Method for calibrating constellation diagram data in vector synthesis type phase shifter test

Abstract

The invention discloses a method for calibrating constellation diagram data in a vector synthesis type phase shifter test. The method comprises the following steps: dividing a constellation diagram test process into k time periods; testing a pre-test calibration coordinate value of the calibration point in a time period 0, testing a constellation diagram data value in sequence from a time period 1 to a time period k, and testing a post-test calibration coordinate value of the calibration point in a time period k + 1; subtracting the corresponding calibration reference point values from the constellation diagram data values corresponding to the calibration points selected from the time period 1 to the time period k to obtain correction values of the calibration points; and performing linear interpolation on the correction values to obtain interpolation correction values, and subtracting the corresponding interpolation correction values from the constellation diagram data values from the time period 1 to the time period k to obtain calibrated constellation diagram data. According to the method, the problem that the vector synthesis type phase shifter has large phase deviation and amplitude deviation in long-time constellation diagram data testing is solved, and good support is provided for subsequent accurate point selection of the phase shifter.

Description

Technical Field

[0001] This invention relates to the field of digital information transmission, and more particularly to a method for calibrating constellation diagram data in the testing of vector synthesis phase shifters. Background Technology

[0002] Vector synthesizer phase shifters are a circuit architecture used in phase shifter modules of phased array transceiver chips. The basic principle of a vector synthesizer phase shifter is as follows: First, the input signal is decomposed into two mutually orthogonal signals, namely the I-channel and Q-channel signals. Then, the amplitudes of the I-channel and Q-channel signals are independently adjusted. Finally, they are vector-summed to produce an output signal with a certain angle. The amplitudes of the I-channel and Q-channel signals are each independently adjusted by an m-bit digital control signal, thus allowing vector synthesis of the output signal. A phase shifter requires only one state. There are several state points, therefore it is necessary to start from the vector synthesis phase shifter. Select from the states There are 64 phase shift states. To ensure good phase and amplitude performance of the phase shifter, m is usually required to be greater than or equal to n. Taking a 6-phase shifter as an example, 64 phase shift states are required, but the internal digital controller for controlling the I and Q amplitudes of a vector synthesizer phase shifter typically uses 6 or 7 bits. If the vector synthesizer phase shifter uses 6 bits for internal control, 4096 states are generated. If the vector synthesizer phase shifter uses 7 bits for internal control, 16384 states are generated.

[0003] Vector Synthesizing Phase Shifter While states can be obtained through simulation or actual testing, simulation data can only serve as preliminary performance evaluation and is insufficient to meet the needs of actual products for selecting points. Vector synthesizer phase shifters can generate a large number of states, meaning that testing the data used for point selection requires a significant amount of time. For example, if the I and Q channels of a vector synthesizer phase shifter each use 7 control bits, scanning all 16384 states using an automated program with a 3-second interval between two data frames would require approximately 13.6 hours of testing. During long-term testing, environmental factors such as temperature and the instability of the testing system itself can cause deviations in the phase and amplitude of the testing system relative to the initial time. These phase and amplitude deviations will cause the measured state of the vector synthesizer phase shifter to deviate from its true value in the constellation diagram, resulting in the selected point deviating significantly from the optimal value.

[0004] Generally, when a vector network analyzer operates at a stable temperature of 25°C ± 5°C for 60 minutes, its amplitude fluctuation at the 40GHz frequency point is within 0.2dB, and its phase fluctuation is within 5°. When the vector network analyzer is connected to a test piece, the phase and amplitude deviations of the entire test system will be greater under long-term testing. Figure 1The diagram shows how the phase and amplitude deviations of the test system relative to the initial moment change over time. It can be seen that as time increases, the phase and amplitude deviations of the test system become increasingly larger, leading to significant errors in the constellation diagram of the tested vector synthesis phase shifter.

