Full-automatic measurement and control method and system of passive device with self-calibration and dynamic compensation

Abstract

The application relates to a passive device full-automatic measurement and control method and system with self-calibration and dynamic compensation, and belongs to the technical field of passive device performance testing. The method comprises the following steps: inputting a test signal into a test path and controlling a preset switch matrix to cooperate with the conduction of the test path; the test path comprises a first signal path and a second signal path; the first signal path is used for connecting the test signal to a preset test channel in the switch matrix after the test signal passes through a measured passive device; the second signal path is used for directly connecting the test signal to a preset reference channel in the switch matrix; for each test channel, the difference between a reference measurement value output by the test signal before entering the test channel and the reference measurement value output by the test signal after entering the test channel and when passing through the reference channel is obtained, the actual measurement value output by the test signal when entering the test channel is corrected by using the difference, and the performance index of the measured passive device is calculated according to the corrected actual measurement value. The application has the effect of realizing the integration of calibration and testing.

Description

Technical Field

[0001] This application relates to the field of passive device performance testing technology, and in particular to a fully automated measurement and control method and system for passive devices with self-calibration and dynamic compensation. Background Technology

[0002] In radio frequency communication systems, passive devices (such as power dividers, couplers, and filters) are widely used in base stations, radar, satellite communications, and other fields. For passive devices with multiple output ports, amplitude balance and phase balance are key performance indicators, reflecting the consistency of signal strength and phase at each output port, respectively.

[0003] Currently, the amplitude and phase balance testing of passive devices is typically automated using a vector network analyzer in conjunction with a switching matrix. The basic connection is as follows: the first port of the vector network analyzer is connected to the input of the passive device under test (DUT); each output port of the DUT is connected to multiple test channels of the switching matrix (i.e., each test channel corresponds one-to-one with an output port of the DUT); and the common output of the switching matrix is ​​connected to the second port of the vector network analyzer. During testing, the transmission coefficients of each output port are measured by sequentially switching the test channels of the switching matrix, and the amplitude and phase balance is obtained by subtracting the values ​​from the test channel values.

[0004] However, in existing testing schemes, the calibration and testing processes are independent, typically employing offline calibration. This means that the device under test (DUT) must be disconnected and the calibration kit reconnected during calibration, and then reconnected to the DUT for testing after calibration. This approach fails to integrate calibration and testing processes. Furthermore, during testing, devices such as vector network analyzers, switch matrices, and connecting cables are susceptible to real-time drift due to changes in ambient temperature and the instrument's own heat generation. Current technologies cannot monitor and compensate for this drift in real time during testing, making it difficult to guarantee the accuracy and stability of test results. Therefore, improvements are needed. Summary of the Invention

[0005] To achieve integrated calibration and testing, this application provides a fully automated measurement and control method and system for passive devices with self-calibration and dynamic compensation.

[0006] In a first aspect, this application provides a fully automated measurement and control method for passive devices with self-calibration and dynamic compensation, comprising:

[0007] The test signal is input into the test path, and a preset switch matrix is ​​controlled to cooperate with the conduction of the test path; wherein, the test path includes a first signal path and a second signal path, the first signal path is used for the test signal to be connected to a preset test channel in the switch matrix after passing through the passive device under test; the second signal path is used for the test signal to be directly connected to a preset reference channel in the switch matrix.

[0008] For each test channel, the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel is obtained, and the actual measurement value output when the test signal enters the test channel is corrected by the difference.

[0009] The performance indicators of the passive device under test are calculated based on the corrected actual measurement values. The performance indicators include at least amplitude balance and phase balance.

[0010] By employing the above technical solution, selectively switching the test signal to either the first signal path via the passive device under test (DUT) or the second signal path directly connected to the reference channel, and acquiring the difference between the reference measured value output by the reference channel before and after each test channel measurement, the actual measured value is corrected using this difference. This eliminates errors caused by system drift during the test. Since the correction involves acquiring the reference measured value and calculating the difference before and after each test channel measurement, it accurately reflects the real-time drift occurring during the measurement of that test channel. This makes the corrected actual measured value closer to the true characteristics of the passive device under test. Based on this, amplitude balance and phase balance are calculated, improving the accuracy and stability of the test results.

[0011] Optionally, for each test channel, acquiring the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel, and using the difference to correct the actual measurement value output by the test signal when entering the test channel, includes:

[0012] The test signal is output through the reference channel to obtain the initial reference measurement value;

[0013] Whenever the reference measurement value is updated and there are undetected test channels, the difference in future reference measurement values ​​is predicted based on the updated reference measurement value, and the measurement mode is determined based on the difference in value; wherein, the measurement mode is a single-channel measurement mode or a group measurement mode;

[0014] According to the determined test mode, the test channel to be tested is selected from the untested test channels, the actual measurement value of the test signal when it passes through the test channel is obtained, and after the test channel completes the measurement, the test signal is passed through the reference channel again and the reference measurement value is obtained to update the reference measurement value.

[0015] Whenever the reference measurement value is updated, the difference between the reference measurement value before and after the update is calculated. When the difference exceeds a preset threshold, the corresponding test channel before and after the update is re-registered as an undetected test channel. If the difference does not exceed the preset threshold, the actual measurement value of the corresponding test channel before and after the update is corrected using the difference.

[0016] By adopting the above technical solution, an initial reference measurement value is obtained, and the method of determining whether to use a single-channel measurement mode or a group measurement mode is based on the predicted future difference changes when the reference measurement value is updated each time. The channel under test is selected for measurement according to the determined mode, and the reference measurement value is updated after the measurement is completed. When the difference between the reference measurement value before and after the update exceeds a preset threshold, the corresponding test channel is put back into the test queue for retesting to ensure data validity; when the difference does not exceed the limit, the actual measurement value is directly corrected using this difference. This solution reduces the number of reference measurements and improves testing efficiency by using group measurements while ensuring correction accuracy.

[0017] Optionally, the corrected actual measurement value includes the actual amplitude and the actual phase;

[0018] The step of reclassifying the corresponding tested channels before and after the update as untested test channels includes:

[0019] Maintain in real time the maximum and minimum values ​​of the actual amplitude, as well as the maximum and minimum values ​​of the actual phase, for all currently tested test channels.

[0020] The test channel corresponding to the difference between the reference measurement values ​​before and after the update exceeds a preset threshold is taken as the test channel to be estimated. Based on the actual measurement value and the difference of the test channel to be estimated during the measurement, the estimated amplitude and estimated phase after the correction of the test channel to be estimated are calculated.

[0021] When the estimated amplitude is not between the maximum and minimum values ​​of the currently determined actual amplitude, and the estimated phase is not between the maximum and minimum values ​​of the currently determined actual phase, the corresponding test channel to be estimated is re-designated as an undetected test channel to await re-detection.

[0022] By adopting the above technical solution, it is clarified that the corrected actual measurement value includes both actual amplitude and actual phase, providing a basis for subsequent extreme value judgment. By maintaining the maximum and minimum values ​​of the actual amplitude and actual phase of the measured test channels in real time, channels with deviations exceeding the limits are designated as channels to be estimated, and their estimated amplitude and estimated limit are calculated. When both the estimated amplitude and estimated phase are outside the range of the current mechanism, the test channel is re-added to the test queue for retesting. This solution avoids indiscriminate retesting of all out-of-limit test channels, retesting only those channels that may affect the extreme value range, reducing unnecessary repeated measurements and improving the efficiency of the retesting process.

[0023] Optionally, the step of reclassifying the corresponding tested channels before and after the update as untested test channels further includes:

[0024] Based on the amplitude and phase value sequences of all currently measured test channels, predict the evolution direction of amplitude extrema and phase extrema respectively;

[0025] The amplitude-side dynamic threshold and the phase-side dynamic threshold are determined according to the evolution direction, respectively; wherein, the dynamic threshold is negatively correlated with the number of currently measured test channels and positively correlated with the evolution rate of extreme values;

[0026] For the test channel whose estimated amplitude is between the maximum and minimum values ​​of the currently determined actual amplitude, when the difference between the estimated amplitude of the test channel and the maximum value of the currently determined actual amplitude is less than the dynamic threshold on the side of the maximum amplitude, or the difference between the estimated amplitude and the minimum value of the currently determined actual amplitude is less than the dynamic threshold on the side of the maximum amplitude, the corresponding test channel to be estimated will be re-registered as an undetected test channel to wait for re-detection.

