signal having a waveform of code A or code A- shown in the following equation.

■ A, - (1 + A) ... (1)
1--(1-A)...
(2) Waveform conversion by offset shifter A3 is shown in FIG. In the figure, (a) is the waveform of code A, (
b) is the waveform of A+ obtained by calculating equation (1), (
C) is the waveform of A- obtained by calculating equation (2). (D) shows the waveform of the signal restored to the original code A by subtracting A+ in (B) and A- in (C). 4
is an offset shifter B which similarly offsets the code B waveform signal output from the B-to-1 generator 2 and outputs the code B+ and code B waveform signals shown in the following equation.

B, −−(1+B) ・・・ (3)■ B −−(1−B) ・・・ (
4) 5 is an electro-optical modulator that modulates the manually input light of wavelength λ with a signal of code A and outputs it, and 6 modulates the input light of wavelength λ2 with a signal of code A- and outputs it. An electro-optic modulator 7 modulates manually input light of wavelength λ3 with a signal of code B and outputs it; 8 modulates input light of wavelength λ4 with a signal of code B-. It is an electro-optic modulator that outputs. This wavelength λ1. λ2. λ3. λ
The light No. 4 is selected to have a wavelength difference as small as possible without being affected by the wavelength dispersion of the optical fiber to be measured and without reducing the crosstalk caused by the demultiplexer.

9 has a wavelength of λ1. λ2. λ3. Combine light waves of λ4 to 1
A multiplexer that converts the light into two light waves, and the output light is sent to the optical coupler 10.
is introduced into the optical fiber to be measured 12 via the optical connector 11. The optical coupler 10 has directionality, and is designed so that the light input to the C terminal of the optical coupler 10 shown in the figure is output from the b terminal, and the light incident from the b terminal is output from the C terminal. There is.

13 is the wavelength λ1.1 returned from the optical fiber 12 to be measured.
λ2. λ5. The reflected light, etc., which is the combined light of λ4, is divided into wavelengths λ
1. λ2. λ2. A demultiplexer separates the demultiplexed wavelengths λ1 to λ4, respectively. λ2゜λ1. Outputs light of λ4.

Reference numeral 14 denotes a photoelectric converter into which light of wavelength λ1 is input and converts it into an electrical signal, and its output signal is a signal expressed as SA, where S (t) is the response characteristic of the optical fiber 12 to be measured.

15, 16, and 17 are similarly set to wavelength λ2. Light of λ3゜λ4 is input and converted into electric signals, and the electric signals Si, SB
, , is a photoelectric converter that outputs SB-. 18 is signal S
Subtractor A, 1 which receives A+ and SA- and calculates the difference between them.
9 is a subtracter B which receives the signals SB and SB- and calculates the difference between them.

20 is a correlator A121 that generates a pulse wave by calculating the correlation between the output signal of the subtracter A and the output signal of the A go 1 generator 1.
is a correlator B that calculates the correlation between the output signal of the subtracter B and the output signal of the B-to-1 generator 2 to generate a pulse wave. These two outputs are added by an adder 22, the side lobes are canceled, and only the main pulse is generated. 23 is an AD converter that converts the output of the adder 22 into a digital signal;
Reference numeral 4 denotes a memory that stores electrical signals generated by reflected light from optical pulses emitted multiple times. When the signals have been stored a predetermined number of times, they are read out to an averaging circuit 25, averaged, and displayed on a display 26. Is displayed.

Next, the operation of the embodiment configured as described above will be explained. The code A signal generated by the A go 1 generator 1 is subjected to arithmetic processing according to equations (1) and (2) in the offset shifter A3, and outputs a code A signal or a code A- signal.

Further, the B-to-1 generator 2 generates a code B signal, which is processed by the equations (3) and (4) in the offset shifter B4 to generate the code B signal. , outputs a code B- signal.

The electro-optic modulator 5 modulates the input light of the wavelength λ1 using the manually input signal of the coat A4, and outputs the light of the wavelength λ1 encoded by the code A and the like. Electro-optic modulators 6, 7 and 8 receive input codes i, B, .
B- signal causes the wavelength λ2. λ1. The light of wavelength λ4 is modulated to produce wavelengths λ2. λ
Outputs light of 3°λ4. These four optical signals are combined by a multiplexer 9 into a signal of λ1+λ2+λ3+λ4, which is inputted from an optical coupler 10 to an optical fiber 12 to be measured via an optical connector 11.