[0005] Taking a vector synthesis phase shifter with a 7-bit controller as an example, the 7 control bits of the I and Q channels can each generate 128 control codes, and the combination of the I and Q channels can produce 16384 control states. During the testing of the vector synthesis phase shifter, the traditional control code scanning method... Figure 2 As shown, the test scans layer by layer from the IQ control code (0,0) to the IQ control code (127,127). According to... Figure 2 The traditional control code scanning method shown produces the following constellation diagram: Figure 3 As shown, the tested constellation charts deviate more and more from the true values ​​over time. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for calibrating constellation diagram data in the testing of vector synthesis phase shifters.

[0007] The objective of this invention is achieved through the following technical solution: A first aspect of the present invention provides a method for calibrating constellation diagram data in the testing of a vector synthesis phase shifter, comprising the following steps: The process of testing the constellation diagram is divided into k time periods. Each time period includes constellation points generated by the control codes of the vector synthesis phase shifter, and a data point is selected from the constellation points in each time period as a calibration point. The pre-test calibration coordinate value of the calibration point is tested at time period 0, the constellation map data value is tested sequentially from time period 1 to time period k, and the post-test calibration coordinate value of the calibration point is tested at time period k+1. Subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the pre-test calibration coordinate values ​​to obtain the pre-test relative coordinate values; subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the post-test calibration coordinate values ​​to obtain the post-test relative coordinate values. The average relative coordinate values ​​of the calibration points under the same control code are added together with the calibration coordinate values ​​before and after the test, and the average is obtained. The coordinate values ​​of the first calibration point with the calibration coordinate values ​​before the test are added to the average relative coordinate values ​​of the calibration points to obtain the calibration reference points of the constellation diagram for each time period. Subtract the corresponding calibration reference point value from the constellation diagram data value corresponding to the selected calibration point in time period 1 to time period k to obtain the correction value of the calibration point; perform linear interpolation on the correction value to obtain the interpolation correction value, and then subtract the corresponding interpolation correction value from the constellation diagram data value in time period 1 to time period k to obtain the calibrated constellation diagram data.

[0008] Furthermore, the vector synthesis phase shifter decomposes the input signal into two mutually orthogonal signals, namely the I-channel signal and the Q-channel signal, and the N-bit control bits of the I-channel and Q-channel can be respectively generated IQ control code, the control code of the I channel changes from 0 to The control code for the Q-path changes from 0 to .

[0009] Furthermore, the scanning method for the constellation chart data values ​​for time period 1 to time period k is to proceed from the outer circle to the inner circle, with each circle's four edges corresponding to four time periods.

[0010] Furthermore, the test constellation diagram data values ​​for time period 1 to time period k only scan the data of the first M layers of the outer ring, that is, the scan of IQ control code (0,0) terminates at IQ control code (M-1,M).

[0011] Furthermore, a data point is selected from the constellation points in each time period as a calibration point, specifically the first data point of the corresponding time period.

[0012] Furthermore, in the last time period, an additional last constellation point is added as the end calibration point.

[0013] The beneficial effects of this invention are: In an exemplary embodiment of the present invention, the problem of large phase and amplitude deviations in long-term test constellation diagram data of vector synthesis phase shifters is solved, providing good support for accurate point selection of subsequent phase shifters; at the same time, the proposed constellation diagram data scanning method avoids large phase and amplitude fluctuations in the data before and after, thereby making the test data more accurate. Attached Figure Description