[0027] For the test channel whose estimated phase is between the maximum and minimum values ​​of the currently determined actual amplitude, when the difference between the estimated phase and the maximum value of the currently determined actual phase is less than the dynamic threshold on the maximum phase value side, or the difference between the estimated phase and the minimum value of the currently determined actual phase is less than the dynamic threshold on the minimum phase value side, the corresponding test channel will be re-designated as an undetected test channel to await re-detection.

[0028] By employing the above technical solution, the evolution direction of extreme values ​​is predicted based on the amplitude and phase value sequences of the tested channels, thereby determining the dynamic thresholds for both the amplitude and phase sides. These dynamic thresholds are negatively correlated with the number of tested channels and positively correlated with the evolution rate. For test channels whose estimated amplitude falls within the current extreme value range, if the difference between their estimated amplitude and the current maximum or minimum amplitude is less than the corresponding dynamic threshold, they are identified as potential extreme value test channels and reinstated into the test queue; the phase side is handled similarly. This solution introduces evolution trend information, making the determination of extreme value channels more accurate and avoiding the omission of high-probability extreme value channels or the misjudgment of low-probability channels due to fixed thresholds.

[0029] Optionally, the method further includes:

[0030] An extreme value probability model is established, which records the probability that each test channel becomes an amplitude extreme value or a phase extreme value in historical test data; the extreme value refers to the maximum or minimum value.

[0031] At the start of the test, the initial measurement order is determined according to the extreme value probability model. During the test, the measurement order of the remaining untested test channels is dynamically adjusted according to the actual measurement values ​​of the measured channels.

[0032] Furthermore, when performing the step of selecting the test channel from the untested test channels, the selection shall be made in accordance with the latest measurement order;

[0033] When the extreme ranges of amplitude and phase of the measured channels converge to within the preset accuracy, the test channels with extreme probability lower than the preset screening threshold in the remaining undetected test channels are judged as extreme value-irrelevant channels and the measurement is skipped.

[0034] By adopting the above technical solution, an extreme value probability model is established to record the historical extreme value probabilities of each test channel. At the start of the test, the initial measurement order is determined according to this extreme value probability model, prioritizing the measurement of test channels that are more likely to become extreme values ​​historically. During the test, the measurement order of the remaining test channels is dynamically adjusted based on the actual measurement values ​​of the already measured test channels, so that measurement resources are preferentially allocated to test channels that are likely to become extreme values. When the amplitude and phase extreme value ranges of the already measured test channels converge to within a preset accuracy, test channels with extreme value probabilities lower than the screening threshold among the remaining unmeasured test channels are judged as irrelevant channels and skipped from measurement. This solution prioritizes the measurement of high-probability extreme value channels, enabling the extreme value range to converge quickly, thereby minimizing the actual number of measurements and improving overall test efficiency while ensuring accurate extreme value judgment.

[0035] Optionally, the method further includes:

[0036] Based on the extreme value probability model, a mutual exclusion relationship is established between test channels. The mutual exclusion relationship is used to indicate that measurement channels with extreme value probabilities higher than a preset mutual exclusion threshold should not be simultaneously assigned to the same measurement group.

[0037] When performing the step of selecting the test channel to be tested from the untested test channels, if the current measurement mode is grouped, the composition of the current measurement group is determined according to the mutual exclusion relationship, so that test channels with mutual exclusion relationships are not selected into the same measurement group at the same time.

[0038] By adopting the above technical solution, a mutual exclusion relationship is established between test channels based on the extreme value probability model, indicating that test channels with extreme value probabilities higher than a preset mutual exclusion threshold should not be simultaneously assigned to the same measurement value. When the test channel selection step is currently in group measurement mode, the composition of the measurement group is determined based on this mutual exclusion relationship, distributing mutually exclusive test channels into different measurement groups. This solution avoids multiple high-probability extreme value channels being concentrated in the same group. When the group needs to be retested as a whole due to drift exceeding limits, it will not simultaneously affect multiple high-probability extreme value channels, thereby reducing retest costs and improving the robustness of group measurements.

[0039] Optionally, the method further includes:

[0040] The reference channel is periodically self-tested, and the reference measurement value of the reference channel is compared with the known nominal value of the calibration component;

[0041] When the comparison deviation exceeds the predetermined threshold, the reference channel is determined to be abnormal, and an alarm message is output.

[0042] By adopting the above technical solution, the reference channel is periodically self-tested. The reference measurement value of the reference channel is compared with the nominal value of the known calibration component. When the comparison deviation exceeds a predetermined threshold, the reference channel is deemed abnormal and an alarm message is output. This solution can promptly detect performance degradation or malfunction of the reference channel itself, avoiding the failure of the correction benchmark of the entire test system due to the abnormality of the reference channel, thereby ensuring the reliability of the test results.

[0043] Secondly, this application provides a fully automatic measurement and control system for passive devices with self-calibration and dynamic compensation, including,

[0044] A signal flow control module is used to input test signals into a test path and control a preset switch matrix to cooperate with the conduction of the test path; wherein, the test path includes a first signal path and a second signal path, the first signal path is used for the test signal to enter a preset test channel in the switch matrix after passing through the passive device under test; the second signal path is used for the test signal to directly enter a preset reference channel in the switch matrix.

[0045] The test result correction module is used to obtain, for each test channel, the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel, and use the difference to correct the actual measurement value output by the test signal when entering the test channel;

[0046] The performance index calculation module is used to calculate the performance index of the passive device under test based on the corrected actual measurement value. The performance index includes at least amplitude balance and phase balance.

[0047] Thirdly, this application provides a fully automatic measurement and control device for passive devices with self-calibration and dynamic compensation, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any of the first aspects.

[0048] Fourthly, this application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described in any of the first aspects.

[0049] In summary, this application includes the following beneficial technical effects:

[0050] In this application, by selectively switching the test signal to either a first signal path via the passive device under test (DUT) or a second signal path directly connected to the reference channel, and by acquiring the difference between the reference measured value output by the reference channel before and after each test channel measurement, the actual measured value is corrected using this difference. This eliminates errors caused by system drift during the test. Since the correction involves acquiring the reference measured value and calculating the difference before and after each test channel measurement, it accurately reflects the real-time drift occurring during the measurement of that test channel. This makes the corrected actual measured value closer to the true characteristics of the DUT. Based on this, amplitude balance and phase balance are calculated, improving the accuracy and stability of the test results. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart illustrating the fully automated measurement and control method for passive devices with self-calibration and dynamic compensation disclosed in the embodiments of this application.

[0053] Figure 2This is a structural block diagram of the fully automatic measurement and control system for passive devices with self-calibration and dynamic compensation disclosed in the embodiments of this application.

[0054] Explanation of reference numerals in the attached diagram: 201, Signal flow control module; 202, Test result correction module; 203, Performance index calculation module. Detailed Implementation

[0055] The following is in conjunction with the appendix Figure 1-2 This application will be described in further detail.

[0056] This application discloses a fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation (hereinafter referred to as the automatic measurement and control method), the execution subject of which is a fully automatic measurement and control system for passive devices with self-calibration and dynamic compensation (hereinafter referred to as the automatic measurement and control system). The following will be combined with... Figure 1 This section elaborates on the specific implementation process of automatic measurement and control methods in an automatic measurement and control system.

[0057] S101, input the test signal into the test path and control the preset switch matrix to cooperate with the conduction of the test path; wherein, the test path includes a first signal path and a second signal path, the first signal path is used to allow the test signal to enter the preset test channel in the switch matrix after passing through the passive device under test; the second signal path is used to allow the test signal to directly enter the preset reference channel in the switch matrix.

