阅读说明：本技术 一种基于十项误差模型的矢量网络分析仪的自校准方法 (Self-calibration method of vector network analyzer based on ten-term error model ) 是由 丁旭 *** 于 2020-06-05 设计创作，主要内容包括：本发明公开了一种基于十项误差模型的矢量网络分析仪的自校准方法,具体包括如下步骤：101)初始化步骤、102)校准标准件参数设置步骤、103)校准标准件参数采集步骤、104)修正步骤、105)变换处理步骤、106)计算得出相应参数步骤、107)测试步骤；本发明提供一种基于十项误差模型的矢量网络分析仪的自校准方法,其只需要校准标准件Thru的延时τ与插入损耗IL及负载标准件Match(Load)的直流电阻RM,其它校准标准件参数在校准过程中自动计算出来。(The invention discloses a self-calibration method of a vector network analyzer based on a ten-term error model, which specifically comprises the following steps: 101) initializing, 102) calibrating standard component parameter setting, 103) calibrating standard component parameter acquisition, 104) correcting, 105) converting, 106) calculating to obtain corresponding parameters, and 107) testing; the invention provides a self-calibration method of a vector network analyzer based on a ten-term error model, which only needs to calibrate the delay tau of a standard element Thru, the insertion loss IL and the direct current resistance RM of a load standard element match (load), and other calibration standard element parameters are automatically calculated in the calibration process.)

1. A self-calibration method of a vector network analyzer based on a ten-term error model is characterized by comprising the following steps:

101) an initialization step: setting an initial state of a vector network analyzer, wherein the initial state comprises a start-stop frequency point, a frequency step, output power, a medium frequency bandwidth and average times;

102) and (3) calibrating standard part parameter setting: setting the necessary pre-parameters of the calibration standard, including: the direct delay and insertion loss are matched with the direct current resistance of the load, and the direct delay and insertion loss are used for obtaining a calibration plane correction matrix;

103) a calibration standard part parameter acquisition step: connecting single-port calibration standard parts in sequence: reflecting Open circuit, reflecting Short circuit, matching Load and direct parameter acquisition of a dual-Port calibration standard component to obtain Gamma _ Measure _ Open _ Port, Gamma _ Measure _ Short _ Port, Gamma _ Measure _ Load _ Port, ISO _ F, ISO _ R, SMeasure _ Thru and Gamma _ F, Gamma _ R;

104) a correction step of correcting Gamma _ F, Gamma _ R, SMeasure _ Thru by ISO _ F, ISO _ R to obtain Г_Fic、Г_Ric、SMeasure_Thru_icISO _ F, ISO _ R, Г_Fic、Г_RicCorrection SMeasure _ Thru_icSMeasure _ Thru is obtained_sc；

105) A transformation processing step: using SMeasure _ Thru_scProcessing the corresponding data of the steps 103) and 104) in a mode of matrix transformation of the relation between the measured scattering parameters and the scattering transmission parameters, and performing Gamma _ Measure _ Open _ Port, Gamma _ Measure _ Load _ Port and SMeasure _ Thru on the data_scAnd supposing that Gamma _ Load is 0, calculating to obtain Gamma _ Open through an LRM + (TRM +) (Line/Thru-reflex-Match) algorithm and an automatic root-finding algorithm, calculating to obtain the inductance value of the Load standard component through the Gamma _ Open and the direct current resistance, and recalculating to obtain Gamma _ Load through the direct current resistance and the inductance value of the Load standard component;

calculating Gamma _ Short by using Gamma _ Measure _ Short _ Port, Gamma _ Measure _ Load _ Port, SMeasure _ Thrusc and Gamma _ Load through LRM + algorithm and automatic root-finding algorithm;

106) calculating to obtain corresponding parameters: calculating data in steps 103), 104) and 105) to obtain a forward directional item, a reverse directional item, a forward source matching, a reverse source matching, a forward reflection tracking item, a reverse reflection tracking item and a transmission error item;

107) testing the device under test to ISO _ F, ISO _ R, Г_Fic、Г_RicCorrecting original test value SDUT to obtain SDUT_SCConverting it into a form of scattering transmission parameters TDUT_SCConstructing an error correction matrix by the forward directional item, the reverse directional item, the forward source matching, the reverse source matching, the forward reflection tracking item, the reverse reflection tracking item and the transmission error item obtained in the step 106), and correcting the TDUT_SCObtaining TDUT_CRevising TDUT by using the calibration plane revision matrix in the step 101)_CObtaining the true value TDUT of the device under test_ACTWill TDUT_ACTConversion from a scattering transmission parametric form to a scattering parametric form SDUT_ACT；

Will SDUT_ACTAnd outputting and storing the test result in the s2p format, and displaying the actual value of the tested device in a graphical interface.