[0014] Figure 1 This is a schematic diagram illustrating how the phase and amplitude deviation values ​​of a test system relative to the initial moment change over time in the prior art. Figure 2 This is a schematic diagram of the traditional control code scanning method during constellation diagram testing of vector synthesis phase shifters in existing technologies; Figure 3 This is a schematic diagram of the constellation diagram obtained by testing a vector synthesis phase shifter in the existing technology using the traditional control code scanning method. Figure 4A flowchart illustrating a method for calibrating constellation diagram data in testing a vector synthesis phase shifter, as provided in an exemplary embodiment of the present invention; Figure 5 This is a schematic diagram of time segmentation of constellation diagram data for a test vector synthesis phase shifter proposed in an exemplary embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the scanning method for testing constellation diagram data for time period 1 to time period k, as proposed in an exemplary embodiment of the present invention. Figure 7 This is a schematic diagram of a constellation diagram for time period 1 to time period k, as proposed in an exemplary embodiment of the present invention. Figure 8 This is a schematic diagram illustrating the scanning calibration point method for time period 0 and time period k+1 as proposed in an exemplary embodiment of the present invention; Figure 9 This is a schematic diagram of a constellation of calibration points proposed in an exemplary embodiment of the present invention; Figure 10 This is a schematic diagram illustrating an interpolation method for correcting the x-axis value of constellation diagram data according to an exemplary embodiment of the present invention; Figure 11 This is a schematic diagram illustrating an interpolation method for correcting the y-axis value of constellation diagram data according to an exemplary embodiment of the present invention; Figure 12 This is a schematic diagram illustrating the positions of calibration constellation points, test constellation points, and actual constellation points as proposed in an exemplary embodiment of the present invention. Detailed Implementation

[0015] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0016] See Figure 4 , Figure 4 A flowchart illustrating a method for calibrating constellation diagram data in vector synthesis phase shifter testing, as proposed in an exemplary embodiment of the present invention, is shown, comprising the following steps: The process of testing the constellation diagram is divided into k time periods. Each time period includes constellation points generated by the control codes of the vector synthesis phase shifter, and a data point is selected from the constellation points in each time period as a calibration point. like Figure 5As shown, the pre-test calibration coordinate value of the calibration point is tested at time period 0, the constellation map data value is tested sequentially from time period 1 to time period k, and the post-test calibration coordinate value of the calibration point is tested at time period k+1. Subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the pre-test calibration coordinate values ​​to obtain the pre-test relative coordinate values; subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the post-test calibration coordinate values ​​to obtain the post-test relative coordinate values. The average relative coordinate values ​​of the calibration points under the same control code are added together with the calibration coordinate values ​​before and after the test, and the average is obtained. The coordinate values ​​of the first calibration point with the calibration coordinate values ​​before the test are added to the average relative coordinate values ​​of the calibration points to obtain the calibration reference points of the constellation diagram for each time period. Subtract the corresponding calibration reference point value from the constellation diagram data value corresponding to the selected calibration point in time period 1 to time period k to obtain the correction value of the calibration point; perform linear interpolation on the correction value to obtain the interpolation correction value, and then subtract the corresponding interpolation correction value from the constellation diagram data value in time period 1 to time period k to obtain the calibrated constellation diagram data.

[0017] Specifically, in this exemplary embodiment, taking a vector synthesis phase shifter with N control bits as an example, the N control bits of its I and Q paths can generate This type of control code, the combination of I-channel and Q-channel can generate There are several control states. During long-term testing, the phase and amplitude deviations of the test system are large; however, during short-term testing, the phase and amplitude deviations are smaller. Therefore, long-term testing can be divided into a series of time periods, and a data point can be selected from each time period as a calibration point. Figure 4 This diagram illustrates the time-segmented data segmentation for testing the constellation diagram of a vector synthesized phase shifter according to this exemplary embodiment. First, the data tested from time segment 1 to time segment k constitutes the points in the vector synthesized phase shifter constellation diagram. The data tested from time segment 0 to time segment k+1 are selected as calibration points from time segment 1 to time segment k. The phase and amplitude deviations of the data from time segment 1, time segment 2 to time segment k can be calibrated using the calibration points measured from time segment 0 and time segment k+1. Long-term phase and amplitude deviations of the test system are calibrated using calibration points measured over short periods.

[0018] Specifically, firstly, in time period 0, calibration point data selected from time period 1 to time period k are tested, typically selecting data from the middle of the corresponding time period, to obtain the pre-test calibration coordinate values. Then, data from time period 1 to time period k are tested; these are the constellation diagram data values ​​of the uncalibrated vector synthesizer phase shifter. Next, the data from the selected calibration points in time period 1 to time period k are tested again to obtain the post-test relative coordinate values. After the tests are completed, the calibration point data from time period 0 and time period k+1 are processed.