[0058] S102, for each test channel, obtain the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel, and use the difference to correct the actual measurement value output when the test signal enters the test channel.

[0059] S103, calculate the performance indicators of the passive device under test based on the corrected actual measurement values. The performance indicators include at least amplitude balance and phase balance.

[0060] In implementation, the automatic measurement and control system is communicatively connected to a vector network analyzer, a switch matrix, and a two-way mechanical switching switch. This connection controls the operating status of the aforementioned equipment, collects measurement data, and performs data processing and error correction. The vector network analyzer generates test signals and receives return signals. It has a first port and a second port. The first port outputs the test signal, and the second port receives the signal returned via the test path. The vector network analyzer measures the transmission coefficient, a complex number containing amplitude and phase components, reflecting the amplitude and phase changes of the signal after passing through the test path, respectively. The switch matrix (specifically, an RF switch matrix) has multiple input channels and a common output terminal, which is connected to the second port of the vector network analyzer. Under the control of the automatic measurement and control system, the switch matrix can independently select any input channel to the common output terminal, achieving channel switching functionality. The switch matrix is ​​a load-bearing switch matrix; its unselected input channels are internally terminated with a 50Ω load. Specifically, when each input channel is not selected, its input terminal is connected to ground through a 50Ω resistor to avoid signal reflection interference with measurement accuracy in an open-circuit state.

[0061] Specifically, the input channels of the switch matrix are divided into one reference calibration channel and multiple test channels. The reference calibration channel and each test channel are designed with equal length and equal phase within the switch matrix. Specifically, during printed circuit board routing, the transmission path length from the input terminal of the reference calibration channel to the common output terminal is used as the reference length. For test channels with a transmission path length shorter than the reference length, a serpentine routing method is used to increase their transmission line length, i.e., the path is extended by routing the transmission lines in a curved and meandering manner. For test channels with a transmission path length longer than the reference length, the distance is shortened by optimizing the routing path or adjusting the input port position. All input channels use the same microstrip line structure, with consistent parameters such as linewidth, line thickness, and substrate dielectric constant, ensuring that the phase characteristics of each channel are identical under the condition of equal length. Through the above design, the amplitude attenuation and phase change introduced when the test signal is transmitted through different input channels are basically consistent, enabling the measurement value of the reference calibration channel to accurately reflect the system state of the test channel.

[0062] The 1 / 2 mechanical switch has one input terminal and two output terminals. The input terminal connects to the first port of a vector network analyzer, the first output terminal connects to the input terminal of the passive device under test (DUT), and the second output terminal connects to the reference calibration channel of the switch matrix. Under the control of an automatic measurement and control system, this 1 / 2 mechanical switch can selectively connect the input terminal to either the first or second output terminal, thus switching the signal path. Correspondingly, the input terminal of the DUT is connected to the first output terminal of the 1 / 2 mechanical switch, and each output port of the DUT is connected to the corresponding test channel of the switch matrix via connecting cables. The number of test channels is equal to the number of output ports of the DUT, and they correspond one-to-one.

[0063] When the system is used for the first time or when the hardware configuration is changed, the automatic measurement and control system performs an initialization calibration procedure.

[0064] The automatic measurement and control system controls a two-way mechanical switch to switch the test signal to the second output terminal, allowing the test signal to be directly connected to the reference calibration channel of the switch matrix. The automatic measurement and control system controls the switch matrix to sequentially select each test channel and the reference calibration channel. When each channel is selected, the vector network analyzer outputs a test signal through its first port and receives a return signal through its second port, measuring the transmission coefficient of the currently selected channel. The transmission coefficient is the test result directly output by the vector network analyzer, usually expressed as a complex number with a real part and an imaginary part, such as x + jy, where x is the real part, y is the imaginary part, and j is the imaginary unit (satisfying j² = -1). This complex number can also be converted to amplitude and phase form: amplitude A = √(x² + y²), phase θ = arctan(y / x). The automatic measurement and control system reads the transmission coefficient measured by the vector network analyzer through the communication interface and stores it in its internal memory.

[0065] The automatic measurement and control system records the transmission coefficient S_ref of the reference calibration channel and the transmission coefficient S_ch_i of each test channel. For test channel i, the automatic measurement and control system calculates its channel error ΔS_i = S_ch_i - S_ref relative to the reference calibration channel. This subtraction is a complex subtraction, that is, subtracting the real and imaginary parts separately: real part error = real part (S_ch_i) - real part (S_ref), imaginary part error = imaginary part (S_ch_i) - imaginary part (S_ref). Using real and imaginary parts for subtraction avoids phase entanglement problems and ensures accurate calculation results. This channel error ΔS_i reflects the inherent differences between test channel i and the reference calibration channel in terms of transmission path within the switch matrix, connecting cables, and adapters. These differences may be positive or negative; retaining the sign helps in subsequent accurate correction. The automatic measurement and control system stores the inherent channel error of each test channel in its internal memory for use in subsequent tests. This completes the initialization calibration step.

[0066] At the start of the test, the automatic measurement and control system controls the vector network analyzer to output a test signal through its first port. This test signal is an radio frequency (RF) signal, and its frequency range is set according to the operating frequency band of the passive device under test (DUT). The automatic measurement and control system controls a two-way mechanical switch to selectively switch the test signal to either the first or second output port, thereby selecting the first or second signal path for the test signal flow. The first signal path refers to the path through which the test signal passes through the DUT and then enters the switch matrix test channel; the second signal path refers to the path through which the test signal bypasses the DUT and directly enters the switch matrix reference calibration channel.

[0067] During testing, the automatic measurement and control system switches between the two paths mentioned above according to measurement requirements. Specifically, for each test channel, the automatic measurement and control system acquires data using a timing sequence of "previous reference measurement → test channel measurement → subsequent reference measurement." That is, before measuring a certain test channel, the automatic measurement and control system first controls a two-way mechanical switch to switch to the second output terminal, allowing the test signal to be directly connected to the reference calibration channel, and simultaneously controls the RF switch matrix to select the reference calibration channel. The vector network analyzer measures the transmission coefficient of the reference calibration channel, and the automatic measurement and control system records this reference measurement value, denoted as the previous reference measurement value R_before. R_before is a complex number, stored in the form of a complex number with a real part plus an imaginary part.

[0068] Subsequently, the automatic measurement and control system controls the two-way mechanical switch to switch to the first output terminal, so that the test signal passes through the passive device under test and enters the corresponding test channel. At the same time, it controls the RF switch matrix to select the currently tested test channel i. The vector network analyzer measures the transmission coefficient of the test channel, and the automatic measurement and control system reads the measured value and records it as the actual measured value M_i. M_i is a complex number and is stored in the form of a complex number with a real part plus an imaginary part.

[0069] After completing the measurement of this test channel, the automatic measurement and control system again controls the two-way mechanical switch to switch to the second output terminal, and controls the RF switch matrix to select the reference calibration channel to obtain the reference measurement value R_after. R_after is a complex number, stored in the form of a complex number with a real part plus an imaginary part.

[0070] In summary, the timing test process for this test channel, consisting of "pre-reference measurement → test channel measurement → post-reference measurement", is completed.

[0071] Next, the automatic measurement and control system calculates the difference between the subsequent reference measurement value and the previous reference measurement value, ΔR = R_after - R_before. This subtraction is a complex subtraction, that is, subtracting the amplitude component (i.e., the real part) and the phase component (i.e., the imaginary part) separately: real part drift = real part(R_after) - real part(R_before), imaginary part drift = imaginary part(R_after) - imaginary part(R_before). This difference ΔR reflects the real-time drift of the system due to factors such as changes in ambient temperature and instrument heating during the measurement period of this test channel.

