Optical power segment fitting reporting deviation evaluation and checkpoint selection method

Abstract

The present invention discloses an optical power segmented fitting reporting deviation evaluation and verification point selection method, comprising a curve CD distributed in a triangular area surrounded by a fitting straight line of the segment and fitting straight lines of two adjacent segments; the area determines the maximum deviation value range between each segment fitting and the real power reporting; for the most edge segment DE, since there is only one adjacent segment CD, the deviation is distributed in a triangular DEH area formed by the segment fitting straight line DE, the adjacent segment fitting straight line extension line CD, and the horizontal straight line EH through the edge verification point; for the non-edge segment CD, there are two adjacent segments BC and DE; the deviation is distributed in the segment fitting straight line CD and the triangular CDG area surrounded by the two adjacent segment fitting straight line extension lines DE and BC; the method evaluates whether there is a risk of RxDDMI reporting deviation exceeding a specified threshold in each reporting segment, reduces unnecessary verification, accurately selects verification points, and improves verification reliability.

Description

[0001] This application is a divisional application of the application filed on November 20, 2023, with application number 202311541261X and invention name “Optical power segmented fitting reporting deviation evaluation and verification point selection method”. Technical Field

[0002] The invention relates to the technical field of optical power monitoring circuits, and in particular to an optical power segmented fitting reporting deviation evaluation and check point selection method. Background Art

[0003] In many optical service-related devices (including but not limited to optical modules, optical modems BOB, etc.), optical power monitoring circuits are designed to monitor the power of input optical signals so as to detect changes in the input optical signals in a timely manner.

[0004] Reference Figure 1-3 , Figure 1 This is a typical optical power monitoring circuit block diagram. Figure 2 is the IP curve of InGaAs / I nP pin photodetector; Figure 3 To fit the power response curve and the real power response curve;

[0005] The optical power monitoring circuit generally includes a power detector, an amplifier and an ADC (analog-to-digital converter): Its working principle is as follows:

[0006] When the input optical signal enters the power detector (PD), it will generate a current signal that is positively correlated with the input power. After passing through the amplification circuit, it will be converted into a voltage signal, and its signal amplitude is positively correlated with the input power. Finally, the voltage signal is sampled by the analog-to-digital converter (ADC) and converted into the numerical value of the digital signal. In addition to the above hardware circuit, in order to correctly report the optical power, it is also necessary to know the response relationship between the Rx power and the AD value. After the power detection circuit detects the intensity AD value corresponding to the input power, it is necessary to calculate the corresponding RxDDM reporting power according to the power response relationship. This is the working principle and working process of the power monitoring circuit;

[0007] In the production process, two steps are required, namely calibration and verification.

[0008] The RxDDM I of the power monitoring circuit needs to go through two steps in the production process, namely calibration and verification. Calibration is to generate the response relationship between the AD of the RxDDM I and the input optical power. Verification is to check whether the response relationship generated by calibration is consistent with the actual situation.

[0009] To calibrate RxDDM I, you need to set several input lights with different powers, and then query the corresponding RxDDM I AD value reporting. The power reporting relationship of RxDDM I can be fitted through several sets of optical power values ​​and corresponding AD values.

[0010] The RxDDM I power response curve is approximately linear in the main power response range, but when the power response range is large, there may be obvious deviations in the intervals with lower or higher power reports;

[0011] In the RxDDM I power reporting range, several different calibration points are selected (for example, PA1-PA5 as shown in the figure). The RxDDM I power response curve is divided into several segments through these calibration points. Each small segment is considered to be an approximate straight line. According to the real power and AD value of two adjacent calibration points, the power response relationship of the segment can be fitted.

[0012] On the one hand, the denser the segments and the smaller the intervals, the closer the power response of each segment is to a linear relationship. The broken lines connecting the segments are closer to the real power response curve.

[0013] On the other hand, the denser the segments and the smaller the intervals, the higher the time cost of calibration.

[0014] In actual work, designers will give appropriate calibration power points based on empirical data of power response, while ensuring the accuracy of power reporting, reducing the number of calibration points as much as possible to improve work efficiency.

[0015] When the RxDDM I power is reported, the reported power of the point can be calculated according to the AD value and the power response relationship of each segment calibrated in the calibration link.