[0019] To minimize random errors in the calibration point data, this exemplary embodiment processes time periods 0 and k+1 as follows: First, subtract the data of the first calibration point in time period 0 from the data of each calibration point in time period 0 (i.e., subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the pre-test calibration coordinate values ​​to obtain the pre-test relative coordinate values); subtract the data of the first calibration point in time period k+1 from the data of each calibration point in time period k+1 (i.e., subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the post-test calibration coordinate values ​​to obtain the post-test relative coordinate values). This yields the relative coordinate values ​​of the calibration points in time periods 0 and k+1 relative to the first calibration point. Then, add the relative coordinate values ​​of the calibration points under the same control code in time periods 0 and k+1 and take the average (i.e., add the relative coordinate values ​​of the calibration points under the same control code in the pre-test and post-test calibration coordinate values ​​and take the average), thereby obtaining the average relative coordinate value of the calibration points. Finally, the average relative coordinate values ​​of the calibration points are added to the coordinate value of the first calibration point in time period 0 (the coordinate value of the first calibration point before the test) to obtain the calibration reference points for each time period of the constellation diagram.

[0020] If calibration reference points are used to directly calibrate the data for each time period, the data at the ends of the time periods will exhibit significant deviations. Therefore, interpolation processing of the calibration correction values ​​is necessary. During calibration, the corresponding calibration reference point value is subtracted from the selected calibration point value in time periods 1 to k to obtain the correction value for the calibration point. To calibrate the phase and amplitude deviations in the constellation diagram at different time points, linear interpolation is performed on the discrete correction values ​​that change over time. Then, the data in time periods 1 to k are subtracted from the corresponding linearly interpolated correction values, thus obtaining the calibrated constellation diagram data.

[0021] As described above, the present exemplary embodiment solves the problem of large phase and amplitude deviations in long-term test constellation diagram data for vector synthesis phase shifters, providing good support for accurate point selection of the subsequent phase shifter; at the same time, the constellation diagram data scanning method proposed in this exemplary embodiment avoids large phase and amplitude fluctuations in the data before and after, thereby making the test data more accurate.

[0022] The following content will elaborate on the specific implementation methods in more detail: More preferably, in an exemplary embodiment, the vector synthesis phase shifter decomposes the input signal into two mutually orthogonal signals, namely an I-channel signal and a Q-channel signal, and the N-bit control bits of the I-channel and Q-channel can be respectively generated IQ control code, the control code of the I channel changes from 0 to The control code for the Q-path changes from 0 to More preferably, in an exemplary embodiment, the scanning method for the test constellation chart data values ​​for time period 1 to time period k is to proceed sequentially from the outer circle to the inner circle, with each circle's four edges corresponding to four time periods.

[0023] Specifically, in this exemplary embodiment, Figure 6 This exemplary embodiment illustrates the scanning method for the test constellation diagram data in time periods 1 to k. The control code for the I-channel changes from 0 to... The control code for the Q-path changes from 0 to To save time, only the M layer data of the outer ring is scanned. Figure 6 The time period in the text corresponds to: Figure 5 In the time period from 1 to k, for example, in the first time period, the control code of the Q channel is 0, and the control code of the I channel changes from 0 to... This corresponds to the data measured in time period 1; for the second segment, the control code for channel I is... The control code for the Q-path changes from 0 to This corresponds to the data measured in time period 2, and the subsequent correspondences follow the same pattern.

[0024] More preferably, in an exemplary embodiment, the test constellation diagram data values ​​for time period 1 to time period k only scan the data of the first M layers of the outer ring, that is, the scan of IQ control code (0,0) terminates at IQ control code (M-1,M).