2. The self-calibration method of the vector network analyzer based on the ten-term error model as claimed in claim 1, wherein: step 103) parameters specifically collected:

301) gamma _ Measure _ Open _ Port, i.e., Γ MOX: when two ports to be calibrated of the vector network analyzer are connected with the reflection Open circuit, the respective reflection coefficients are measured;

302) gamma _ Measure _ Short _ Port, i.e., Γ MSX: when two ports to be calibrated of the vector network analyzer are connected with a reflection Short, the respective reflection coefficients are measured;

303) gamma _ Measure _ Load _ Port, i.e., Γ MLX: when two ports to be calibrated of the vector network analyzer are connected with a load Match, measuring respective reflection coefficients;

304) ISO _ F: when two ports to be calibrated of the vector network analyzer are connected with matched load Match, the forward transmission coefficient is measured;

305) ISO _ R: when two ports to be calibrated of the vector network analyzer are connected with matched load Match, the measured reverse transmission coefficient is obtained;

306) SMeasure _ Thru: when two ports to be calibrated of the vector network analyzer are connected with a straight-through standard component Thru, the measured scattering parameters are obtained;

307) gamma _ F, i.e., F: when two ports to be calibrated of the vector network analyzer are connected with a straight-through standard component Thru, the measured forward switch item;

308) gamma _ R, R: and when the two ports to be calibrated of the vector network analyzer are connected with the straight-through standard element Thru, the measured reverse switch item is obtained.

Technical Field

The invention relates to the field of radio frequency microsystems, in particular to a self-calibration method of a vector network analyzer based on a ten-term error model.

Background

The S parameter (scattering parameter) is the most common and important test parameter in the field of radio frequency microwave, a vector network analyzer needs to be used for measurement, the vector network analyzer is different from other instruments, the vector network analyzer needs to be calibrated before testing to correct system errors, and the quality of a calibration method determines the accuracy of a calibration result and finally directly influences the quality of the test result. The calibration type is generally divided into three forms of coaxial, on-chip and waveguide according to the type of a test port; the calibration process is a post-processing mathematical operation process, necessary preposed data is obtained by connecting the test calibration standard component, an error matrix is obtained by a corresponding calibration method, and in the real test process, the original test data is corrected by the error matrix through mathematical operation to finally obtain the real data of the tested component.

Common calibration methods are: a Short-Open-Load-Thru (SOLT) or TOSM (Thr-Open-Short-Match) method based on 12 item error models of the three-receiver architecture vector network analyzer, a TRL (Thru-Reflect-Line) method based on 8 item error models of the four-receiver architecture vector network analyzer and the like.

The most common SOLT (TOSM) method is simple to operate, has high dependence on the precision of a calibration standard part, and the parameters of the calibration standard part must be completely and accurately defined, the precision of the calibration standard part generally gives three-order model parameters, but the model parameters are gradually out of alignment along with the rise of frequency, the high-frequency test precision is poor, and the abrasion of each connection brings certain deviation to the model parameters, so that the method is not suitable for high-frequency high-precision test; other common TRL (Thru-reflector-Line) methods, although having a small accuracy dependency on calibration standards and high accuracy, are limited by the frequency range of the test, which requires the ratio of start-stop frequencies to be less than 1: 8, when the frequency is very low (generally less than or equal to 1GH), the Line of the transmission Line is too long in size and difficult to manufacture and use, and meanwhile, the method has high requirements on the operation of personnel and is very easy to damage and limit the use range.

