On-chip calibration piece, error model generation method and system, and test method and system
Through the in-chip calibration parts and error model generation method, the error problem of RF differential probes for test results is solved, and accurate calibration of high-frequency and large bandwidth tests is achieved, which simplifies the design difficulty.
Patent Information
- Application Number
- CN202310588673.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-23
- Publication Date
- 2024-11-26
AI Technical Summary
Existing calibration methods, such as SOLT and TRL methods, cannot effectively remove the near-end crosstalk and far-end crosstalk errors of RF differential probes for test results, especially in high-frequency tests, and the calibration results are significantly deviated, and the TRL method cannot calibrate to the difference. The probe tip and delay line length are limited, making it difficult to meet the needs of large bandwidth testing.
A piece calibration part is designed, and multiple conductive parts are arranged on the substrate to form a variety of calibration structures, such as short circuit, open circuit and load structure. Combined with the error model generation method, the actual measured scattering parameter matrix is obtained through a vector network analyzer. and the intrinsic scattering parameter matrix, fit the target error model for calibration and testing.
It improves the accuracy of calibration results, can effectively remove errors of differential probes, is suitable for high-frequency and large bandwidth testing, and can be calibrated to the tip of differential probes, simplifying design difficulty.
Smart Images

Figure CN119024241A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of calibration technology, and in particular to on-wafer calibration components, error model generation methods and systems, and testing methods and systems. Background Art
[0002] Vector Network Analyzer (VNA) is a common instrument for measuring scattering parameters (s-parameter, S parameters). The S parameters measured by it can be used to analyze link information. Generally, VNA and the device under test (DUT) cannot be directly connected, and a transit connection structure such as a coaxial line or a coaxial line plus a differential probe is required. This part is called a fixture. The performance of the fixture itself needs to be removed from the test results. This process of removing the performance of the fixture itself is called de-embedding or calibration. On-wafer testing uses RF differential probes as fixtures. The test reference plane needs to be adjusted from the output port of the VNA to the tip of the differential probe to remove the influence of the differential probe itself on the test results.
[0003] The commonly used calibration methods at present are Short-Open-Load-Through (SOLT) method and Through-Reflect-Line (TRL) method, and the corresponding calibration parts used are also different. The SOLT method only uses four calibration parts: Figure 1a The short circuit calibration component having the short circuit calibration structure SG shown in FIG. Figure 1b The open calibration component with the open calibration structure OG shown in FIG. Figure 1c The load calibration member having the load calibration structure RG shown in FIG. Figure 1d The through calibration component with through calibration structure ZG is shown. The TRL method uses only three types of calibration components: through calibration component, reflective calibration component, and line calibration component.
[0004] However, the SOLT method ignores the near-end crosstalk and far-end crosstalk between the differential probes near the calibration reference plane. Since the distance between the differential probes is very close, this error cannot be ignored. With the decrease in the spacing between the differential probes and higher frequency testing, the error becomes more obvious, resulting in a large deviation in the calibration result.
[0005] In addition, the TRL method can only calibrate to the midpoint of the through-component, not the tip of the differential probe. Also, the length of the delay line calibration component is limited, and large bandwidth testing requires multiple delay lines of different lengths, otherwise there will be resonance problems. Summary of the invention
[0006] The embodiments of the present application provide an on-wafer calibration component, an error model generation method and system, and a testing method and system to improve the accuracy of calibration results.
[0007] In the first aspect, an embodiment of the present application provides an on-chip calibration part, comprising: a substrate and a metal layer disposed on the substrate. The metal layer comprises: a plurality of conductive parts, and by making the plurality of conductive parts constitute at least two of a short-circuit calibration structure, an open-circuit calibration structure, and a load calibration structure, a plurality of different calibration structures can be provided on an on-chip calibration part, thereby improving the accuracy of calibration. Also, the number of the plurality of conductive parts may not be greater than the total number of signal pins and ground pins of the differential probes connected to the vector network analyzer, and the signal pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of the plurality of conductive parts, and the ground pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of the plurality of conductive parts, and the vector network analyzer connects at least four differential probes. It is understandable that when a differential probe includes a plurality of ground pins, the plurality of ground pins may be connected to the same conductive part, or each ground pin may be connected one-to-one to one of the conductive parts.