[0016] The RxDDM I power response is not a strict straight line. The segmented fitting can only approximate the relationship of the power response curve. There will definitely be a certain deviation between the straight line of each segmented fitting and the actual power response, and this deviation needs to be verified.

[0017] Some common verification schemes currently:

[0018] If the power point is repeatedly calibrated to verify the reported deviation, this solution cannot verify the deviation between the fitted straight line and the actual power response.

[0019] Select some check points in the middle of each segment and query the deviation between the corresponding reported power and the actual power.

[0020] For ONT BOB products, the RxDDM I verification time may account for about 10% of the total BOB commissioning time.

[0021] The main problem with current checkpoint selection is that incorrect checkpoint selection wastes testing time and may not accurately intercept risks. Summary of the invention

[0022] The purpose of the present invention is to provide a method for evaluating the deviation of optical power segmented fitting reporting and selecting a checkpoint to solve the following technical problems: The main problem with the current checkpoint selection is that incorrect checkpoint selection wastes test time and may not accurately intercept risks.

[0023] The purpose of the present invention can be achieved through the following technical solutions:

[0024] The method for selecting an evaluation point for optical power segmented fitting reporting deviation includes the following steps:

[0025] Any curve CD in the non-main linear region is located at the upper left of the straight line CD, at the lower right of the extension line of the previous segment straight line BC, and at the lower left of the extension line of the next segment straight line DE; the curve CD is distributed in the triangular area surrounded by the fitting straight line of this segment and the fitting straight lines of the two adjacent segments; the area determines the maximum deviation value range between the fitting of each segment and the actual power report;

[0026] For the segment DE at the edge, since there is only one adjacent segment CD, the deviation is distributed in the triangle DEH area formed by the segment fitting straight line DE, the adjacent segment fitting straight line extension line CD, and the horizontal straight line EH passing through the edge check point;

[0027] For the non-edge segment CD, there are two adjacent segments BC and DE; the deviation is distributed in the triangular CDG area enclosed by the segment fitting straight line CD and the extension lines DE and BC of the two adjacent segment fitting straight lines.

[0028] As a further solution of the present invention: the main linear region report is very close to linear; no verification is performed for this interval.

[0029] As a further solution of the present invention: for the constructed triangle DEH area, if the deviation is less than the threshold, no actual verification is required; if it exceeds the threshold, verification should be performed.

[0030] As a further solution of the present invention: for the enclosed triangular CDG area, if the deviation is less than the threshold, the deviation between the segmented fitting straight line and the true power will not exceed the threshold, otherwise there is a risk of reporting that the deviation exceeds the threshold, and it is recommended to perform a check.

[0031] The method for selecting an evaluation point for optical power segmented fitting reporting deviation includes the following steps:

[0032] Select the midpoints E and F of the two fitting lines AB and AC. The AD values ​​of E and F corresponding to their respective fitting lines are the same. The actual reported power corresponding to this AD is between E and F. Take the average value D of E and F as the verification point. Set the Rx actual power to D and query the RxDDMI report. Determine whether the deviation between the reported value at point D and the actual power is less than the threshold.

[0033] After obtaining the three ADB checkpoints, a second-order reporting curve can be fitted using the least squares method or other methods;

[0034] Calculate the point G of maximum deviation between the fitted curve and straight line AB.

[0035] As a further solution of the present invention: if the threshold is exceeded, it will fail directly; if <deviation threshold value*tightening coefficient, no further verification is required.

[0036] As a further solution of the present invention: if it is close to the threshold value, further verification is recommended.

[0037] As a further solution of the present invention: the second-order reporting curve is y=ax 2 +bx+c.

[0038] As a further solution of the present invention: the straight line curve is y=kx+h.

[0039] As a further solution of the present invention: if the calculated deviation < the deviation threshold value * the tightening coefficient, then the actual power check is not performed;

[0040] Otherwise, it is necessary to check the actual power reporting deviation of point G.

