JP2005110211A - Transmission quality measuring apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To present a transmission quality measuring apparatus and method for measuring a Q factor with high accuracy based on asynchronous code judgement. <P>SOLUTION: The stochastic distribution of input signals is measured by a photoelectric transducer 10, a binary identification circuit 12, a threshold sweep circuit 14 and an averaging circuit 16, and its result is stored in a memory 18. A differentiation circuit 20 differentiates the stochastic distribution with a threshold to calculate stochastic density distribution. A data extraction circuit 21 extracts data in a portion corresponding to the outside of an I waveform from the stochastic density distribution. An average value/standard deviation calculation circuit 22a calculates a space side average μ<SB>0</SB>and a standard deviation σ<SB>0</SB>from an output of the circuit 21, and an average value/standard deviation calculation circuit 22b calculates a mark side average value μ<SB>1</SB>and a standard deviation σ<SB>1</SB>from the output of the circuit 21. A pseudo Q factor calculation circuit 24 calculates ¾μ<SB>0</SB>-μ<SB>1</SB>¾/(σ<SB>0</SB>+σ<SB>1</SB>). <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a transmission quality measuring apparatus and method, and more specifically to a method and apparatus for easily measuring transmission quality in an optical transmission system, for example, a pseudo Q value.

  As a method of measuring the Q value of a transmission line in an optical transmission system, conventionally, the probability density distribution (PDF: Probability distribution function) of received signal amplitude is calculated by asynchronous code determination, and the Q value is estimated from the probability density distribution. Have been proposed (see, for example, Patent Document 1, Patent Document 2, and Non-Patent Document 1).

  FIG. 4 shows a schematic block diagram of a conventional Q value measuring apparatus. The photoelectric converter 110 converts an optical signal input from the optical transmission path into an electrical signal. The electrical signal output from the photoelectric converter 110 has a waveform substantially corresponding to the waveform of the optical signal input from the optical transmission line. The data identification circuit 112 binarizes the output signal of the photoelectric conversion circuit 110 according to the threshold voltage Vth from the threshold sweep circuit 114. The data identification circuit 112 is composed of, for example, a D flip-flop. The threshold sweep circuit 114 changes the threshold voltage Vth applied to the data identification circuit 112 stepwise between the voltages V0 and V1 every 100 ms within a fixed time T as shown in FIG. The voltage V0 is a space voltage, and the voltage V1 is a mark voltage.

  The averaging circuit 116 averages the output of the binary identification circuit 112 at the same threshold voltage Vth and writes the average value in the memory 118. FIG. 6 shows an example of the distribution of the average value written in the memory 118 with respect to the threshold value. The horizontal axis in FIG. 6 indicates the threshold voltage Vth, and the horizontal axis indicates the average value output from the averaging circuit 116. FIG. 6 shows a probability distribution with respect to the threshold value Vth of an optical signal (more specifically, an output signal of the photoelectric converter 110) input from the optical transmission line.

  The differentiating circuit 120 differentiates the probability distribution stored in the memory 118 with the threshold voltage Vth. FIG. 7 shows an example of the differential waveform, that is, an example of probability density distribution. The horizontal axis of FIG. 7 indicates the threshold voltage Vth, and the horizontal axis indicates the differential value with respect to Vth, that is, the probability density. The probability density distribution calculated in this way includes two peaks, a peak indicating the power distribution on the space side and a peak indicating the power distribution on the mark side, as clearly shown in FIG.

The average value / standard deviation calculation circuits 122a and 122b fit a Gaussian function to each of these two peaks, and calculate the average values μ ₀ and μ ₁ and standard deviations σ ₀ and σ ₁ of the peaks. The pseudo Q value calculation circuit 124 calculates Q = | μ ₀ −μ ₁ | / (σ ₀ + σ ₁ ) from the outputs μ ₀ , μ ₁ , σ ₀ , and σ ₁ of the average value / standard deviation calculation circuits 122 a and 122 b.
According to the above, the Q value is calculated. The calculation of the Q value by the above equation is described in Patent Literature 3 and Patent Literature 4.