[0008] In order to form different calibration structures, the multiple conductive parts can include: a first conductive part group and a second conductive part group arranged along a first direction, wherein the first conductive part group includes a plurality of first conductive parts corresponding to the signal pins and a plurality of second conductive parts corresponding to the ground pins; the second conductive part group includes: a plurality of third conductive parts corresponding to the signal pins and a plurality of fourth conductive parts corresponding to the ground pins. In addition, the first conductive part group and the second conductive part group can respectively form a plurality of different calibration structures. In other words, the plurality of first conductive parts and the plurality of second conductive parts can form a plurality of first calibration structures, and the plurality of first calibration structures include at least two of a short-circuit calibration structure, an open-circuit calibration structure, and a load calibration structure. In addition, the plurality of third conductive parts and the plurality of fourth conductive parts can form a plurality of second calibration structures, and the plurality of second calibration structures include at least two of a short-circuit calibration structure, an open-circuit calibration structure, and a load calibration structure.
[0009] For example, the spacings between adjacent conductive parts in the first conductive part group can be made the same. For example, the spacings between adjacent conductive parts in the second conductive part group can also be made the same.
[0010] Optionally, the spacing between adjacent conductive portions in the first conductive portion group may be made the same as the spacing between adjacent conductive portions in the second conductive portion group.
[0011] By making the plurality of first conductive parts correspond to the plurality of third conductive parts one by one, and arranging the first conductive parts and the third conductive parts corresponding to each other along the first direction, the first conductive parts and the third conductive parts corresponding to the signal pins can be arranged correspondingly.
[0012] In addition, the length of the first conductive portions arranged along the first direction may be the same as the length of the third conductive portions in the first direction.
[0013] By making the plurality of second conductive parts correspond to the plurality of fourth conductive parts one by one, and arranging the corresponding second conductive parts and fourth conductive parts along a set direction, the first conductive part and the third conductive part corresponding to the ground pin can be arranged correspondingly.
[0014] In addition, the length of the second conductive portions arranged along the first direction in the first direction may be the same as the length of the fourth conductive portions in the first direction.
[0015] For example, the lengths of the conductive parts corresponding to different signal pins in the first conductive part group may be different or the same. For example, the lengths of the conductive parts corresponding to different signal pins in the second conductive part group may be different or the same.
[0016] In order to simplify the design difficulty, the multiple first calibration structures can correspond to the multiple second calibration structures one by one, and the corresponding first calibration structures and second calibration structures are the same. In this way, the first calibration structures and the second calibration structures can be designed uniformly, simplifying the design difficulty.
[0017] In order to simplify the design difficulty, the first calibration structure and the second calibration structure corresponding to each other can be arranged in a mirror-symmetrical or center-symmetrical manner, thereby simplifying the design difficulty.
[0018] In order to form different calibration structures, the multiple first conductive parts and the multiple second conductive parts in the first conductive part group can also form multiple third calibration structures, and the multiple third calibration structures can include one of the short-circuit calibration structure, the open-circuit calibration structure and the load calibration structure, so that the conductive parts in the first conductive part group can communicate with the same calibration structure. And, the multiple third conductive parts and the multiple fourth conductive parts in the second conductive part group can form multiple fourth calibration structures, and the multiple fourth calibration structures can include at least one of the remaining calibration structures among the short-circuit calibration structure, the open-circuit calibration structure and the load calibration structure except the third calibration structure, so that the conductive parts in the second conductive part group can communicate with the same calibration structure or any two calibration structures.
[0019] In the second aspect, the embodiment of the present application also provides an on-chip calibration part, including: a substrate and a metal layer arranged on the substrate. Wherein, the metal layer includes: a plurality of conductive parts, by making the plurality of conductive parts constitute the same calibration structure, and the calibration structure includes one of a short-circuit calibration structure, an open-circuit calibration structure and a load calibration structure, the same calibration structure can be set on the on-chip calibration part. And, the number of the plurality of conductive parts may not be greater than the total number of signal pins and ground pins of the differential probes connected to the vector network analyzer, and the signal pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of the plurality of conductive parts, and the ground pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of the plurality of conductive parts, and the vector network analyzer connects at least four differential probes. It is understandable that when a differential probe includes a plurality of ground pins, the plurality of ground pins can be connected to the same conductive part, or each ground pin can be connected one-to-one to a conductive part.
[0020] In addition, by making the lengths of different conductive parts corresponding to different signal pins different, the calibration piece is arranged in a non-mirror symmetric manner in the first direction and / or the second direction, thereby improving the accuracy of calibration.
[0021] Exemplarily, the multiple conductive parts may include: a first conductive part group and a second conductive part group arranged along the first direction, wherein the first conductive part group includes a plurality of first conductive parts corresponding to the signal pins and a plurality of second conductive parts corresponding to the ground pins; the second conductive part group includes: a plurality of third conductive parts corresponding to the signal pins and a plurality of fourth conductive parts corresponding to the ground pins. In addition, the plurality of first conductive parts correspond to the plurality of third conductive parts one-to-one, and the first conductive parts corresponding to each other are arranged along the first direction, so that the first conductive parts and the third conductive parts corresponding to the signal pins can be arranged correspondingly. In addition, the plurality of second conductive parts correspond to the plurality of fourth conductive parts one-to-one, and the second conductive parts corresponding to each other are arranged along the first direction, so that the second conductive parts and the fourth conductive parts corresponding to the ground pins can be arranged correspondingly.
