JP3006025B2 - Judgment method of code error rate of optical transmission characteristics and automatic recognition evaluation method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to a method for judging whether or not a bit error rate of an optical transmission characteristic is good and an automatic recognition system, and in particular, to automatically determine whether or not a bit error rate of an optical transmission characteristic of an optical carrier is good. The present invention relates to a method for judging whether or not a bit error rate is good or bad, and a method for automatically recognizing and judging whether or not a bit error rate is good or bad for an optical transmission characteristic of an optical carrier device, and a method for automatically recognizing a bit error rate of a light transmission characteristic of a bit error rate.

[Prior Art] A conventional method for determining the quality of a bit error rate of optical transmission characteristics is to input an NRZ signal and a clock signal generated from a pulse pattern signal generator to a device under test, and the device under test clocks the NRZ signal. After converting to an optical signal synchronized with the signal and input to the optical attenuator with an optical fiber, the optical attenuator is trial and error to the optical power value at which measurement data is likely to be obtained at the order of the error rate desired by the operator. Operate to set the optical power value.
The optical signal attenuated according to the optical power is input again to the device under test by the optical fiber, converted into an electric signal, and input to the error rate measuring device. The error rate measuring device measures and displays the bit error rate of the signal from the device under test according to the gate opening / closing time according to the specified order of the error rate. The operator measures the trend of the mantissa of the displayed error rate data,
Plot the data on a standard error rate sheet according to each error rate order. Next, the plotted data is used to find an approximate linear trend, and the error rate at the minimum value of the optical power is used as a reference point, and a straight line that makes the slope of the straight line the smallest is intuitively drawn. , Subjectively determine if the data is scattered,
The quality of the optical transmission characteristics of the device under test was determined. Also,
Although facilities that partially automate the measurement have been realized,
When displaying the measurement error rate paper on the CRT on the microcomputer software and plotting the data of the measuring instrument, the measurement data was displayed in terms of the distance from the minimum point on the vertical axis of the standard error rate paper . In the determination of linearity, a straight line is estimated using the least squares method, and the distance between the estimated straight line and the data is measured to determine the linearity.

[Problem to be Solved by the Invention] In the above-mentioned conventional method for determining the quality of the bit error rate of the optical transmission characteristic, the measurer changes the optical power to the specified error rate by changing the optical attenuator and sets the optical power to the specified value. There was a drawback that it required a lot of skill to begin with, and it required a lot of man-hours for beginners. If the optical power value cannot be adjusted to the specified value, the code error rate in the specified order cannot be obtained, or a data group with a linear tendency cannot be obtained. Since the plotting is performed again on the paper, there is a disadvantage that the measurement time is increased and a large number of man-hours are required for determining the data processing. Also, when drawing a standard code error rate paper on software, it is difficult to accurately display the measured code error rate data on the error rate axis indicated by the integral value of the residual error function. There was a disadvantage. Further, even if a straight line is estimated by using a numerical calculation method such as a least square method using a microcomputer, such a case is possible when the bending method of the error rate measurement data is large or when the variation of the measurement data is large. Under the influence of the measurement data group,
There is a drawback that a correct error rate estimation line cannot be obtained. Therefore, the judgment by the computer is limited to the reference data, and the measurer performs the adjustment and the judgment of the data subjectively. there were.

[Means for Solving the Problems] A method for determining the quality of a bit error rate of an optical transmission characteristic according to the present invention comprises: measuring a light power value of a laser of a device under test; A procedure for measuring the optical transmission bit error rate, a procedure for accurately displaying the bit error rate data corresponding to the optical power output from the error rate measuring instrument on the measurement graph, and an optical power for the data displayed in the graph. Passing through the error rate data at the lowest value of, and on the basis of this point, a procedure for obtaining a straight line having the smallest slope; and code error rate data used for obtaining a straight line having the smallest slope by the means, Automatically determining whether the arrangement of the remaining measurement data that does not lie on a straight line excluding the points is linear, based on the straight line obtained by the above means, and It is characterized by.

The method for automatically recognizing and evaluating the bit error rate of the optical transmission characteristic according to the present invention is a means for measuring the optical power value of the laser of the device under test and measuring the optical transmission code error rate of the device under test according to the optical power value. And a means for accurately displaying the bit error rate data corresponding to the optical power output from the error rate measuring device on the measurement graph, and comparing the data displayed on the graph with the error data at the lowest value of the optical power. As described above, on the basis of this point, means for finding a straight line having the smallest slope, code error rate data used for finding a straight line having the smallest slope by the means, and a straight line excluding the reference point A means for automatically evaluating whether there is linearity when the arrangement of the remaining measurement data is based on the straight line obtained by the means, and the straight line obtained by the means and the error rate graph from the origin. Express distance explicitly, The distance, characterized in that it comprises an automatic recognizing means the quality of the optical transmission properties of the object to be measured from the evaluation value of the linearity of the error rate data.