[0072] The automatic measurement and control system uses the aforementioned drift amount to dynamically correct the actual measured value. The correction employs complex subtraction, meaning that for test channel i, the corrected actual measured value is M_i' = M_i - ΔR. This subtraction also applies to the real and imaginary parts separately: corrected real part = real part (M_i) - real part (ΔR), corrected imaginary part = imaginary part (M_i) - imaginary part (ΔR). The corrected actual measured value M_i' eliminates the influence of real-time system drift during measurement. Then, the automatic measurement and control system further subtracts the channel error of the corresponding test channel stored in the initial calibration step from the corrected actual measured value to obtain the true transmission coefficient T_i = M_i' - ΔS_i of the i-th output of the passive device under test. This subtraction is also a complex subtraction. T_i is the true transmission characteristic of the i-th output of the passive device under test, which can be converted into amplitude component A_i and phase component θ_i, where A_i = √(real part² + imaginary part²) and θ_i = arctan(imaginary part / real part).

[0073] The automatic measurement and control system follows the steps described above, sequentially executing the complete process of "pre-reference measurement—channel measurement—post-reference measurement—correction—subtraction of channel error" for all test channels. During the measurement process, the automatic measurement and control system monitors the drift amount (i.e., difference ΔR) calculated each time in real time. When the absolute value of the real part or the absolute value of the imaginary part of ΔR exceeds a preset threshold, it indicates that the system drift is too large during the measurement of that channel, and the reliability of the measurement data is insufficient. The preset threshold can be set according to the system calibration results or empirical values. For example, the real part threshold can be set to 0.01, and the imaginary part threshold can be set to 0.01. Those skilled in the art can adjust it according to the actual accuracy requirements. At this time, the automatic measurement and control system marks the channel as invalid and puts the test channel back into the preset test queue for subsequent remeasurement. When both the real and imaginary parts of ΔR are within the preset threshold range, the automatic measurement and control system accepts the corrected data as a valid result and uses it for subsequent performance index calculations. In other embodiments, for adjacent test channels, the automatic measurement and control system can directly use the post-reference measurement value of the previous test channel as the pre-reference measurement value of the next test channel, thereby reducing the number of reference measurements and improving test efficiency.

[0074] Finally, after completing the measurement and correction of all test channels, the automatic measurement and control system obtains the true transmission coefficient T_i for each channel. Each T_i contains an amplitude value A_i and a phase value θ_i. The automatic measurement and control system calculates the amplitude balance, which is the difference between the maximum and minimum amplitude values ​​in all test channels: Amplitude Balance = max(A_i) - min(A_i), in decibels. The automatic measurement and control system calculates the phase balance, which is the difference between the maximum and minimum phase values ​​in all test channels: Phase Balance = max(θ_i) - min(θ_i), in degrees. The automatic measurement and control system outputs the calculated amplitude balance and phase balance as key performance indicators of the passive device under test for subsequent quality assessment or data analysis.

[0075] Optionally, S102 specifically includes the following sub-steps:

[0076] S1021 outputs the test signal via the reference channel to obtain the initial reference measurement value;

[0077] S1022, Whenever the reference measurement value is updated and there are undetected test channels, the difference change of the future reference measurement value is predicted based on the updated reference measurement value, and the measurement mode is determined based on the difference change; wherein, the measurement mode is a single-channel measurement mode or a group measurement mode;

[0078] S1023, according to the determined test mode, select the test channel to be tested from the untested test channels, obtain the actual measurement value when the test signal passes through the test channel to be tested, and after the test channel to be tested completes the measurement, pass the test signal through the reference channel again and obtain the reference measurement value to update the reference measurement value.

[0079] S1024. Whenever the reference measurement value is updated, the difference between the reference measurement value before and after the update is calculated. If the difference exceeds a preset threshold, the corresponding test channel before and after the update is re-registered as an undetected test channel. If the difference does not exceed the preset threshold, the actual measurement value of the corresponding test channel before and after the update is corrected using the difference.

[0080] Maintain in real time the maximum and minimum values ​​of the actual amplitude, as well as the maximum and minimum values ​​of the actual phase, for all currently tested test channels.

[0081] Specifically, S1024, "re-application of the corresponding test channels before and after the update as untested test channels," includes the following steps:

[0082] The test channel corresponding to the difference between the reference measurement values ​​before and after the update exceeds the preset threshold is taken as the test channel to be estimated. Based on the actual measurement value and the difference of the test channel to be estimated during measurement, the estimated amplitude and estimated phase of the test channel to be estimated after correction are calculated.

[0083] When the estimated amplitude is not between the maximum and minimum values ​​of the currently determined actual amplitude, and the estimated phase is not between the maximum and minimum values ​​of the currently determined actual phase, the corresponding test channel to be estimated will be re-designated as an untested test channel to await retesting.

[0084] Based on the amplitude and phase value sequences of all currently measured test channels, predict the evolution direction of amplitude extrema and phase extrema respectively;

[0085] The amplitude-side dynamic threshold and the phase-side dynamic threshold are determined according to the evolution direction; among them, the dynamic threshold is negatively correlated with the number of currently measured test channels and positively correlated with the evolution rate of extreme values.

[0086] For test channels whose estimated amplitude is between the maximum and minimum values ​​of the currently determined actual amplitude, if the difference between the estimated amplitude of the test channel and the maximum value of the currently determined actual amplitude is less than the dynamic threshold on the side of the maximum amplitude, or the difference between the estimated amplitude and the minimum value of the currently determined actual amplitude is less than the dynamic threshold on the side of the maximum amplitude, the corresponding test channel will be re-registered as an undetected test channel to await re-detection.

[0087] For a test channel whose estimated phase is between the maximum and minimum values ​​of the currently determined actual amplitude, if the difference between the estimated phase and the maximum value of the currently determined actual phase is less than the dynamic threshold on the maximum phase value side, or the difference between the estimated phase and the minimum value of the currently determined actual phase is less than the dynamic threshold on the minimum phase value side, the corresponding test channel will be re-registered as an undetected test channel to await re-detection.

[0088] S301, Establish an extreme value probability model. The extreme value probability model records the probability that each test channel becomes an amplitude extreme value or a phase extreme value in historical test data; extreme value refers to the maximum or minimum value.

[0089] S302: At the start of the test, the initial measurement order is determined according to the extreme value probability model. During the test, the measurement order of the remaining untested test channels is dynamically adjusted according to the actual measurement values ​​of the measured channels.

[0090] S303: When the extreme range of amplitude and extreme range of phase of the measured channel converge to within the preset accuracy, the test channel with an extreme value probability lower than the preset screening threshold in the remaining undetected test channels is determined as an extreme value-irrelevant channel and the measurement is skipped.

[0091] S304. Based on the extreme value probability model, establish a mutual exclusion relationship between test channels. The mutual exclusion relationship is used to indicate that measurement channels with extreme value probabilities higher than the preset mutual exclusion threshold should not be simultaneously assigned to the same measurement group.

[0092] When performing the step "Select the test channel to be tested from the untested test channels" in S1023, the selection is performed according to the latest measurement order; if the current measurement mode is in group measurement mode, the composition of the current measurement group is determined according to the mutual exclusion relationship, so that test channels with mutual exclusion relationship are not selected into the same measurement group at the same time.

[0093] In implementation, the automatic control system uses a cyclic structure to perform the test. The cyclic structure consists of S1021 to S1024, and the cyclic condition is that there is a test channel that has not yet completed a valid measurement.

[0094] Before entering the loop for the first time, all untested test channels are stored in a preset test queue. Upon entering the loop for the first time, the automatic measurement and control system controls a two-way mechanical switch to switch to the second output terminal, allowing the test signal to directly access the reference calibration channel. Simultaneously, the system controls the RF switch matrix to select the reference calibration channel. The vector network analyzer measures the transmission coefficient of the reference calibration channel, and the automatic measurement and control system reads this measurement value, recording it as the initial reference measurement value R0. This measurement value is a complex number, stored in the internal memory of the automatic measurement and control system in the form of a real part plus an imaginary part. Then, the automatic measurement and control system predicts the future difference changes in the reference measurement value based on the current reference measurement value (initially R0, and subsequently the most recently updated reference measurement value), and determines the measurement mode used in the current loop accordingly. The prediction method uses a lookup table. The automatic measurement and control system internally stores a prediction mapping table, which records the correspondence between the amplitude component (unit: dB) of the reference measurement value and the predicted drift state and recommended group size. This mapping table is obtained through prior experimental calibration and specifically includes three fields: reference measurement value amplitude range field, predicted drift state field, and recommended group size field. For example:

[0095] When the reference measurement range field is greater than 0.50, the corresponding predicted drift state is drastic fluctuation, and the corresponding recommended group size is 1 channel per group (single channel mode).