[0022] Exemplarily, the length of the first conductive portion along the first direction may be smaller than the length of the corresponding third conductive portion along the first direction.
[0023] Exemplarily, a plurality of first conductive portions may be arranged along the second direction, and the lengths of the plurality of first conductive portions arranged along the second direction in the first direction may increase sequentially; and / or a plurality of third conductive portions may be arranged along the second direction, and the lengths of the plurality of third conductive portions arranged along the second direction in the first direction may increase sequentially.
[0024] Exemplarily, the lengths of the conductive parts of the corresponding ground pins along the first direction may be the same, or different, or partially the same and partially different.
[0025] For example, the spacings between adjacent conductive parts in the first conductive part group may be the same, and the spacings between adjacent conductive parts in the second conductive part group may also be the same.
[0026] Optionally, the spacing between adjacent conductive portions in the first conductive portion group may be made the same as the spacing between adjacent conductive portions in the second conductive portion group.
[0027] In a third aspect, an embodiment of the present application further provides a method for generating an error model, comprising:
[0028] When the signal pin and the ground pin of the differential probe connected to the vector network analyzer are connected to multiple calibration parts for testing, the measured scattering parameter matrix corresponding to each calibration part is obtained. For example, multiple calibration parts are provided, and the signal pin and the ground pin of the differential probe connected to the vector network analyzer are connected to the multiple calibration parts in sequence, and the test signal is output to the connected calibration parts for testing. The vector network analyzer will test and obtain the measured scattering parameter matrix corresponding to each calibration part in the multiple calibration parts. Therefore, the measured scattering parameter matrix corresponding to each calibration part can be obtained from the vector network analyzer. In addition, the measured scattering parameter matrix is a matrix with N rows and N columns.
[0029] The measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration piece are input into the initial M*M error model for fitting, and the fitted target M*M error model is obtained, where M=2N, and N represents the total number of signal needles of all differential probes; the target M*M error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix. For example, the intrinsic scattering parameter matrix is a matrix with N rows and N columns.
[0030] In addition, the intrinsic scattering parameter matrix corresponding to each calibration piece can be obtained in advance through calibration piece measurement or simulation. Exemplarily, these intrinsic scattering parameter matrices can be stored in the controller or in other devices, and can be directly called when executing the method.
[0031] In order to obtain the target M*M error model, the measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration piece can be input into the initial M*M error model for fitting, and the fitted target M*M error model can be obtained, which includes the following process: First, the measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration piece are input into the initial M*M error model for fitting, and N characteristic coefficient matrices are solved. Then, the N characteristic coefficient matrices are substituted into the initial M*M error model to obtain the target M*M error model.
[0032] The target M*M error model can be made to conform to the following relationship:
[0033] SA=(TA-SM*TC)-1 *(SM*TD-TB);
[0034] Among them, SA represents the intrinsic scattering parameter matrix, SM represents the measured scattering parameter matrix, TA represents the first characteristic coefficient matrix among the N characteristic coefficient matrices, TB represents the second characteristic coefficient matrix among the N characteristic coefficient matrices, TC represents the third characteristic coefficient matrix among the N characteristic coefficient matrices, and TD represents the fourth characteristic coefficient matrix among the N characteristic coefficient matrices.
[0035] Exemplarily, the plurality of calibration components include: one or more of the on-chip calibration components of the first aspect or each embodiment of the first aspect, as well as a short-circuit calibration component, an open-circuit calibration component, a load calibration component, and a through calibration component.
[0036] In a fourth aspect, the present application also provides a testing method, including:
[0037] When the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to the device under test for testing, the measured scattering parameter matrix of the device under test is obtained, wherein the measured scattering parameter matrix corresponding to the device under test is a matrix with N rows and N columns.
[0038] The measured scattering parameter matrix of the device under test is input into the target M*M-term error model to obtain the calibrated intrinsic scattering parameter matrix of the device under test; the target M*M-term error model is obtained based on the error model generation method of the third aspect or various embodiments of the third aspect.
[0039] In addition, the corresponding relationship between the intrinsic scattering parameter matrix Sdut and the measured scattering parameter matrix Sdutm input into the target M*M error model is: Sdut = (TA-Sdutm*TC) -1 *(Sdutm*TD-TB), Sdutm represents the measured scattering parameter matrix corresponding to the device under test, and Sdut represents the intrinsic scattering parameter matrix corresponding to the device under test.