Disclosure of Invention

The invention overcomes the defects of the prior art, and provides a vector network analyzer eLRRM + (eLMRR +/eTMRR +/eTRRM +) (ehenated-Line/Thru-reflector-Open-reflector-Short-Match-Plus) self-calibration method based on 10 error models, which combines the convenience and the convenience of SOLT operation, is similar to a TRL method, has low dependence on the parameters of a calibrator and is not limited by the frequency bandwidth to be tested; and on the basis of the compatibility of common 8-term error models, the influence of an isolation term on a test result is additionally considered, and the 10-term error model is changed into the 10-term error model, so that the precision of a calibration result is further improved. The method only needs to calibrate the delay tau and the insertion loss IL of the standard element Thru and the direct current resistance RM of the load standard element match (load), and the parameters of other calibration standard elements are automatically calculated by adopting a conventional processing mode in the calibration process.

The technical scheme of the invention is as follows:

a self-calibration method of a vector network analyzer based on a ten-term error model specifically comprises the following steps:

101) an initialization step: setting an initial state of a vector network analyzer, wherein the initial state comprises a start-stop frequency point, a frequency step, output power, a medium frequency bandwidth and average times;

102) and (3) calibrating standard part parameter setting: setting the necessary pre-parameters of the calibration standard, including: calculating a calibration plane correction matrix according to the through delay and insertion loss and the direct current resistance matched with the load;

103) a calibration standard part parameter acquisition step: two ports to be calibrated of the vector network analyzer are sequentially connected with a single-port calibration standard component: reflecting Open circuit, reflecting Short circuit, matching Load, calibrating a standard component with double ports, and directly acquiring parameters to obtain Gamma _ Measure _ Open _ Port, Gamma _ Measure _ Short _ Port, Gamma _ Measure _ Load _ Port, ISO _ F, ISO _ R, SMeasure _ Thru and Gamma _ F, Gamma _ R;

104) a correction step of correcting Gamma _ F, Gamma _ R, SMeasure _ Thru by ISO _ F, ISO _ R to obtain Г_Fic、Г_Ric、SMeasure_Thru_icISO _ F, ISO _ R, Г_Fic、Г_RicCorrection SMeasure _ Thru_icSMeasure _ Thru is obtained_sc；

105) A transformation processing step: using SMeasure _ Thru_scProcessing the corresponding data of the steps 103) and 104) in a mode of matrix transformation of the relation between the measured scattering parameters and the scattering transmission parameters, and performing Gamma _ Measure _ Open _ Port, Gamma _ Measure _ Load _ Port and SMeasure _ Thru on the data_scAnd assuming Gamma _ Load is 0, calculated by LRM + (TRM +) (Line/Thru-reflex-Match) algorithm and auto-root-finding algorithmObtaining Gamma _ Open, obtaining the inductance value of the Load standard component through calculation of direct current resistance of the Gamma _ Open, and obtaining Gamma _ Load through recalculation of the direct current resistance and the inductance value of the Load standard component;

calculating Gamma _ Short by using Gamma _ Measure _ Short _ Port, Gamma _ Measure _ Load _ Port, SMeasure _ Thrusc and Gamma _ Load through LRM + algorithm and automatic root-finding algorithm;

106) calculating to obtain corresponding parameters: calculating data in steps 103), 104) and 105) to obtain a forward directional item, a reverse directional item, a forward source matching, a reverse source matching, a forward reflection tracking item, a reverse reflection tracking item and a transmission error item;

107) testing the device under test to ISO _ F, ISO _ R, Г_Fic、Г_RicCorrecting original test value SDUT to obtain SDUT_SCConverting it into a form of scattering transmission parameters TDUT_SCConstructing an error correction matrix by the forward directional item, the reverse directional item, the forward source matching, the reverse source matching, the forward reflection tracking item, the reverse reflection tracking item and the transmission error item obtained in the step 106), and correcting the TDUT_SCObtaining TDUT_CRevising TDUT by using the calibration plane revision matrix in the step 101)_CObtaining the true value TDUT of the device under test_ACTWill TDUT_ACTConversion from a scattering transmission parametric form to a scattering parametric form SDUT_ACT；

Will SDUT_ACTAnd outputting and storing the test result in the s2p format, and displaying the actual value of the tested device in a graphical interface.