[0096] When the reference measurement range field is between 0.30 and 0.50, the corresponding predicted drift state is fluctuating, and the corresponding recommended group size is 1 channel per group (single channel mode).

[0097] When the reference measurement range field is between 0.15 and 0.30, the corresponding predicted drift state is moderate, and the corresponding recommended group size is 2 channels per group (group measurement mode).

[0098] When the reference measurement value amplitude range field is between 0.05 and 0.15, the corresponding predicted drift state is stable, and the corresponding recommended group size is 4 channels per group (group measurement mode).

[0099] When the reference measurement range field is less than 0.05, the corresponding predicted drift state is extremely stable, and the corresponding recommended group size is 8 channels per group (group measurement mode).

[0100] The automatic measurement and control system converts the current reference measurement value into an amplitude component, looks up the corresponding amplitude range in the above mapping table, and obtains the recommended group size. If the recommended group size is 1, the current measurement mode is a single-channel measurement mode; if the recommended group size is greater than 1, the current measurement mode is a grouped measurement mode, and the group size is the recommended value.

[0101] Next, the automatic measurement and control system performs the measurement according to the measurement mode determined in S1022.

[0102] If the current measurement mode is single-channel, the automatic measurement and control system executes the following sub-steps: Take a measurement test channel from the queue to be tested as the current test channel; control the mechanical switch to switch to the second output terminal, select the reference calibration channel, and obtain the previous reference measurement value R_before; control the mechanical switch to switch to the first output terminal, select the current test channel, and obtain the actual measurement value M; control the mechanical switch to switch to the second output terminal, select the reference calibration channel, and obtain the subsequent reference measurement value R_after; store R_after as the updated reference measurement value for prediction in the next cycle.

[0103] If the current measurement mode is grouped, the automatic measurement and control system controls the following sub-steps: k test channels are sequentially retrieved from the queue to be measured as the current measurement group, where k is the recommended group size determined by S1022; the two-way mechanical switch is switched to the second output terminal to select the reference calibration channel and obtain the reference measurement value R_before before the group is formed; each test channel within the measurement group is selected in this way, and the actual measurement values ​​M_1, M_2...M_k of each test channel are obtained sequentially; the two-way mechanical switch is switched to the second output terminal to select the reference calibration channel and obtain the reference measurement value R_after after the group is formed; R_after is stored as the updated reference measurement value for prediction in the next cycle.

[0104] The automatic measurement and control system calculates the drift (i.e., difference) ΔR = R_after - R_before for the current measurement cycle (one channel in single-channel mode, one group of channels in group mode). This subtraction is a complex subtraction, that is, subtracting the real part and the imaginary part separately to obtain the real drift Δx and the imaginary drift Δy. The automatic measurement and control system determines whether the absolute value of Δx exceeds a preset amplitude threshold (set to 0.01 in this embodiment, corresponding to 0.1 dB) and whether the absolute value of Δy exceeds a preset phase threshold (set to 0.01 radians in this embodiment, corresponding to approximately 0.6 degrees). The above thresholds can be adjusted according to the system calibration results or actual accuracy requirements.

[0105] When both Δx and Δy are within the threshold range, it indicates that the system drift during measurement is acceptable. The automatic measurement and control system uses ΔR to correct the actual measured values ​​acquired in the current measurement cycle: for single-channel mode: M' = M - ΔR; for group mode: M'_j = M_j - ΔR (j = 1 to k). The corrected measured value M' or M'_j is the result after deducting the real-time drift. The automatic measurement and control system further subtracts the channel error of the corresponding test channel stored in the initial calibration phase to obtain the true transmission coefficient T of that channel. The automatic measurement and control system converts the true transmission coefficient T into amplitude component A and phase component θ, and stores them in the measured channel database.

[0106] When Δx or Δy exceeds the threshold range, it indicates that the system drift is too large during the measurement period, and the reliability of the actual measured values ​​obtained in the current measurement cycle is insufficient. The automatic measurement and control system puts the corresponding test channel in the current measurement cycle back into the test queue, waiting for subsequent remeasurement. For group mode, all channels in the group are put back into the test queue. These channels will redetermine the measurement mode and be measured again in subsequent loops based on the latest reference measurement values. After completing the above processing, the automatic measurement and control system determines whether the test queue is empty. If it is not empty, it returns to S1022 and uses the latest updated reference measurement values ​​for the prediction and measurement of the next loop; if it is empty, the main loop ends and enters the performance index calculation stage.

[0107] During the above-mentioned cyclic test, the automatic measurement and control system performs the following two tasks in parallel outside the main loop, and the results are used for the abnormal retest determination in S1024.

[0108] Task 1: Maintenance of Extreme Values ​​for Measured Channels. The automatic measurement and control system maintains a database of measured channels, recording the channel number, amplitude value A_i, and phase value θ_i for each test channel that has completed a valid measurement (i.e., the drift is within limits and has been corrected). After each main loop completes a valid measurement of a channel (i.e., obtains the true transmission coefficient T of that channel), the automatic measurement and control system updates the following extreme values: maximum amplitude A_max = max(A_i of all measured channels); minimum amplitude A_min = min(A_i of all measured channels); maximum phase θ_max = max(θ_i of all measured channels); minimum phase θ_min = min(θ_i of all measured channels).

[0109] Task 2: Prediction of Extreme Value Evolution Direction. The automatic control system maintains the amplitude value sequence [A_{n-9}, A_{n-8}, ..., A_n] and phase value sequence [θ_{n-9}, θ_{n-8}, ..., θ_n] of the most recent Z (Z is a preset positive integer, such as 10) measured test channels, where n is the total number of currently measured channels. The automatic measurement and control system uses a linear regression method to predict the extreme value evolution direction. Taking the amplitude value as an example, the sequence index t=1, 2, ..., 10 is used as the independent variable, and the amplitude value A is used as the dependent variable. The linear regression slope k_A is calculated as follows:

[0110] k_A = [10·Σ(t·A_t)-Σt·ΣA_t] / [10·Σt²-(Σt)²]; where ∑ represents summation over t = 1 to 10; · represents multiplication.

[0111] If k_A > 0.01 (i.e., the amplitude increases by more than 0.01 dB on average per test channel), the predicted amplitude extremum evolves towards the maximum value, meaning the amplitude value of subsequent test channels may continue to increase, and the current A_max may be refreshed. If k_A < -0.01 (i.e., the amplitude decreases by more than 0.01 dB on average per test channel), the predicted amplitude extremum evolves towards the minimum value, meaning the amplitude value of subsequent test channels may continue to decrease, and the current A_min may be refreshed. If |k_A| ≤ 0.01, the predicted amplitude extremum tends to stabilize. The same method is used to predict the evolution direction of the phase value, with the slope threshold set to 0.01 radians (approximately 0.6 degrees). If k_θ > 0.01, the predicted phase evolves towards the maximum value; if k_θ < -0.01, the predicted phase evolves towards the minimum value; if |k_θ| ≤ 0.01, the predicted phase tends to stabilize.

[0112] When drift exceeds the limit and the corresponding channel needs to be put back into the test queue in the main loop S1024, the automatic measurement and control system does not unconditionally accept the retest of all channels. Instead, it selects channels based on their likelihood of becoming extreme values ​​in order to reduce the cost of retesting.