[0041] Beneficial effects of the present invention:

[0042] According to the characteristics of the power response curve, for RxDDMI using segmented fitting, calibration is not required for intervals with good linearity; the reporting of high-power and low-power intervals may deviate significantly from linearity, but should be a continuous monotonic curve. These segments can predict the distribution of reporting deviations based on the adjacent segment fitting straight line and the segment fitting straight line; if there is no risk of exceeding the threshold, no calibration is required; if there is a risk of exceeding the threshold, calibration is recommended; it is not recommended to select the original calibration power point for the calibration power point of each segment, but a point near the midpoint of the segment should be selected for strict calibration. If the midpoint strict calibration fails, the least squares method can be used to fit the curve for a strict analysis of the theoretical deviation. If there is a risk, the maximum deviation point of the fitting curve is taken for actual power calibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below in conjunction with the accompanying drawings.

[0044] Figure 1 This is a block diagram of a typical optical power monitoring circuit structure;

[0045] Figure 2 It is the IP curve of InGaAs / I nP pin photodetector;

[0046] Figure 3 It is the fitted power response curve and the true power response curve;

[0047] Figure 4 is a distribution diagram of reported deviations of optical power segmented fitting according to the present invention;

[0048] Figure 5 is a curve graph evaluated in the optical power segment fitting process of the present invention;

[0049] Figure 6 It is a curve diagram verified in the optical power segmented fitting process of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Embodiment 1

[0052] The present invention is a method for evaluating the reporting deviation and selecting the calibration points of optical power segmented fitting. The distribution range of the reporting deviation is analyzed for the RxDDMI using segmented fitting, and whether calibration is required is determined according to the analysis results. Calibration is not required in risk-free areas, and calibration is performed in risky areas.

[0053] Figure 4 is a distribution diagram of the reported deviation of the optical power segment fitting of the present invention, wherein Figure 4 The chart shows that the horizontal axis is the Rx optical power value, and the vertical axis is the AD value corresponding to the optical power;

[0054] PA1-PA5 are calibration power points; when the input power is PA1, the AD value corresponding to the power monitoring circuit is AD1; when the input power is PA2, the AD value corresponding to the power monitoring circuit is AD2; and so on, multiple calibration points (PA1, AD1), (PA2, AD2), (PA3, AD3), (PA4, AD4), (PA5, AD5) can be obtained;

[0055] Adjacent calibration points are connected into line segments, and 4 line segments of 5 calibration points form a broken line, which is the power-AD fitting response curve of the power response circuit;

[0056] PB1 / PB3 / PB4 are the theoretical maximum power deviation points evaluated based on the reported deviation distribution of optical power segment fitting. Segment 2 in the figure is considered to have good linearity and does not need to be verified, so it is not analyzed.

[0057] Features of the RxDDMI power response curve: The main range of power response is approximately linear, with low risk and no calibration required; the high-power and low-power response ranges can be regarded as continuous monotonic curves, with a gradually increasing slope in the low-power region and a gradually decreasing slope in the high-power region;

[0058] According to the above characteristics of the power response curve, after determining the power response relationship of each segment, the power reporting deviation range of this segment can be evaluated according to the response relationship of this segment and the response relationship of the adjacent channel, so as to determine whether there is a risk of the reporting deviation exceeding the threshold;

[0059] Figure 5 is a curve diagram of the optical power segment fitting of the present invention, such as Figure 5 As shown, the method comprises the following steps:

[0060] S1. The report of the main linear area is very close to linear (as shown in the figure A->B); no verification is required for this interval;

[0061] S2. Any curve CD in the non-main linear region is located at the upper left of the straight line CD, and at the lower right of the extension line of the previous segment straight line BC, and at the lower left of the extension line of the next segment straight line DE; the curve CD is distributed in the triangular area surrounded by the fitting straight line of this segment and the fitting straight lines of the two adjacent segments; this area also determines the maximum deviation value range between the fitting of each segment and the actual power report;

[0062] S3. For the most edge segment (DE), since there is only one adjacent segment (CD), the deviation is distributed in the triangle (as shown in Figure DEH) formed by the segment fitting straight line (DE), the adjacent segment fitting straight line extension line (CD), and the horizontal straight line (EH) through the edge check point; the theoretical maximum deviation is the EH deviation; if the deviation is less than the threshold, no actual check is required; if it exceeds the threshold, a check should be performed;

[0063] S4. For non-edge segments (CD), there are two adjacent segments (BC and DE); the deviation is distributed in the triangular CDG area surrounded by the fitting straight line of this segment (CD) and the extension lines of the two adjacent fitting straight lines (DE and BC); the theoretical maximum deviation value point is the deviation from the intersection point G of the extension lines of the two adjacent fitting straight lines to the intersection point I of the horizontal line through G and the fitting straight line of this segment; if this deviation is less than the threshold, the deviation between the fitting straight line of this segment and the true power will not exceed the threshold, otherwise there is a risk of reporting the deviation exceeding the threshold, and it is recommended to perform a verification.