Example Next, an example of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a measuring system according to one embodiment of the present invention. The microcomputer 1 generates a setting signal a of the pulse pattern signal generator 2, which is input to the pulse pattern signal generator 2. The pulse signal generator 2 generates a clock signal b having a frequency F1 (MHz) and a transmission data number X.
An NRZ signal c having (M bit / sec) is generated and input to the DUT 3. The NRZ signal c is converted into an optical main signal d on the device under test, and a standard optical power value P ₁ (dBm) required by the optical attenuator 4.
Is converted into a main signal light data signal e having the following. On the other hand, the interference wave signal f
Is generated and input to the frequency generation circuit 7, and the frequency F ₂ (MH
The interference wave data signal g of z) is output and input to the E / O converter 8. The interference wave data signal g is converted into an interference wave optical signal h by an E / O converter 8, and a standard optical power value required by an optical attenuator 9.
The signal is converted into an interference-wave standard optical signal i having P ₂ (dBm) and input to the optical combiner 5. In the optical combiner 5, the main signal optical data signal e and the interference-wave standard optical signal i are merged to form an optical data signal j.
Occurs. The optical data signal j is input to the optical attenuator 10. Here, an optical attenuator control signal n for adjusting to an optical power value previously input by the microcomputer 1
Is input to the optical attenuator 10. The optical attenuator 10 attenuates the optical power value of the optical data signal j and outputs an optical attenuated data signal k. The optical attenuated data signal k is input to the optical splitter 11, split into two, and then input to the optical sensor 12 and the device under test 3. The optical attenuation data signal k is converted into an optical power signal 1 by the optical sensor 12 and taken into the optical power meter 13. The optical power signal 1 is converted into numerical data by an optical power meter, and is again input to the microcomputer 1 as an optical power measurement data signal m. The microcomputer 1 checks whether or not the optical power value is specified in advance. If the optical power value does not match the specified optical power value, the microcomputer 1 again generates an optical attenuator control signal n,
Is adjusted, and this operation is repeated until the specified optical power measurement value is reached. Optical attenuation data signal k input to DUT 3
Is again converted into a product output signal o by the E / O conversion circuit of the device under test 3 and input to the bit error rate measuring device 14. The code error rate measuring device 14 counts the number of code errors of the product output signal o,
A code error rate with respect to the number of transmission data X [M bit / sec] is calculated, an error rate data signal p is output, and is taken into the microcomputer 1.

Next, the operation of this embodiment will be described in more detail.
FIG. 2 is a flowchart of a method for processing the measurement data information taken into the microcomputer 1. The operator specifies the number N of measurement points in step 15. Next, in step 16, an optical power value corresponding to the order of the bit error rate to be measured is specified, and in step 17, the measured optical power value L (i) (i = 1, 2,...)
N) is stored in the memory of the microcomputer 1. Further, in step 18, the microcomputer 1
The optical attenuator 10 shown in FIG.
m]. In step 19, the bit error rate measuring device 14 measures the bit error rate measuring device 14 at each measured light power value L (i) [dBm]. In step 20, the bit error rate Err (i) is stored in the memory of the microcomputer 1. Is stored in In step 21, the microcomputer 1 sets the error rate
Err (i) = a (i) · 10 ^{−b (i)} is ^converted to a mantissa part a (i) (0 <
a <(i) <10 real number) and exponent part b (i) (3 <≦ b
(I) an integer of ≤12). The relational expression between the optical power value L (i) and the bit error rate Pe (i) in the optical digital transmission system is L (i) = 20 · Log ([L (i)]) [dBm] (1) Pe (i ) = 1/2 · erfc ([L (i)]) (2) i = 1, 2, 3,... N. Here, [L (i)] is an antilog of L (i), and erfc (a) is a residual error function. The integral of the residual error function of the equation (3) cannot be obtained analytically. Therefore,
In order to accurately approximate equation (2), in step 22, the value of the error rate on the vertical axis coordinate of the standard code error rate paper is 1.0 * 1.
The vertical axis coordinate value of 0- ^{b (i) which} is error-free (the state where the code error rate is 0) is Linα (b (i)), and the vertical axis coordinate value of a point which is 5.0 * 10- ^{b (i)} is Expressed as Linβ (b (i)), and based on these vertical coordinate values, an exponent part of Err (i) is used to plot the measurement error rate Err (i) on the approximate value of equation (2). When linear conversion is performed according to the value of b (i) and the value of the mantissa part a (i), 1) When 1 ≦ b (i) ≦ 3, Pe (i) = Linα (3) (4) 2) b (I)> 12 Pe (i) = Linα (12) (5) 3) 3 <b (i) ≦ 12 Case 1) 0 <a (i) <Pe (i) = Linβ (B (i)) + [Linα (b (i)) − Linβ (b (i))] Log (a (i)) / log (5) (6) Case 2) 5 <a (i) When <10 Pe (i) = Linβ (b (i)) + [Linβ (b (i) −1)) − Linα (b (i))] · Log (a (i) / 5 / Log becomes (2) (7). Each optical power value and the linear conversion code error rate at that time are stored in the memory of the microcomputer 1 in step 23 as two-dimensional coordinate values (L (i), Pe (i)).