[0113] The automatic control system performs the following judgment process for each test channel that needs to be retested:

[0114] Step 1: Calculate the estimated amplitude and estimated phase. Let the actual measured value of the test channel under abnormal measurement be M (complex form), and the intra-group drift be ΔR (complex form). The automatic measurement and control system calculates the estimated true transmission coefficient T_est = M - ΔR - ΔS_i, where ΔS_i is the inherent channel error (complex form) of the test channel under test. Convert T_est into amplitude A_est (i.e., estimated amplitude) and phase θ_est (i.e., estimated phase).

[0115] Step 2: Extreme Value Range Determination. The automatic measurement and control system queries the extreme values ​​(A_max, A_min, θ_max, θ_min) of the currently measured channel. A_max is the maximum actual amplitude of the amplitude value sequence of the currently measured test channel; A_min is the minimum actual amplitude of the amplitude value sequence of the currently measured test channel; θ_max is the maximum actual phase of the phase value sequence of the currently measured test channel; θ_min is the minimum actual phase of the phase value sequence of the currently measured test channel.

[0116] If A_est < A_min and θ_est < θ_min, or A_est > A_max and θ_est > θ_max, then the predicted amplitude and predicted phase of the test channel to be estimated both fall outside the current extreme value range and cannot become a new extreme value. Therefore, it is determined to be an extreme value-independent channel and is skipped for remeasurement.

[0117] Step 3: Dynamic Threshold Determination. For channels whose estimated amplitude is between A_min and A_max, or whose estimated phase is between θ_min and θ_max, the automatic measurement and control system further uses a dynamic threshold for determination. The automatic measurement and control system determines the dynamic threshold based on the evolution direction predicted by parallel task two:

[0118] If the predicted evolution direction is towards the maximum value, then the dynamic threshold TH_A_max on the amplitude maximum side is 0.02 / (1+N_measured / 50), where N_measured is the number of measured test channels; the dynamic threshold TH_θ_max on the phase maximum side is 0.02 / (1+N_measured / 50).

[0119] If the predicted evolution direction is towards the minimum value, then the dynamic threshold TH_A_min on the minimum amplitude side is 0.02 / (1+N_measured / 50); the dynamic threshold TH_θ_min on the minimum phase side is 0.02 / (1+N_measured / 50).

[0120] If the prediction tends to stabilize, the dynamic thresholds on both sides are both set to 0.01 / (1+N_measured / 50).

[0121] In the above formula, the denominator (1+N_measured / 50) causes the dynamic threshold to decrease as the number of measured test channels increases, reflecting the law that the extreme value range tends to stabilize as the number of measurements increases. The coefficients 0.02 and 0.01 are determined based on system calibration and can be adjusted by those skilled in the art according to actual accuracy requirements. The automatic measurement and control system calculates the following differences: ΔA_max = A_max - A_est; ΔA_min = A_est - A_min; Δθ_max = θ_max - θ_est; Δθ_min = θ_est - θ_min.

[0122] When ΔA_max < TH_A_max or ΔA_min < TH_A_min, the test channel to be estimated is determined to be a potential amplitude extremum channel, and retesting is retained; when Δθ_max < TH_θ_max or Δθ_min < TH_θ_min, the test channel to be estimated is determined to be a potential phase extremum channel, and retesting is retained. Only when all differences are greater than the corresponding dynamic threshold is it determined to be an extremum-independent channel, and retesting is skipped.

[0123] The automatic measurement and control system accumulates historical test data during long-term operation and establishes an extreme value probability model. This extreme value probability model is stored in the internal memory of the automatic measurement and control system and is continuously updated as the number of tests on different passive devices increases. Specifically, the automatic measurement and control system maintains four counters for each test channel:

[0124] C_Amax[i]: The number of times that test channel i reaches its maximum amplitude;

[0125] C_Amin[i]: The number of times test channel i reaches the minimum amplitude;

[0126] C_θmax[i]: The number of times test channel i reaches the maximum phase value;

[0127] C_θmin[i]: The number of times test channel i becomes the minimum phase value.

[0128] After each test, the automatic measurement and control system records the channel number corresponding to the maximum amplitude, the channel number corresponding to the minimum amplitude, the channel number corresponding to the maximum phase, and the channel number corresponding to the minimum phase, and increments the value of the corresponding counter by 1.

[0129] Let the total number of tests be N_total. The automatic measurement and control system calculates the extreme value probability of each test channel:

[0130] P_Amax[i]= C_Amax[i] / N_total;

[0131] P_Amin [i]= C_Amin [i] / N_total;

[0132] P_θmax[i]= C_θmax[i] / N_total;

[0133] P_θmin[i]= C_θmin[i] / N_total.

[0134] The comprehensive extreme value probability P[i] of test channel i takes the maximum value among the above four terms, that is, P[i] = max(P_Amax[i], P_Amin[i], P_θmax[i], P_θmin[i]).

[0135] At the start of the test, the automatic measurement and control system determines the initial measurement order based on the extreme value probability model. Specifically, all test channels that have not yet been measured are sorted from high to low according to their comprehensive extreme value probability P[i] to form an initial measurement queue. Channels with higher extreme value probabilities are placed at the front of the queue and measured first.

[0136] During the testing process, after each valid measurement of a test channel is completed, the automatic measurement and control system dynamically adjusts the measurement order of the remaining untested test channels based on the actual measurement values ​​of the measured channels. The adjustment rules are as follows:

[0137] Let the channel number corresponding to the current actual maximum amplitude be I_Amax, and its extreme value probability be P_Amax[I_Amax] (that is, the probability that this test channel becomes the maximum amplitude). Let the channel number with the highest probability of maximum amplitude in the extreme value probability model be I_Amax_model, and its probability be P_Amax_max = max(P_Amax[i]).

[0138] Condition 1: If I_Amax = I_Amax_model and P_Amax[I_Amax] ≥ 0.1, then the actual maximum amplitude is determined to be consistent with the historical model. In this case, the priority of the adjacent channels of channel I_Amax (whose numbers differ by ±1 and are within the range of 1 to the total number of channels) is multiplied by a coefficient of 1.2. Adjacent channels are physically adjacent to the extreme value channels, and their amplitude and phase characteristics are similar, and they may also have high amplitude values.

[0139] Condition 2: If I_Amax ≠ I_Amax_model and P_Amax[I_Amax] < 0.1, then the actual maximum amplitude value is determined to occur on the channel with the lower probability. In this case, the priority of the channel I_Amax_model with the highest probability of maximum amplitude value in the extreme value probability model is multiplied by a coefficient of 0.5 to reduce its priority.

[0140] Similarly, let the channel number corresponding to the current actual maximum phase value be I_Pmax, and its extreme value probability be P_Pmax[I_Pmax] (that is, the probability that this test channel becomes the maximum phase value). Let the channel number with the highest probability of the maximum phase value in the extreme value probability model be I_Pmax_model, and its probability be P_Pmax_max=max(P_Pmax[i]).

[0141] Condition 3: If I_Pmax = I_Pmax_model and P_Pmax[I_Pmax] ≥ 0.1, then the actual maximum phase value is determined to be consistent with the historical model. In this case, the priority of the adjacent channels of channel I_Pmax (with numbering difference ±1) is multiplied by a coefficient of 1.2.

[0142] Condition 4: If I_Pmax ≠ I_Pmax_model and P_Pmax[I_Pmax] < 0.1, then the actual maximum phase value is determined to occur on the channel with the lower probability. In this case, the priority of the channel I_Pmax_model with the highest probability of maximum phase value in the extreme value probability model is multiplied by a coefficient of 0.5 to reduce its priority.

[0143] For each remaining untested test channel i, its initial priority coefficient f_i = 1. The four conditions mentioned above are checked sequentially. If channel i meets any of the conditions, f_i is multiplied by the coefficient corresponding to that condition. For example, if test channel i simultaneously meets the amplitude adjustment coefficient of 1.2 and the phase adjustment coefficient of 0.5, then f_i = 1 × 1.2 × 0.5 = 0.6. If channel i does not meet any conditions, f_i remains 1.