[0025] Specifically, in this exemplary embodiment, the vector synthesis phase shifter has many states, resulting in a long testing time. To ensure performance, the vector synthesis phase shifter needs to have the highest possible gain or the lowest possible insertion loss. Therefore, low-gain points near the origin in the constellation diagram are of no help in selecting points for the phase shifter and can be skipped during testing, thus saving testing time. The constellation diagram only scans the outermost M-layer data, and the scanning direction of the control code is as follows... Figure 6 As shown, the scan proceeds gradually from the outer circle to the inner circle, from the IQ control code (0, 0) to the control code (M-1, M), with a total time interval k equal to 4*M. The constellation diagram generated from time interval 1 to time interval k is shown below. Figure 7 As shown, during the scanning process from the starting coordinate point to the ending coordinate point, the phase and amplitude changes between test points are relatively small, thus making the test more accurate.

[0026] More preferably, in an exemplary embodiment, the step of selecting a data point from the constellation points in each time period as a calibration point specifically selects the first data point of the corresponding time period as the calibration point. More preferably, in an exemplary embodiment, in the last time period, an additional last constellation point is added as the ending calibration point.

[0027] Specifically, in this exemplary embodiment, for ease of calculation, the first point at the beginning of each time period is selected as the calibration point. In the last time period, the last constellation point is additionally added as a calibration point. The scanning method of the calibration points in this invention is as follows: Figure 8 As shown, the calibration points are scanned sequentially from the outer ring inwards, with the IQ control code scanning from (0, 0) to (M-1, M), requiring a total of 4*M+1 points to be scanned. If the outer ring 11 layers of data are selected, only 45 calibration points need to be scanned. The coordinates of the calibration points in the constellation diagram are as follows: Figure 9 As shown.

[0028] The specific scanning test method in this exemplary embodiment is described using the preferred scanning method described above: Time interval 0 and time interval k+1 according to Figure 8 The scanning is performed using the calibration point scanning method, while time period 1 to time period k is scanned according to... Figure 6 The test constellation data is scanned in the manner shown. After scanning, the location of the calibration reference points on the constellation needs to be obtained. Since only the outer M layer data is scanned, there are 4*M+1 calibration points. The control codes for the four calibration points in the first layer are as follows: , , , Then the control codes for the four calibration points of the k-th layer are: , , , The coordinates of the five calibration points in the final M layer (including the coordinates of the last scan point) are: , , , And the last point The lengths of the four time periods in the first layer are: The lengths of the four time periods in the k-th layer are: .

[0029] Let the coordinate data of the test scan calibration point during time period 0 (i.e., the pre-test calibration coordinate values ​​of the calibration point) be... Where x represents the position of the calibration point on the I-axis of the constellation diagram, y represents the position of the calibration point on the Q-axis of the constellation diagram, a represents that the calibration point belongs to time period 0, and i represents the data of the i-th calibration point. Similarly, the calibration point data for time period k+1 (i.e., the calibration coordinate values ​​of the calibration point after testing) can be represented as: 'b' indicates that the calibration point belongs to time period k+1; the coordinate data of the calibration reference point can be expressed as... Therefore, the coordinates of the i-th calibration reference point can be obtained as follows, by using the deviations of the calibration point from the initial calibration point in the average time period 1 and time period k+1 to reduce the random error of the test: Discrete correction values ​​can be expressed as Where d represents the discrete correction value (i.e., the correction value of the calibration point); the test constellation data (i.e., the constellation data values ​​from time period 1 to time period k) can be represented as Where x represents the position of the calibration point on the I-axis of the constellation diagram, y represents the position of the calibration point on the Q-axis of the constellation diagram, s represents the constellation diagram data tested, i represents the i-th time period or the i-th time interval, j represents the j-th data point in the i-th time interval, and the length of the i-th time interval is represented by the letter e, and its value is... The `floor` function rounds down to the nearest integer. Therefore, the last data point in the `i`th time interval can be represented as... The value of the last time period length e is... The i-th discrete correction value can be calculated using the following formula: The last discrete correction value is calculated using the following formula: .