The linear transformation code error rate Pe is displayed on the vertical axis of the standard code error rate paper.
(I), the measured light power value is taken on the horizontal axis, the angle with the counterclockwise direction being positive is θ, the distance between the error rate line and the origin is ρ, the light power value L (i), the linear conversion code error rate Pe (I)
Is a variable, the error rate line to be obtained can be expressed as L (i) · cos θ + Pe (i) · sin θ = ρ (8) Equation (8) shows that when the values of θ and ρ are fixed, L
On the -Pe plane, the straight line shown in FIG. 3 is expressed, and L, Pe
Is fixed, a point on the L-Pe plane is a Hough transform showing a curve on the θ-ρ plane shown in FIG. Equation (8) expresses the product characteristics from the actually measured data by the distance ρ between the origin O on the L-Pe plane and the error rate straight line and the slope θ of the straight line in the current judgment method. It has the advantageous feature that it can be expressed in Two-dimensional coordinates (L (1), Pe that are the minimum of the measured light power value)
Assuming that (1)) is a reference point, these coordinates are present on the error rate line to be obtained. Therefore, the inclination between this point and each measurement point is obtained by means 24. The inclination θ at that time is θ (i) = tan ⁻¹ [((Pe (i) −Pe (1)) / ((L (i) −L
(1))]... (9) ρ (i) = L (i) · cos θ (i) + Pe (i) · sin θ (i) (10) (Θ
(I), ρ (i)) are stored in the memory of the microcomputer 1. The minimum value of each θ (i) obtained by equation (9) is searched for in step 26, and i = K at that time is set. The value of θ (K) is stored in the memory of the microcomputer 1 in step 27 as θa. Then the two-dimensional coordinate value at the time of the procedure 28 θ = θ _a value of [rho (i) in (L (i), Pe (i)) was calculated using equation (10), steps 29
Is stored in the memory of the microcomputer 1 as θ-ρ plane transformed coordinates (θ _a , ρ _a (i)). In equation (9), ρ _a (j) for i = K is set as ρ _a and (θ _a ,
Assuming that ρ _a ) is the reference coordinate of the θ-ρ plane conversion and L and Pe are variables, the ideal error rate estimation straight line to be obtained on the L-Pe plane can be calculated by the following equation (8): ρ _a = L · cos θ _a + Pe · sin θ _a ... (11) and in step 30 as an ideal error rate estimation straight line candidate,
It is stored in the memory of the microcomputer 1. The microcomputer 1 determines the standard code error rate determined in step 31. Read the predicted optical power level for
Equation (11) L = ρ _a −Pe · sin θ _a / cos θ _a (12) is calculated, and it is determined in step 33 that L is within the standard of L <. If this standard is not satisfied, it is determined that the set value of the measured light power is defective, and the procedure returns to step 15 to instruct the measurer to reset the measurement point. L <expression that satisfies the standards (11) has a minimum value theta _a to theta, straight starting point on the L-Pe plane since always passes through the (L (1), Pe ( 1)), ρ a Is also minimized. Therefore, the error rate characteristic is θ
Since it is known that the smaller the values of ρ and ρ are, the higher the quality is, it can be said that Expression (11) indicates an ideal error rate straight line. According to the Hough transform theorem, when a group of points on the same straight line on the L-Pe plane is transformed into a group of curves on the θ-ρ plane shown in FIG.
For 1), two-dimensional coordinates (L (i), Pe (i)) (i = 1,
As shown in FIG. 4, the degree of variation in K) is determined by substituting the two-dimensional coordinates (L (i), Pe (i)) into equation (8) and fixing the θ and ρ parameters separately. Then, the degree of convergence to (θ _a , ρ _a ) on the θ-ρ plane may be checked. Now, θ = θ _a as steps 28, .rho.a obtained in Step 29 (i) is present on a straight line parallel to the ρ axis indicated by the vertical axis in FIG. 4. This indicates a straight line group parallel to the equation (11) on the L-Pe plane. Optical power ± ΔL for equation (11)
Since points within [dBm] are regarded as the same straight line existing in the shaded portion in FIG. 6, Δρ is calculated from the geometrical relationship diagram in FIG. 6 by Δρ = | ρ _a (i) −ρ _a | ≦ ΔL Cos (θ _a ) (13) Therefore, when the error rate measurement value on the L-Pe plane starts to bend, ρ _a (i) cannot satisfy the equation (13). It can be automatically determined whether or not the values have linearity. In addition, each two-dimensional coordinate data (L
(I), Pe (i)) approaches equation (11) as much as Δρ
→ 0 and the degree of convergence to (θ _a , ρ _a ) on the θ-ρ plane is good, so that the sign of ρ _a (i) −ρ _a is negative and Δ
It can be said that the smaller the ρ, the better the linearity of the relative position of each measurement point directly to the estimation of the ideal error rate, that is, the better the product characteristics.