[0144] The automatic measurement and control system recalculates the comprehensive score of each remaining channel according to P[i]×f_i, and reorders them from highest to lowest score to form an updated measurement queue. Channels with higher scores are measured first.

[0145] The automatic measurement and control system monitors the convergence degree of the extreme values ​​of the measured test channels in real time during the testing process. The extreme value range is considered to have converged when all of the following conditions are met: the number of measured test channels exceeds 30% of the total number of test channels (this percentage can be adjusted according to the testing accuracy requirements); the amplitude values ​​of the five most recently measured test channels all fall within the middle 80% interval between the current A_min and A_max; the middle 80% interval refers to the interval defined by shifting the current minimum amplitude value A_min upwards by 10% of the extreme value range length to the current maximum amplitude value A_max downwards by 10% of the extreme value range length, i.e., [A_min + 0.1 × (A_max - A_m)]. [θ_min + 0.1 × (θ_max - θ_min)]; The phase values ​​of the five most recently measured test channels all fall within the middle 80% range between the current θ_min and θ_max; the middle 80% range refers to the interval defined by shifting the current phase minimum value θ_min upwards by 10% of the extreme value range to the current phase maximum value θ_max downwards by 10% of the extreme value range, i.e., [θ_min + 0.1 × (θ_max - θ_min), θ_max - 0.1 × (θ_max - θ_min)].

[0146] Once the extreme value range converges, the automatic measurement and control system iterates through the remaining untested test channels. For each remaining untested test channel, its comprehensive extreme value probability P[i] is queried. When P[i] < 0.05 (i.e., the 5% screening threshold), the automatic measurement and control system classifies the test channel as an extreme value-independent channel and skips the actual measurement. The extreme value probabilities of these test channels are extremely low, and even if their actual values ​​fluctuate, they are highly unlikely to exceed the currently determined extreme value range, thus not affecting the final amplitude balance and phase balance calculation results. For skipped channels, the automatic measurement and control system uses their historical average value as an estimate, or directly ignores their participation in the balance calculation.

[0147] Optionally, the automatic measurement and control system can establish mutual exclusion relationships between test channels based on the extreme value probability model. A mutual exclusion threshold of 0.2 (i.e., 20%) is set, and test channels with a comprehensive extreme value probability P[i] ≥ 0.2 are marked as high-probability channels. The mutual exclusion relationship is defined as follows: any two high-probability test channels should not simultaneously belong to the same measurement group. This mutual exclusion relationship is applied during the grouping process in the group measurement mode. When executing the group measurement in S1023, the automatic measurement and control system selects channels sequentially to construct measurement groups according to the current measurement order. During the construction process, if the next candidate channel has a mutual exclusion relationship with an existing channel in the current test group (i.e., both are high-probability channels), the test channel is skipped, and the system continues to select the next channel until the number of channels in the group reaches the current group size or there are no more selectable channels. Through the constraint of the mutual exclusion relationship, the automatic measurement and control system ensures that high-probability channels are evenly distributed among different measurement groups, reducing the risk of multiple high-probability channels being retested simultaneously due to anomalies in a single group.

[0148] After the main loop ends, the automatic measurement and control system obtains the true transmission coefficients T_i for all test channels. Each T_i contains an amplitude value A_i and a phase value θ_i. The automatic measurement and control system calculates the amplitude balance: Amplitude balance = A_max - A_min, in decibels. The automatic measurement and control system calculates the phase balance: Phase balance = θ_max - θ_min, in degrees. The automatic measurement and control system outputs the calculated amplitude balance and phase balance as key performance indicators of the passive device under test for subsequent quality assessment or data analysis.

[0149] Optionally, the automatic measurement and control method may also include the following steps:

[0150] Regularly perform self-tests on the reference channel and compare the reference measurement values ​​of the reference channel with the known nominal values ​​of the calibration components;

[0151] When the comparison deviation exceeds the predetermined threshold, the reference channel is determined to be abnormal, and an alarm message is output.

[0152] In practice, to ensure the reliability of the entire testing system and prevent the correction benchmark from failing due to performance degradation or malfunction of the reference channel itself, the automatic measurement and control system periodically performs a self-test procedure on the reference channel.

[0153] The automatic measurement and control system has an internal self-test timer, and the self-test cycle can be set according to the importance of the test task and the system stability requirements. In this embodiment, the self-test cycle is set to 24 hours, meaning that a self-test is automatically triggered every 24 hours. Users can also manually trigger the self-test through the control interface. When the self-test is triggered, the automatic measurement and control system pauses the current test task (if any) and performs the following steps:

[0154] The automatic measurement and control system controls the two-way mechanical switch to switch to the second output terminal, allowing the test signal to be directly connected to the reference calibration channel of the RF switch matrix, and simultaneously controls the RF switch matrix to select the reference calibration channel. This state is consistent with the state of obtaining the reference measurement value during normal testing. The automatic measurement and control system controls the RF switch matrix to switch the input terminal of the reference calibration channel to the state of connecting to a known calibration component. The calibration component is a metrologically calibrated standard component with a known nominal transmission coefficient S_cal, including the nominal amplitude value A_cal and the nominal phase value θ_cal. The calibration component can be any one of a through-circuit calibration component, an open-circuit calibration component, a short-circuit calibration component, or a load calibration component. In this embodiment, a through-circuit calibration component is preferred, with a nominal transmission coefficient of 1∠0° (i.e., amplitude 0dB, phase 0°). The automatic measurement and control system controls the vector network analyzer to output the test signal through its first port and receive the signal returned from the reference calibration channel and the calibration component through its second port, measuring the transmission coefficient S_meas in the current state. S_meas is a complex number, including the measured amplitude A_meas and the measured phase θ_meas.

[0155] The automatic measurement and control system calculates the deviation between the measured value of the reference channel and the nominal value of the calibration component: amplitude deviation ΔA = |A_meas - A_cal|, in dB; phase deviation Δθ = |θ_meas - θ_cal|, in degrees; since this embodiment uses a through calibration component, A_cal = 0 dB, θ_cal = 0°, therefore ΔA = |A_meas|, Δθ = |θ_meas|.

[0156] The automatic measurement and control system compares the calculated deviation with a preset threshold. The preset threshold is determined based on the system accuracy requirements and the accuracy of the calibration components. In this embodiment, the amplitude deviation threshold is set to 0.1 dB, and the phase deviation threshold is set to 1 degree. These thresholds can be adjusted according to actual test requirements. For example, for high-precision testing, the thresholds can be narrowed to 0.05 dB and 0.5 degrees. When ΔA ≤ 0.1 dB and Δθ ≤ 1 degree, the reference channel is considered to be working normally, and the self-test passes. The automatic measurement and control system records the self-test result, resumes the test task (if any), and continues normal operation. When ΔA > 0.1 dB or Δθ > 1 degree, the reference channel is considered abnormal. The automatic measurement and control system performs the following abnormality handling: displays alarm information on the control interface, including the abnormality type (amplitude or phase exceeding limits), the measured deviation value, and the occurrence time; records the abnormality information to the system log for subsequent fault analysis; if the system has a backup reference channel, it automatically switches to the backup reference channel; if no backup reference channel is available, the automatic measurement and control system suspends all test tasks and waits for manual intervention.

[0157] The reference channel transmission coefficient S_meas recorded during the self-test can be used to correct system drift calculations in subsequent tests. The automatic measurement and control system compares S_meas with the reference channel baseline value S_ref_base recorded during initial calibration, calculates the drift of the reference channel itself, and subtracts this drift in subsequent tests to further improve test accuracy. In addition to timed self-tests, the automatic measurement and control system also supports the following triggering methods: User manual triggering: Click the "Self-test" button on the control interface to immediately execute the self-test; Power-on self-test: The system automatically performs a self-test every time it is powered on; Automatic anomaly triggering: When abnormal fluctuations occur in the differences between multiple consecutive reference measurements, the self-test is automatically triggered to confirm the status of the reference channel.