The Q value obtained by the pseudo Q value calculation circuit 124 is calculated based on asynchronous code determination. Originally, the Q value of the optical transmission line is calculated from the BER (bit error rate) of the transmission signal. Therefore, for comparison, it is necessary to convert the Q value obtained by the pseudo Q value calculation circuit 124 into a Q value based on the BER. For conversion, the Q value based on the BER is also measured in advance in the same optical signal transmission state on the same optical transmission line, and based on the Q value (pseudo Q value) obtained by the pseudo Q value calculation circuit 124 and the BER. Obtain a correlation with the Q value. The correction circuit 126 corrects or converts the calculation result of the pseudo Q value calculation circuit 124 into a Q value based on the BER by applying a correlation obtained in advance to the calculation result of the pseudo Q value calculation circuit 124. Thereby, the Q value corresponding to the Q value based on the BER can be obtained.
JP 2000-358015 A UK Patent Application Publication No. 2352597 JP-A-7-154378 US Pat. No. 5,585,954 Mitsuru Sugawara and Kazuhiko Ide, “Proposal of Optical Signal Quality Monitoring Method by Probability Distribution Measurement”, 1999 IEICE Communication Society Conference, B-10-73, p. 250

  In the conventional asynchronous code determination, as shown in FIG. 4, the inner part of the eye waveform of the probability density distribution has a distribution shape deviating from the Gaussian distribution due to the influence of the rise time and fall time of the pulse of the signal light. . In particular, the inner part of the mountain corresponding to the space side deviates greatly from the Gaussian distribution.

  In the conventional example, since a Gaussian function is forcibly applied to a distribution deviating from the Gaussian distribution, high accuracy cannot be obtained.

  An object of the present invention is to provide a transmission quality measuring apparatus and method capable of measuring a Q value with higher accuracy than before by asynchronous code determination.

  The transmission quality measuring device according to the present invention is a device for measuring the transmission quality of an optical signal carrying a code having a space and a mark, the photoelectric converter for converting the optical signal into an electric signal, and the photoelectric converter A cumulative distribution calculating circuit that discriminates binary of the output electric signal by a threshold value that is discretely swept, calculates a cumulative distribution with respect to the threshold value, and a differentiating circuit that calculates a probability density distribution by differentiating the cumulative distribution with the threshold value A data extraction circuit for extracting data corresponding to the outside of the eye waveform from the probability density distribution, and fitting a Gaussian function to the data extracted by the data extraction circuit to obtain an average value and a standard of the space in the probability density distribution Deviation, and average / standard deviation calculation circuit for calculating the average value and standard deviation of the mark, and the average value and standard of the space and the mark Characterized by comprising the index calculation circuit for calculating an indication of the transmission quality of the optical signal from the difference.

  The photoelectric converter and / or the cumulative distribution calculation circuit may be composed of a responsive element or circuit that is lower than the signal frequency of the optical signal.

  The transmission quality measuring method according to the present invention is a method for measuring the transmission quality of an optical signal carrying a code having a space and a mark, the photoelectric conversion step for converting the optical signal into an electric signal, and discrete sweeping A cumulative distribution calculating step of calculating a cumulative distribution with respect to the threshold value by binary identification of the electric signal by the threshold value, a probability density distribution calculating step of differentiating the cumulative distribution by the threshold value and calculating a probability density distribution; An average for calculating the average value and standard deviation of the space and the average value and standard deviation of the mark in the probability density distribution by fitting a Gaussian function to data corresponding to the outside of the eye waveform of the probability density distribution. The transmission quality index of the optical signal is calculated from the value / standard deviation calculation step and the average value and standard deviation of the space and the mark. Characterized by comprising a transmission quality index calculation step that.

  The response speed of the photoelectric conversion step and / or the cumulative distribution calculation step may be lower than the signal frequency of the optical signal.