On the other hand, the two-dimensional coordinate data with respect to the ideal error rate estimated straight line to see theta direction displacement (L (i), Pe ( i)), ρ (i) = fixed to [rho _a, ± front and rear theta _a Assuming that ε is a threshold, the displacement of θ is obtained by transforming equation (8) ρ _a = L (i) · cos θ + Pe (i) · sin θ, V = L (i) ² + Pe (i) ² (15) φ = atn ⁻¹ (L (i) / Pe (i)) (16) By means 35, the microcomputer 1 calculates Expressions (14), (15) and (16). displacement [Delta] [theta] in the theta direction [Delta] [theta] = the means 36 | is calculated ≦ epsilon ... and _{(17) | θ a (i} ) -θ a. θ _a (i) exists at a point on a straight line parallel to the θ axis in the θ-ρ plane shown in FIG. 4, and on the L-Pe plane, as shown in FIG. Is rotated as a tangent to _a circle having a radius ρ _a centered on the origin, and is equivalent to an angle when passing through the two-dimensional coordinates (L (i), Pe (i)). Therefore, as Δθ → 0 in equation (17) approaches, that is, the degree of coincidence of this tangent with the estimated straight line in the angular direction can be evaluated. Δθ is equivalent to the degree of bending of each two-dimensional coordinate data (L (i), Pe (i)) with respect to the estimated straight line. Thus, as a characteristic evaluation criteria for product for each measurement ^{_{data, W a (i) 2 =}} 1/2 [(θ a (i) -θ a) 2 + (ρ a (i) -ρ a)
² ] ... (18) is defined, and the means 37 is calculated by the equation (18). W _a (j) is a point on the circumference around the point (θ _a , ρ _a ) when θ _a (i) and ρ _a (i) shown in FIG. In FIG. 4, a point set on ρ = ρ _a , θ = θ _{a in a} rectangle indicated by a solid line centered on the point (θ _a , ρ _a ). The first term on the right side of equation (18) is the two-dimensional coordinate data (L (i), P
e (i)) indicates the degree of data bending on the L-Pe plane, and the second term on the right side indicates the degree of coincidence of the relative position with respect to the ideal error rate estimation straight line. The smaller the value of W _a (j), the better the degree of convergence to (θ _a , ρ _a ), that is, the higher the degree of coincidence with the ideal error rate estimation straight line. J (i) = ΣW _a (i) ² (19) is calculated as an evaluation function of the data comparison criterion, and the means 39 indicates whether the linearity is good or bad. Equation (19)
Is the frequency of distribution of the two-dimensional coordinate data (L (i), Pe (i)) with respect to the ideal estimated error rate straight line different for each product lot, that is, the numerical value of the quality level. Therefore, the quality level of each product lot can be compared by comparing the calculated values of Expression (19). FIG. 9 shows an example in which, when ε = 4 [DEG] and the allowable judgment standard value is 0.7, Δρ = 0.21 and Δθ = 1.25 [deg], which are within the standard. FIG. 10 shows an example in which the measured error rate straight line has no optical power value with respect to the standard code error rate, and thus is determined to be defective.