Further, step 103) specifically acquiring parameters:

301) gamma _ Measure _ Open _ Port, i.e., Γ MOX: when two ports to be calibrated of the vector network analyzer are connected with the reflection Open circuit, the respective reflection coefficients are measured;

302) gamma _ Measure _ Short _ Port, i.e., Γ MSX: when two ports to be calibrated of the vector network analyzer are connected with a reflection Short, the respective reflection coefficients are measured;

303) gamma _ Measure _ Load _ Port, i.e., Γ MLX: when two ports to be calibrated of the vector network analyzer are connected with matched loads match (load), measuring respective reflection coefficients;

304) ISO _ F: when two ports to be calibrated of the vector network analyzer are connected with matched loads match (load), the measured forward transmission coefficient is obtained;

305) ISO _ R: when two ports to be calibrated of the vector network analyzer are connected with matched loads match (load), the measured reverse transmission coefficient is obtained;

306) SMeasure _ Thru: when two ports to be calibrated of the vector network analyzer are connected with a straight-through standard component Thru, the measured scattering parameters are obtained;

307) gamma _ F, i.e., F: when two ports to be calibrated of the vector network analyzer are connected with a straight-through standard component Thru, the measured forward switch item;

308) gamma _ R, R: and when the two ports to be calibrated of the vector network analyzer are connected with the straight-through standard element Thru, the measured reverse switch item is obtained.

Compared with the prior art, the invention has the advantages that: the method combines the convenience and the rapidness of the SOLT operation, has low dependence on the parameters of the calibration piece similar to the TRL method, and is not limited by the test frequency bandwidth; and on the basis of the compatibility of common 8-term error models, the influence of an isolation term on a test result is additionally considered, and the 10-term error model is changed into the 10-term error model, so that the precision of a calibration result is further improved. The method only needs to calibrate the delay tau and the insertion loss IL of the standard element Thru and the direct current resistance RM of the load standard element match (load), and other calibration standard element parameters are automatically calculated in the calibration process.

The method is compatible with the 8-term error model, and on the basis, the influence of the isolation term on the high-frequency S parameter calibration of the vector network analyzer is considered, and the 8-term error model is improved into the 10-term error model by adding the isolation correction term. The influence of the signal coupling leakage between the ports on the test result can be well corrected in the high-frequency test (more than or equal to 60GHz) after the isolation item is considered.

The improved 10-term error model calibration operation steps are completely consistent with the common SOLT calibration method based on the 12-term error model, but the method is a self-calibration method, namely the dependence on the parameters of the calibration standard is low, and the related parameters of the calibration standard are automatically calculated by the calibration processTherefore, the accuracy and the application range of calibration are greatly improved. The method only needs 3 preposed parameters: DC resistance R of load standard component match (load)_MThe delay tau and the insertion loss IL of the straight-through (Thru/Line) greatly reduce the manufacturing difficulty of the calibration standard.

The invention is a broadband calibration method, the calibration process is not limited by the frequency range, different calibration standard parts are not required to be replaced for different frequencies, and the TRL method does not need a plurality of sections of transmission lines to deal with different frequencies. Especially for on-chip calibration applications, significant area savings and significant cost reductions are possible.

Drawings

FIG. 1 is a signal flow diagram of a 10-term error model of the present invention;

fig. 2 is a main flow chart of the calibration method of the present invention.

Detailed Description

The following is combined with

The figures and the detailed description further illustrate the invention.

As shown in fig. 1 and fig. 2, a self-calibration method of a vector network analyzer based on a ten-term error model specifically includes the following steps:

101) an initialization step: setting an initial state of a vector network analyzer, wherein the initial state comprises a start-stop frequency point, a frequency step, output power, a medium frequency bandwidth and average times;

102) and (3) calibrating standard part parameter setting: setting the necessary pre-parameters of the calibration standard, including: calculating a calibration plane correction matrix according to the direct delay and insertion loss and the direct current resistance of the load standard component;

103) a calibration standard part parameter acquisition step: two ports to be calibrated of the vector network analyzer are sequentially connected with a single-port calibration standard component: reflecting Open circuit, reflecting Short circuit, matching Load and direct parameter acquisition of a dual-Port calibration standard component to obtain Gamma _ Measure _ Open _ Port, Gamma _ Measure _ Short _ Port, Gamma _ Measure _ Load _ Port, ISO _ F, ISO _ R, SMeasure _ Thru and Gamma _ F, Gamma _ R;

parameters specifically collected:

301) gamma _ Measure _ Open _ Port, i.e., Γ MOX: when two ports to be calibrated of the vector network analyzer are connected with Open, respective reflection coefficients are measured;