  According to the present invention, the portion corresponding to the inside of the eye waveform of the probability density distribution, which tends to be out of the shape of the Gaussian distribution, is excluded in advance, and the Gaussian function is fitted. The average value and standard deviation of the space, and the average value and standard deviation of the mark can be calculated. As a result, it is possible to calculate a higher accuracy and transmission quality index than the conventional one, for example, a pseudo Q value.

  Further, when the photoelectric converter and / or the cumulative distribution calculation is composed of a responsive element or circuit lower than the signal frequency of the optical signal, or the response speed of the photoelectric conversion step and / or the cumulative distribution calculation step is that of the optical signal. When the frequency is lower than the signal frequency, a cheaper circuit element can be used, and the transmission quality index can be monitored with a device cheaper than the conventional one.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 shows a schematic block diagram of an embodiment of the present invention. The photoelectric converter 10 converts an optical signal input from the optical transmission path into an electrical signal. The electric signal output from the photoelectric converter 10 has a waveform substantially corresponding to the waveform of the optical signal input from the optical transmission line. The data identification circuit 12 identifies the output signal of the photoelectric conversion circuit 10 in binary according to the threshold voltage Vth from the threshold sweep circuit 14. The data identification circuit 12 is composed of, for example, a D flip-flop. The threshold sweep circuit 14 changes the threshold voltage Vth applied to the data identification circuit 12 stepwise between the voltages V0 and V1 every 100 ms within a fixed time T as shown in FIG. The voltage V0 is a space voltage, and the voltage V1 is a mark voltage. The averaging circuit 16 averages the output of the binary identification circuit 12 at the same threshold voltage Vth and writes the average value in the memory 18. The differentiating circuit 20 differentiates the probability distribution stored in the memory 18 with the threshold voltage Vth. The operation so far is exactly the same as the conventional example shown in FIG.

  The data extraction circuit 21 extracts only data corresponding to the outside of the eye waveform from the probability density distribution obtained by the differentiation circuit 20. Of course, the data extraction circuit 21 may remove data corresponding to the inside of the eye waveform from the probability density distribution obtained by the differentiation circuit 20. The inside of the peaks of the two peaks in the probability density distribution may be regarded as the inside of the eye waveform. If the number of sample points after extraction by the data extraction circuit 21 is sufficiently large, the data of two peak peaks in the probability density distribution may be removed. If the number of sample points after extraction by the data extraction circuit 21 is small, the probability Leave the data for the peaks of the two peaks in the density distribution.

Each of the average value / standard deviation calculation circuits 22a and 22b fits a Gaussian function to the distribution corresponding to the outside of the eye waveform with respect to the probability density distribution data extracted by the data extraction circuit 21, and averages the respective peaks. Values μ ₀ and μ ₁ and standard deviations σ ₀ and σ ₁ are calculated. FIG. 2 illustrates a range in which the average value / standard deviation calculation circuits 22a and 22b fit a Gaussian function on the probability density distribution by the differentiation circuit 20.

As pointed out earlier, asynchronous code determination is affected by the rise time and fall time of the pulse of signal light, and the inner part of the eye waveform of the probability density distribution has a distribution shape that deviates from the Gaussian distribution. In the embodiment, since the data extraction circuit 21 removes the data of the inner part of the eye waveform of the probability density distribution, the Gaussian function is fitted to each mountain, so that the average values on the space side and the mark side are more accurate than before. μ ₀ and μ ₁ and standard deviations σ ₀ and σ ₁ can be calculated.