The backscattered light, Fresnel reflected light, etc. that have returned from the optical fiber 12 to be measured pass through the optical connector 11 to the b of the optical coupler 10.
The signal returns to the terminal, is output from the C terminal, and is introduced into the duplexer 13. The demultiplexer 13 divides the input optical signals into wavelengths λ.
1. λ2. The light is separated into wavelengths λ3° and λ4, and wavelengths λ1. λ2.
λ9. The light of λ4 is transmitted to photoelectric converters 14, 15, 16, respectively.
Enter in ゜17. Photoelectric converter 14, 15, 16゜17
converts each light into an electrical signal and outputs the signal shown below. That is, the photoelectric converter 14 outputs an electric signal SA+ (an electric signal from the light reflected by the light of code A4), and the photoelectric converter 15 outputs an electric signal SA-, which is manually input to the subtractor A18, and is then outputted from the photoelectric converter 15. The converter 16 outputs an electric signal SB, and the photoelectric converter 17 outputs an electric signal SB-, which is manually input to a subtracter B19.

Here, when the response characteristic at the fault point of the optical fiber 12 under test is 5(t), the reflected wave SA+ of the optical signal with code A is expressed as the superposition integral of the response characteristic S(t) and code A. It can be expressed as the following equation. (*: superposition integral) S A, -S (t) *A and...(
5) The following formula can be obtained similarly for SA-, SB, and SB-.

S i-S (t)*A-...(6) SB or -S (
t)SB+...(7)SB,--5(
t)*B-...(8) Subtractor A18 receives the input SA
+ and SA- are subtracted as shown in the following equation and output. (5
) formula, (6) formula, (1) formula, and (2) formula, SA and -8A--3(t)*A+-8(t)*^-3(
t)*(A+-^-) -3(t)*A...(9) Subtractor B
19 subtracts the input SB+ and SB-, and outputs the signal of the following equation from equations (7), (8), (3), and (4) in the same way as above.

SB+-3B--3(t)*B...(10)
The correlator A20 receives the input signal of equation (9) and the A21 generator 1.
An autocorrelation calculation is performed with the output signal of , and a signal expressed by the following equation having a positive peak near the center as shown in FIG. 3 is output.

(s (t) * A )★A ... (11
) However, ★... The autocorrelation correlator B21 performs autocorrelation calculation between the input signal of equation (10) and the output signal of the B21 generator 2, and a positive peak is generated near the center as shown in FIG. In the other parts, a signal having a waveform of the following equation which cancels the waveform shown in FIG. 3 by addition is output.

(S (t) * B )★B ... (12
) Adder 22 adds the outputs of correlator A20 and correlator B21 to produce a sharp signal with a large peak value as shown in FIG. The same effect as that of 12 is obtained. This output is converted into a digital signal by the AD converter 23 and stored in the memory 24. When the reflected light from a certain number of optical pulses is stored in the memory 24, the averaging circuit 25 extracts the reflected light and calculates the average value to improve the accuracy of detecting the location of the fault point. The averaging circuit 25 causes the display 26 to display the averaged image data.

Next, the magnitude of the output data obtained by the above embodiment is determined. As mentioned above, the response characteristic of the optical fiber 12 to be measured is assumed to be S (t). 5(t) can be obtained by observing the output when a delta function δ(1) with a pulse width of 5(t) is input. The signal applied with this delta function is given by the following equation.

δ(t)+5(t)-J” 5(t)δ(t-τ)dτ
(13) The complementary Golay codes A and B used in this example have the following relationship.

A*A+B*B-2Lδ(1)...(1
4) However, L... The output of the bit number photoelectric converter 14 is calculated from equations (5) to (8) as 'SA, -8(t)*^, -, l'5(t)A, (
t-τ)dτ... (15) SA--3(t)*A--J' 5(t)A-(t-
τ) dτ... (16) SB, -3(t)*B and -, l' 5(t)B and (t
-τ)dτ... (17) SB--8(t)*B--J' 5(t)B-(t-τ
)dτ... (18) Input signal SA of subtractor A18 and subtractor B19. .