[0158] This application also discloses a fully automated measurement and control system for passive devices with self-calibration and dynamic compensation. (Refer to...) Figure 2 ,include:

[0159] The signal flow control module 201 is used to input the test signal into the test path and control the preset switch matrix to cooperate with the conduction of the test path; wherein, the test path includes a first signal path and a second signal path, the first signal path is used to allow the test signal to enter the preset test channel in the switch matrix after passing through the passive device under test; the second signal path is used to allow the test signal to directly enter the preset reference channel in the switch matrix.

[0160] The test result correction module 202 is used to obtain the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel for each test channel, and to use the difference to correct the actual measurement value output when the test signal enters the test channel.

[0161] The performance index calculation module 203 is used to calculate the performance index of the passive device under test based on the corrected actual measurement value. The performance index includes at least amplitude balance and phase balance.

[0162] This application also discloses a fully automatic measurement and control device for passive devices with self-calibration and dynamic compensation. The fully automatic measurement and control device for passive devices with self-calibration and dynamic compensation includes a memory and a processor. The memory stores a computer program that can be loaded by the processor and executed as described above for the fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation.

[0163] This application also discloses a computer-readable storage medium that stores a computer program that can be loaded by a processor and executed as described above for a fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation. The computer-readable storage medium includes, for example, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0164] It should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0165] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit the scope of protection of the application. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on these embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

Claims

1. A fully automated measurement and control method for passive devices with self-calibration and dynamic compensation, characterized in that, include: The test signal is input into the test path, and a preset switch matrix is ​​controlled to cooperate with the conduction of the test path; wherein, the test path includes a first signal path and a second signal path, the first signal path is used for the test signal to be connected to a preset test channel in the switch matrix after passing through the passive device under test; the second signal path is used for the test signal to be directly connected to a preset reference channel in the switch matrix. For each test channel, the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel is obtained, and the actual measurement value output when the test signal enters the test channel is corrected by the difference. The performance indicators of the passive device under test are calculated based on the corrected actual measurement values. The performance indicators include at least amplitude balance and phase balance. For each test channel, the step of acquiring the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel, and using the difference to correct the actual measurement value output by the test signal when entering the test channel, includes: The test signal is output through the reference channel to obtain the initial reference measurement value; Whenever the reference measurement value is updated and there are undetected test channels, the difference in future reference measurement values ​​is predicted based on the updated reference measurement value, and the measurement mode is determined based on the difference in value; wherein, the measurement mode is a single-channel measurement mode or a group measurement mode; According to the determined test mode, the test channel to be tested is selected from the untested test channels, the actual measurement value of the test signal when it passes through the test channel is obtained, and after the test channel completes the measurement, the test signal is passed through the reference channel again and the reference measurement value is obtained to update the reference measurement value. Whenever the reference measurement value is updated, the difference between the reference measurement value before and after the update is calculated. When the difference exceeds a preset threshold, the corresponding test channel before and after the update is re-registered as an undetected test channel. If the difference does not exceed the preset threshold, the actual measurement value of the corresponding test channel before and after the update is corrected using the difference. The corrected actual measurement values ​​include actual amplitude and actual phase; The step of reclassifying the corresponding tested channels before and after the update as untested test channels includes: Maintain in real time the maximum and minimum values ​​of the actual amplitude, as well as the maximum and minimum values ​​of the actual phase, for all currently tested test channels. The test channel corresponding to the difference between the reference measurement values ​​before and after the update exceeds a preset threshold is taken as the test channel to be estimated. Based on the actual measurement value and the difference of the test channel to be estimated during the measurement, the estimated amplitude and estimated phase after the correction of the test channel to be estimated are calculated. When the estimated amplitude is not between the maximum and minimum values ​​of the currently determined actual amplitude, and the estimated phase is not between the maximum and minimum values ​​of the currently determined actual phase, the corresponding test channel to be estimated is re-designated as an undetected test channel to await re-detection. The step of reclassifying the corresponding test channels before and after the update as untested test channels also includes: Based on the amplitude and phase value sequences of all currently measured test channels, predict the evolution direction of amplitude extrema and phase extrema respectively; The amplitude-side dynamic threshold and the phase-side dynamic threshold are determined according to the evolution direction, respectively; wherein, the dynamic threshold is negatively correlated with the number of currently measured test channels and positively correlated with the evolution rate of extreme values; For the test channel whose estimated amplitude is between the maximum and minimum values ​​of the currently determined actual amplitude, when the difference between the estimated amplitude of the test channel and the maximum value of the currently determined actual amplitude is less than the dynamic threshold on the side of the maximum amplitude, or the difference between the estimated amplitude and the minimum value of the currently determined actual amplitude is less than the dynamic threshold on the side of the maximum amplitude, the corresponding test channel to be estimated will be re-registered as an undetected test channel to wait for re-detection. For the test channel whose estimated phase is between the maximum and minimum values ​​of the currently determined actual amplitude, when the difference between the estimated phase and the maximum value of the currently determined actual phase is less than the dynamic threshold on the maximum phase value side, or the difference between the estimated phase and the minimum value of the currently determined actual phase is less than the dynamic threshold on the minimum phase value side, the corresponding test channel will be re-designated as an undetected test channel to await re-detection.

2. The fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation according to claim 1, characterized in that, The method further includes: An extreme value probability model is established, which records the probability that each test channel becomes an amplitude extreme value or a phase extreme value in historical test data; the extreme value refers to the maximum or minimum value. At the start of the test, the initial measurement order is determined according to the extreme value probability model. During the test, the measurement order of the remaining untested test channels is dynamically adjusted according to the actual measurement values ​​of the measured channels. Furthermore, when performing the step of selecting the test channel from the untested test channels, the selection shall be made in accordance with the latest measurement order; When the extreme ranges of amplitude and phase of the measured channels converge to within the preset accuracy, the test channels with extreme probability lower than the preset screening threshold in the remaining undetected test channels are judged as extreme value-irrelevant channels and the measurement is skipped.

3. The fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation according to claim 2, characterized in that, The method further includes: Based on the extreme value probability model, a mutual exclusion relationship is established between test channels. The mutual exclusion relationship is used to indicate that measurement channels with extreme value probabilities higher than a preset mutual exclusion threshold should not be simultaneously assigned to the same measurement group. When performing the step of selecting the test channel to be tested from the untested test channels, if the current measurement mode is grouped, the composition of the current measurement group is determined according to the mutual exclusion relationship, so that test channels with mutual exclusion relationships are not selected into the same measurement group at the same time.

4. The fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation according to claim 1, characterized in that, The method further includes: The reference channel is periodically self-tested, and the reference measurement value of the reference channel is compared with the known nominal value of the calibration component; When the comparison deviation exceeds the predetermined threshold, the reference channel is determined to be abnormal, and an alarm message is output.

5. A fully automatic measurement and control system for passive devices with self-calibration and dynamic compensation, applied to the fully automatic measurement and control method for passive devices with self-calibration and dynamic compensation as described in claim 1, characterized in that, include, The signal flow control module (201) is used to input the test signal into the test path and control the preset switch matrix to cooperate with the conduction of the test path; wherein, the test path includes a first signal path and a second signal path, the first signal path is used to allow the test signal to be connected to a preset test channel in the switch matrix after passing through the passive device under test; the second signal path is used to allow the test signal to be directly connected to a preset reference channel in the switch matrix. The test result correction module (202) is used to obtain, for each test channel, the difference between the reference measurement value output by the test signal before and after entering the test channel and when the test signal passes through the reference channel, and to use the difference to correct the actual measurement value output by the test signal when entering the test channel; The performance index calculation module (203) is used to calculate the performance index of the passive device under test based on the corrected actual measurement value. The performance index includes at least amplitude balance and phase balance.

6. A fully automatic measurement and control device for passive devices with self-calibration and dynamic compensation, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executed as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer program is stored that can be loaded by a processor and executed as described in any one of claims 1 to 4.

Images