Similar to the pseudo Q value calculation circuit 124, the pseudo Q value calculation circuit 24 uses the outputs μ ₀ , μ ₁ , σ ₀ , and σ ₁ of the average value / standard deviation calculation circuits 22a and 22b as Q = | μ ₀ −μ _1. | / (Σ ₀ + σ ₁ )
According to the above, the Q value is calculated. Since the original Q value is calculated from the BER (bit error rate) of the transmission signal, in this embodiment as well, for comparison, the Q value obtained by the pseudo Q value calculation circuit 24 is used as the Q value based on the BER. It is necessary to convert to For conversion, a Q value based on the BER is also measured in advance in the same optical signal transmission state on the same optical transmission line, and based on the Q value (pseudo Q value) obtained by the pseudo Q value calculation circuit 24 and the BER. Obtain a correlation with the Q value. The correction circuit 26 corrects or converts the calculation result of the pseudo Q value calculation circuit 24 into a Q value based on the BER by applying a correlation obtained in advance to the calculation result of the pseudo Q value calculation circuit 24. Thereby, the Q value corresponding to the Q value based on the BER can be obtained.

  FIG. 3 shows a comparative example of the pseudo Q value calculated by the pseudo Q value calculation circuit 24 and the Q value by BER measurement. The horizontal axis indicates the Q value obtained by BER measurement, and the vertical axis indicates the Q value calculated by the pseudo Q value calculation circuit 24. It can be seen that there is a good correlation between the two. The Q value calculated by the pseudo Q value calculation circuit 24 can be used as an index of transmission quality of an optical signal, specifically, SNR (signal to noise ratio). When it is not necessary to convert the Q value due to the BER measurement, the correction circuit 26 is unnecessary.

  The part from the memory 18 to the correction circuit 26 is actually realized by a personal computer, and the part from the differentiation circuit 20 to the correction circuit 26 is implemented by software. Of course, part or all of the part from the differentiation circuit 20 to the correction circuit 26 can be realized by hardware or firmware.

  The O / E converter 10, the binary identification circuit 12, and the averaging circuit 16 may not be able to follow the signal rate of the optical signal input from the optical transmission line. For example, as shown in FIG. 8, an electrical low-pass filter 30 that removes high-frequency components may be disposed between the O / E converter 10 and the binary identification circuit 12. For example, when the signal rate of the input signal light is 10 Gbps, the high-frequency cutoff frequency may be lower than the frequency corresponding to the signal rate, such as 8 GHz, 5 GHz, or 2 GHz, as the electrical low-pass filter 30. This is because even if the high frequency is cut off in this way, the probability density of the peak level and the space level of the optical pulse can be detected by allowing a considerable number of marks and spaces of the input signal light to continue.

  FIG. 9 shows the Q measurement results when the high-frequency cutoff frequency of the electrical filter 30 is set to 7.5 GHz, 4.8 GHz, and 1.866 GHz. The horizontal axis shows the Q value calculated by BER measurement, and the vertical axis shows the Q value measured by this example. When the Q value is large, the deviation from the Q value calculated by the BER measurement becomes large. However, in the application for monitoring the transmission path, it is only necessary to accurately detect the situation where the Q value is low. From FIG. 9, it can be seen that sufficient accuracy can be ensured even in the configuration in which the Q value is monitored by an electrical signal with the high frequency cut off as shown in FIG.

  The fact that the high range of the electrical signal output from the O / E converter 10 may be cut off in this way means that a low-speed photodiode that cannot follow the input optical signal is used as the O / E converter 10. This means that a low-speed circuit and a device can be used as the binary identification circuit 12 and subsequent circuits. As a result, the optical transmission line can be monitored with a cheaper device configuration.

It is a schematic block diagram of one Example of this invention. It is a figure which shows the fitting range of the Gaussian function by a present Example. It is a comparison figure of pseudo | simulation Q value obtained in a present Example, and Q value computed by BER measurement. It is a schematic block diagram of a conventional Q value measuring apparatus. It is a schematic diagram which shows the example of a change of a threshold value. It is a schematic diagram of the cumulative distribution in asynchronous code determination. It is a schematic diagram of probability density distribution in asynchronous code determination. It is a schematic block diagram of a modified embodiment of the present invention. It is a figure which shows the contrast of the measurement result of 2nd Example, and Q value computed by BER measurement.