302) gamma _ Measure _ Short _ Port, i.e., Γ MSX: when two ports to be calibrated of the vector network analyzer are connected with Short, measuring respective reflection coefficients;

303) gamma _ Measure _ Load _ Port, i.e., Γ MLX: when two ports to be calibrated of the vector network analyzer are connected with Match, measuring respective reflection coefficients;

304) ISO _ F: when two ports to be calibrated of the vector network analyzer are connected with Match, the forward transmission coefficient is measured;

305) ISO _ R: when two ports to be calibrated of the vector network analyzer are connected with Match, the measured reverse transmission coefficient is obtained;

306) SMeasure _ Thru: when two ports to be calibrated of the vector network analyzer are connected with Thru, the measured scattering parameters are obtained;

307) gamma _ F, i.e., F: when two ports to be calibrated of the vector network analyzer are connected with Thru, the measured forward switch item;

308) gamma _ R, R: and when the two ports to be calibrated of the vector network analyzer are connected with Thru, the measured reverse switching term is obtained.

104) A correction step of correcting Gamma _ F, Gamma _ R, SMeasure _ Thru by ISO _ F, ISO _ R to obtain Г_Fic、Г_Ric、SMeasure_Thru_icISO _ F, ISO _ R, Г_Fic、Г_RicCorrection SMeasure _ Thru_icSMeasure _ Thru is obtained_sc。

105) A transformation processing step: using SMeasure _ Thru_scAnd processing corresponding data of the steps 103) and 104) in a mode of matrix transformation of the relation between the measured scattering parameters and the scattering transmission parameters, calculating Gamma _ Open by using a Gamma _ Measure _ Open _ Port, a Gamma _ Measure _ Load _ Port, a SMeasure _ Thrusc and an assumed Gamma _ Load ═ 0 through an LRM + (TRM +) (Line/Thru-reflex-Match) algorithm and an automatic root-finding algorithm, calculating a negative standard component inductance value by using the Gamma _ Open and a direct current resistance, and recalculating the Gamma _ Load by using the direct current resistance and the negative standard component inductance value. Therein, fromThe core process of the dynamic root-finding algorithm is as follows: determining a theoretical Phase when the frequency is 0Hz by reflection standard elements Open and Short, wherein PhaseOp |0Hz is 0 DEG, PhaseShort |0Hz is 180 DEG, obtaining two groups of Phase |0Hz through Phase folding-free, overturning-free and 0Hz continuation, searching an initial Phase true root according to the type of the actual reflection standard element, and then judging all true roots through an iterative process according to the adjacent frequency point folding-free Phase difference smaller than 180 deg.

With Gamma _ Measure _ Short _ Port, Gamma _ Measure _ Load _ Port, SMeasure _ Thru_scAnd the Gamma _ Load is calculated again through an LRM + algorithm (which is fully called as a vector network analyzer calibration algorithm Line-reflex-Match-Plus) and an automatic root-finding algorithm to obtain the Gamma _ Short.

106) Calculating to obtain corresponding parameters: calculating data in steps 103), 104) and 105) to obtain a forward directional item, a reverse directional item, a forward source matching, a reverse source matching, a forward reflection tracking item, a reverse reflection tracking item and a transmission error item;

107) testing the device under test to ISO _ F, ISO _ R, Г_Fic、Г_RicCorrecting the original Test value SDUT (uncorrected S parameter of the tested piece) to obtain the SDUT_SCConverting it into a form of scattering transmission parameters TDUT_SCConstructing an error correction matrix by the forward directional item, the reverse directional item, the forward source matching, the reverse source matching, the forward reflection tracking item, the reverse reflection tracking item and the transmission error item obtained in the step 106), and correcting the TDUT_SCObtaining TDUT_CRevising TDUT by using the calibration plane revision matrix in the step 101)_CObtaining the true value TDUT of the device under test_ACTWill TDUT_ACTConversion from a scattering transmission parametric form to a scattering parametric form SDUT_ACT；

Will SDUT_ACTAnd outputting and storing the test result in the s2p format, and displaying the actual value of the tested device in a graphical interface. The algorithm deduction process which is not specifically disclosed in the above is only needed by adopting a conventional processing method.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and decorations can be made without departing from the spirit of the present invention, and these modifications and decorations should also be regarded as being within the scope of the present invention.