SA-, SB, SB- are correlator A20 and correlator B21
And the adder 22 performs the following calculation. (15
), (16), (17), (18) and (14)
) From the formula, (SA, -38-)★A+ (SB+-3B-)★B-
(S(t)*A+-3(t)*A)★^+ (S(
t)*B, -8(t)*B-)★B-8(t)*(A
or -A-)★A +5(t)*(B or -B-)★B From A+-A--A, B+-B, -B, the given formula -5(t
)*(A*A+ B*B)-2Lδ(t)IS(t)
...(1) Equation (19) and (13)
Comparing with the formula, it can be seen that the output is 2L times larger. This means that by sending an optical pulse using a complementary Golay code to the optical fiber under test, the output is 2L times as large as that of simply sending a pulse shaped like a delta function, and the reflectance is This makes it possible to capture reflected light and backscattered light from a small single-mode optical fiber failure point.

Also, in this embodiment, there are four A, i, B, and so on.

B- signals are sent at the same time, and the required measurement value SA,.

Since SA-SB,,SB- can be obtained in one measurement, the improvement rate of the dynamic range goes from r7 to 'T, which improves the dynamic range by 2 times and eliminates obstacles with small reflection coefficients. It is also possible to capture reflected light from a point.

With the conventional autocorrelation type 0TDR, it is necessary to store all four measured values of the reflected light beams incident on the optical fiber under test into memory and perform digital calculations to obtain the results. Processing is complicated and the burden on the signal processing circuit is heavy, which is a limiting condition for speeding up. However, in this example, it is possible to perform analog calculations up to the final result, and digital signal processing is not required. Basically, it is just averaging, so it is easy to speed up the processing, and it is an autocorrelation type 0TD that can be processed at high speed and has high resolution.
R can be obtained.

(Effects of the Invention) As explained in detail above, according to the present invention, an optical pulse of effectively large amplitude can be sent, the improvement rate of the dynamic range is expanded, and the autocorrelation type 0TDR has improved distance resolution. can now be realized, and the practical effects are significant.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is an explanatory diagram of the autocorrelation of the Golay code, Fig. 3 is a diagram of the waveform obtained by the autocorrelation according to Fig. 2, and Fig. 4 is a diagram illustrating the autocorrelation of the Golay code. Figure 5 is a diagram of the waveform obtained by the complementary Golay code of the code in Figure 2, Figure 5 is similar to Figures 3 and 4.
FIG. 6 is an explanatory diagram of the waveform obtained by the sum of the waveforms shown in FIG. 1... A code generator 2... B code generator 3... Offset shifter A 4... Offset shifter B 56.7.8... Electro-optic modulator 9... Multiplexer 10. ...Optical coupler 12
・・Optical fiber to be measured 13 ・・Brancher 14.15,
16.17...Photoelectric converter 18...Subtractor A
19...Subtractor B20...Correlator A
21... Correlator B22... Adder
23...AD converter 25...Averaging circuit XAl 0oO010000 XA2 XA3 oo-+-++o. AX△6 Figure Figure 0-101 4 To-10 Figure

Claims (1)

[Claims] Optical pulses with four different pulse codes created by offsetting two types of complementary pulse code signals are sent to the optical fiber to be measured, and the reflected light from the fault point position is used as the source. After returning to the two types of codes, an autocorrelation type optical fiber fault position inspection machine that performs autocorrelation and adds them each modulates light of four different wavelengths with the four different pulse code signals. 1st, 2nd,
third and fourth electro-optic modulators (5, 6, 7, 8); and the first, second, third and fourth electro-optic modulators (5, 6, 8).
A multiplexer ( 7, 8) combines the lights of four different wavelengths into one optical signal and guides it to the optical fiber under test (12).
9), a demultiplexer (13) that separates backscattered light, Fresnel reflected light, etc. returning from the optical fiber to be measured (12) into the four different wavelengths of light, and the demultiplexer (13). ) converts the separated optical signal into an electrical signal.
15, 16, 17), and a first subtractor (18) that calculates the difference between the output signals of the first photoelectric converter (14) and the second photoelectric converter (15).
and a second subtractor (19) that calculates the difference between the output signals of the third photoelectric converter (16) and the fourth photoelectric converter (17).
a first correlator (20) that calculates the autocorrelation of the output signal of the first subtractor (18); and a second correlator (20) that calculates the autocorrelation of the output signal of the second subtractor (19). correlator (21), the output signal of the first correlator (20) and the output signal of the second correlator (21) are added to equivalently convert the reflected light of the transmitted light pulse into a single light. An autocorrelation type optical fiber fault position inspection machine characterized by comprising an adder (22) for forming pulses.