Explanation of symbols

10: photoelectric converter 12: data identification circuit 14: threshold sweep circuit 16: averaging circuit 18: memory 20: differentiation circuit 21: data extraction circuits 22a and 122b: average value / standard deviation calculation circuit 24: pseudo Q value calculation circuit 26: correction circuit 30: electrical low-pass filter 110: photoelectric converter 112: data identification circuit 114: threshold sweep circuit 116: averaging circuit 118: memory 120: differentiation circuits 122a and 122b: average value / standard deviation calculation circuit 124: pseudo Q value calculation circuit 126: correction circuit

Claims (9)

An apparatus for measuring transmission quality of an optical signal carrying a code having a space and a mark,
A photoelectric converter (10, 30) for converting the optical signal into an electrical signal;
A cumulative distribution calculating circuit (12, 14, 16, 18) for performing binary discrimination on the output electrical signal of the photoelectric converter (10, 30) by a threshold value that is discretely swept and calculating a cumulative distribution with respect to the threshold value;
A differentiating circuit (20) for differentiating the cumulative distribution by the threshold and calculating a probability density distribution;
A data extraction circuit (21) for extracting data corresponding to the outside of the eye waveform from the probability density distribution;
Mean value / standard deviation for fitting the Gaussian function to the data extracted by the data extraction circuit (21) and calculating the average value and standard deviation of the space in the probability density distribution, and the average value and standard deviation of the mark A calculation circuit (22a, 22b);
An index calculation circuit (24) for calculating an index of transmission quality of the optical signal from the average value and standard deviation of the space and the mark
And a transmission quality measuring device.   The photoelectric converter includes a light receiving element (10) that receives the optical signal, and a filter (30) that removes part of the signal frequency component of the optical signal from the output of the light receiving element. The transmission quality measuring device according to claim 1.   The transmission quality measuring apparatus according to claim 1, wherein the photoelectric converter includes a photoelectric conversion element having a response lower than a signal frequency of the optical signal.   The transmission quality measuring apparatus according to any one of claims 1 to 3, wherein the cumulative distribution calculating circuit includes a circuit having a response lower than a signal frequency of the optical signal.   The cumulative distribution calculation circuit includes a binary identification circuit (12, 14) for binary-identifying an output signal of the photoelectric converter (10, 30) by a threshold value that is discretely swept, and an output of the binary identification circuit. 5. An averaging circuit (16, 18) for calculating a cumulative distribution of discriminator circuit output values with respect to a threshold value by averaging Transmission quality measuring device. A method for measuring transmission quality of an optical signal carrying a code having a space and a mark,
A photoelectric conversion step for converting the optical signal into an electrical signal;
A cumulative distribution calculating step of calculating a cumulative distribution with respect to the threshold by discriminating the electric signal in binary using a threshold that is discretely swept;
A probability density distribution calculating step of calculating the probability density distribution by differentiating the cumulative distribution by the threshold value;
Fitting a Gaussian function to the data corresponding to the outside of the eye waveform of the probability density distribution to calculate the average value and standard deviation of the space and the average value and standard deviation of the mark in the probability density distribution・ Standard deviation calculation step;
A transmission quality measuring method comprising: a transmission quality index calculating step for calculating an index of transmission quality of the optical signal from the space and the average value and standard deviation of the mark.   The photoelectric conversion step includes a step of receiving the optical signal by the light receiver (10), and a high frequency removing step of removing the signal frequency component of the optical signal from the output of the light receiver (10). The transmission quality measuring method according to claim 6, wherein:   The transmission quality measuring method according to claim 6, wherein the photoelectric conversion step converts the optical signal into an electric signal with a response lower than the signal frequency of the optical signal.   9. The transmission quality measuring method according to claim 6, wherein the cumulative distribution calculating step identifies the electrical signal in binary with a response lower than the signal frequency of the optical signal.

Images