[Effect of the Invention] The automatic recognition and evaluation method of the bit error rate of the optical transmission characteristic according to the present invention employs a method in which a worker adjusts an optical attenuator while watching the display of a bit error rate measuring device, and adjusts the bit error rate at each optical power value. Is plotted on a standard error rate measurement paper, and a straight line is drawn intuitively on the measured data, and instead of judging the quality of the product characteristics,
The microcomputer automatically adjusts the optical attenuator, plots the measured data of the bit error rate on the standard error rate paper with high accuracy, automatically recognizes the linearity of the measured data, and calculates the evaluation function of the quality level. By quantifying the degree of dispersion of measured data and the degree of data bending with respect to the ideal error rate straight line, it is possible to quantify product quality evaluations that humans have perceived sensibly. There is an effect that data accumulation of product quality can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention, FIGS. 2 (a), (b) and (c) are flowcharts of the inspection operation of the present invention, FIG. 3 is a schematic diagram of an error rate straight line, FIG. 4 is a schematic diagram of a curve group when a point group on the error rate straight line shown in FIG. 3 is Hough-transformed, and FIG. 5 shows a relationship between geometric properties of a linearity determination reference value. FIG. 6 is a diagram showing the region where the ideal error rate estimation straight line group exists when θ _a is fixed, FIG. 6 is a diagram showing the relationship between the geometrical properties of the determination reference values shown in FIG. 5, and FIGS. ρ = ρa, and _a schematic diagram of the angular displacement when the error rate line is rotated so as to pass through the two-dimensional coordinate data (L (i), Pe (i)). FIG. 9 shows ε = 4.
FIG. 10 is a diagram showing an example in which the measured error rate straight line when [DEG] and Δρ = 0.7 is determined to be good, and FIG. 10 is a diagram showing an example in which the measured error rate straight line is determined to be defective. DESCRIPTION OF SYMBOLS 1 ... microcomputer, 2 ... pulse pattern signal generator, 3 ... DUT, 4 optical attenuator, 5 ... light combiner, 6 ... interference wave generator, 7 ... frequency generation circuit, 8 ...
... E / O converter, 9 ... Optical attenuator, 10 ... Optical attenuator, 11 ...
… Optical branching device, 12… Optical sensor, 13… Optical power meter, 14… Bit error rate measuring device, a… Pattern generator setting signal, b… Clock signal, c… NRZ signal, d ......
Optical main signal, e: main signal data signal, f: interference wave signal, g: interference wave data signal, h: interference light signal, i
... interference wave standard signal, j ... optical data signal, k ... optical attenuation data signal, l ... optical power signal, m ... optical power measurement data signal, n ... optical attenuator control signal, o ... Product output signal, p ... Error rate signal.

Claims (2)

(57) [Claims] 1. A procedure for measuring an optical power value of a laser of an object to be measured and measuring an optical transmission code error rate of the object to be measured according to the optical power value, and an optical power output from an error rate measuring device. The procedure for displaying the bit error rate data according to the accuracy in the measurement graph with high accuracy, and passing the error rate data at the lowest value of the optical power to the data displayed in the graph, and based on this point, The procedure for obtaining a small straight line, the bit error rate data used for obtaining the straight line having the minimum slope by the means, and the arrangement of the remaining measurement data not on the straight line excluding the reference point are arranged by the means. A procedure for automatically determining whether the linearity is obtained based on the obtained straight line, and a method for determining whether the bit error rate of the optical transmission characteristic is good or bad. Means for measuring an optical power value of a laser of an object to be measured and an optical transmission code error rate of the object to be measured according to the optical power value; and an optical power output from an error rate measuring device. Means for accurately displaying the bit error rate data according to the measurement graph on the measurement graph, and passing the error rate data at the lowest value of the optical power with respect to the data displayed on the graph. Means for finding a small straight line, the bit error rate data used to find the straight line with the smallest slope in the means, and the arrangement of the remaining measurement data not on the straight line excluding the reference point, Based on the obtained straight line, means for automatically evaluating whether there is linearity, and expresses the distance from the origin of the error rate graph and the straight line obtained by the means explicitly, and this distance, Is the evaluation value of the linearity of the error rate data? A means for automatically recognizing the quality of the optical transmission characteristic of the device under test.