[0064] Embodiment 2

[0065] Figure 6 It is a curve diagram of the verification in the optical power segment fitting process of the present invention. Based on the above embodiment 1, for the segment with the risk of reporting deviation exceeding the threshold, how to further select the verification working point, the specific process is as follows:

[0066] S1. Select the midpoints E and F of the two fitting lines AB and AC. The AD values ​​of the fitting lines corresponding to E and F are the same. The actual reported power corresponding to this AD is between E and F. The average value D of E and F is taken as the verification point. Set the Rx actual power to D and query the RxDDMI report. Determine whether the deviation between the reported value at point D and the actual power is less than the threshold. If it exceeds the threshold, it will fail directly. If it is far from the threshold (such as <deviation threshold value*strictness coefficient 1), no further verification is required. If it is close to the threshold, further verification is recommended.

[0067] The above judgment process appropriately tightens the reporting deviation threshold to improve reliability;

[0068] S2. After the above steps, three ADB check points can be obtained, and a second-order reporting curve can be fitted by the least square method or other methods; for example: y = ax 2 +bx+c.

[0069] S3. Calculate the fitting curve y=ax 2 +bx+c and the line AB y=kx+h, the point of maximum deviation G;

[0070] S4. If the calculated deviation is small (<deviation threshold value*strictness factor 2), actual power verification is not performed;

[0071] S5. Otherwise, it is necessary to perform actual power reporting deviation verification on point G.

[0072] Working principle of the present invention: Features of RxDDMI power response curve: The main interval of power response is approximately linear, with low risk and no calibration required; the high power and low power response intervals can be regarded as continuous monotonic curves, with a gradually increasing slope in the low power area and a gradually decreasing slope in the high power interval. The following is an example of the power response curve in the high power interval:

[0073] The power response linear range takes at least two calibration points, and these segments are very close to linear.

[0074] No verification is required for these segments;

[0075] The high power and low power intervals may deviate significantly from the linear interval. According to the characteristics of continuous monotonic curves, each segmented true power response curve is sandwiched between the fitting straight line of this segment and the fitting of the adjacent segment; based on this characteristic, the deviation range of the true power curve and the segmented fitting straight line can be analyzed.

Claims

1. A method for selecting evaluation points for optical power segmented fitting reporting deviation, characterized in that: The following steps are involved: A and B are two adjacent test points, straight line AB is the fitting straight line between test points A and B, point C is the intersection of the adjacent segmented fitting straight lines of line segment AB, select the midpoints F and E of the two fitting straight lines AB and AC, and the AD values ​​of E and F corresponding to their respective fitting straight lines are the same; the actual reported power corresponding to this AD value is between E and F, and the average value D of E and F is taken as the verification point; set the Rx real power to the Rx optical power at point D, query the RxDDMI report; determine whether the deviation between the reported value at point D and the Rx real power is less than the threshold; Get three ADB checkpoints and fit a second-order reporting curve using the least squares method; Calculate the point G of maximum deviation between the fitted curve and straight line AB.

2. The optical power segmented fitting reporting deviation evaluation point selection method according to claim 1 is characterized in that: If the threshold is exceeded, it will fail directly. If it is <threshold*tightening coefficient, no further verification is required.

3. The optical power segmented fitting reporting deviation evaluation point selection method according to claim 2 is characterized in that: If it is close to the threshold, further verification is performed.

4. The optical power segmented fitting reporting deviation evaluation point selection method according to claim 3 is characterized in that: The second-order reporting curve is y=ax 2 +bx+c.

5. The optical power segmented fitting reporting deviation evaluation point selection method according to claim 4 is characterized in that: The straight line curve is y=kx+h.

6. The optical power segmented fitting reporting deviation evaluation point selection method according to claim 5 is characterized in that: If the calculated deviation is less than the threshold * tightening factor, the actual power calibration will not be performed; Otherwise, it is necessary to check the actual power reporting deviation of point G.

Images