[0030] To calibrate each coordinate point in the constellation chart test, correction values ​​for each coordinate need to be calculated. The calculation method for the x-axis and y-axis correction values ​​of the coordinate points adopts... Figure 10 and Figure 11 Perform as shown. Figure 10Taking the calculation of the correction value of the x-axis of the constellation diagram as an example, the correction value of each constellation diagram is calculated by using linear interpolation based on the known correction values ​​of the calibration points.

[0031] The correction value for the j-th value in the i-th interval is calculated as follows: ,in The `floor` function represents rounding down to the nearest integer, i.e., the integer value of the i-th interval. The value corresponds to the x-axis coordinate of the discrete correction value. The last interval ,in .

[0032] Similarly, the value of y can be calculated. , in The last interval ,in ; The data of the calibrated constellation diagram can be represented as Let z represent the calibrated constellation data. Then, the j-th calibrated data within the i-th time period is: The effect of the calibrated constellation diagram in this exemplary embodiment is as follows: Figure 12 As shown, this addresses the issue of significant phase and amplitude deviations in the test system when testing constellation diagram data over long periods.

[0033] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for calibrating constellation diagram data in the testing of vector synthesis phase shifters, characterized in that: Includes the following steps: The process of testing the constellation diagram is divided into k time periods. Each time period includes constellation points generated by the control codes of the vector synthesis phase shifter, and a data point is selected from the constellation points in each time period as a calibration point. The pre-test calibration coordinate value of the calibration point is tested at time period 0, the constellation map data value is tested sequentially from time period 1 to time period k, and the post-test calibration coordinate value of the calibration point is tested at time period k+1. Subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the pre-test calibration coordinate values ​​to obtain the pre-test relative coordinate values; subtract the coordinate value of the first calibration point from the coordinate values ​​of each calibration point in the post-test calibration coordinate values ​​to obtain the post-test relative coordinate values. The average relative coordinate values ​​of the calibration points under the same control code are added together with the calibration coordinate values ​​before and after the test, and the average is obtained. The coordinate values ​​of the first calibration point with the calibration coordinate values ​​before the test are added to the average relative coordinate values ​​of the calibration points to obtain the calibration reference points of the constellation diagram for each time period. Subtract the corresponding calibration reference point value from the constellation diagram data value corresponding to the selected calibration point in time period 1 to time period k to obtain the correction value of the calibration point; perform linear interpolation on the correction value to obtain the interpolation correction value, and then subtract the corresponding interpolation correction value from the constellation diagram data value in time period 1 to time period k to obtain the calibrated constellation diagram data.

2. The method for calibrating constellation diagram data in the testing of a vector synthesis phase shifter according to claim 1, characterized in that: The vector synthesis phase shifter decomposes the input signal into two mutually orthogonal signals, namely the I-channel signal and the Q-channel signal. The N-bit control bits of the I-channel and the Q-channel can be generated respectively. IQ control code, the control code of the I channel changes from 0 to The control code for the Q-path changes from 0 to .

3. The method for calibrating constellation diagram data in the testing of a vector synthesis phase shifter according to claim 2, characterized in that: The scanning method for the constellation chart data values ​​for time period 1 to time period k is to proceed from the outer circle to the inner circle, with each circle's four edges corresponding to four time periods.

4. The method for calibrating constellation diagram data in the testing of a vector synthesis phase shifter according to claim 3, characterized in that: The test constellation diagram data values ​​for time period 1 to time period k only scan the data of the first M layers of the outer ring, that is, the scan terminates when the IQ control code (0,0) is scanned to the IQ control code (M-1,M).

5. The method for calibrating constellation diagram data in the testing of a vector synthesis phase shifter according to claim 3, characterized in that: In each time period, a data point is selected from the constellation points as a calibration point, specifically the first data point of the corresponding time period.

6. A method for calibrating constellation diagram data in testing a vector synthesis phase shifter according to claim 5, characterized in that: In the last time period, an additional constellation point is added as the end calibration point.

Images