[0040] In a fifth aspect, an embodiment of the present application further provides an error model generation system, including: multiple calibration parts and a controller. Wherein, the controller is used to obtain the measured scattering parameter matrix corresponding to each calibration part when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to multiple calibration parts for testing. And, the measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration part are input into the initial M*M error model for fitting to obtain the fitted target M*M error model, M=2N, N represents the total number of signal pins of all differential probes; the target M*M error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix.
[0041] Exemplarily, the plurality of calibration components include: one or more of the on-chip calibration components in the first aspect or the embodiments in the first aspect, as well as a short-circuit calibration component, an open-circuit calibration component, a load calibration component, and a through calibration component.
[0042] In a sixth aspect, an embodiment of the present application further provides a test system, comprising: a vector network analyzer, at least four differential probes, and a controller. The at least four differential probes are connected to the vector network analyzer through corresponding coaxial lines, respectively. Furthermore, the controller is used to: when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to the device under test for testing, obtain the measured scattering parameter matrix of the device under test, input the measured scattering parameter matrix of the device under test into the target M*M term error model, and obtain the calibrated intrinsic scattering parameter matrix of the device under test.
[0043] The test system can also realize the calibration function, then the test system also includes: multiple calibration parts, the multiple calibration parts include: one or more of the on-chip calibration parts in the first aspect or each embodiment of the first aspect, as well as a short-circuit calibration part, an open-circuit calibration part, a load calibration part, and a through calibration part.
[0044] In addition, the controller is also used to: when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to different multiple calibration pieces for testing, obtain the measured scattering parameter matrix corresponding to each calibration piece. And, input the measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration piece into the initial M*M error model for fitting, and obtain the fitted target M*M error model, where M=2N, and N represents the total number of signal pins of all differential probes; the target M*M error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix.
[0045] In the seventh aspect, the embodiments of the present application also provide a computer storage medium, which stores executable instructions. When the executable instructions are executed by a computer, the error model generation method of the third aspect of the present application or the various embodiments in the third aspect is executed, or the testing method of the fourth aspect of the present application or the various embodiments in the fourth aspect is executed. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1a It is a schematic diagram of the structure of a short-circuit calibration component in the prior art;
[0047] Figure 1b It is a schematic diagram of the structure of an open-circuit calibration component in the prior art;
[0048] Figure 1c It is a structural schematic diagram of a load calibration component in the prior art;
[0049] Figure 1dIt is a structural schematic diagram of a straight-through calibration component in the prior art;
[0050] Figure 2 A schematic diagram of the structure of a test system in an embodiment of the present application;
[0051] Figures 3a to 3e They are schematic diagrams of the structure of an on-wafer calibration component in an embodiment of the present application;
[0052] Figures 4a to 4d They are schematic diagrams of the structure of an on-wafer calibration component in another embodiment of the present application;
[0053] Figure 5a to Figure 5c They are schematic diagrams of the structure of an on-wafer calibration component in another embodiment of the present application;
[0054] Figure 6 This is a flow chart of an error model generation method in one embodiment of the present application;
[0055] Figure 7 A flow chart of a testing method in one embodiment of the present application;
[0056] Figure 8 The figure is a schematic diagram of the structure of a test system in an embodiment of the present application during a test process.
[0057] Reference numerals:
[0058] 10-vector network analyzer; 20-differential probe; 21-signal pin; 22-ground pin; 30-coaxial line; 40-device under test; 100-substrate; 200-metal layer; 211a / 211b-first conductive part; 212a / 212b-second conductive part; 213a / 213b-third conductive part; 214a / 214b-fourth conductive part; h11 / h12 / h13 / h14 / h21 / h22 / h23 / h24-length; Z1-first conductive part group; Z2-second conductive part group; F1-first direction; F2-second direction; SG-short circuit calibration structure; OG-open circuit calibration structure; RG-load calibration structure; ZG-through calibration structure. DETAILED DESCRIPTION
[0059] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. In the embodiments of the present application, "connection" refers to electrical connection, and the connection between two electrical components can be a direct or indirect connection between the two electrical components. For example, A and B are connected, which can be either A and B directly connected, or A and B indirectly connected through one or more other electrical components, such as A and B are connected, or A and C are directly connected, C and B are directly connected, and A and B are connected through C. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application.
[0060] The on-wafer calibration component, error model generation method and system, and test method and system provided in the embodiments of the present application can be applied to calibration and test scenarios based on a vector network analyzer.
[0061] Figure 2 This is a schematic diagram of the structure of a test system in an embodiment of the present application. Figure 2 , the test system may include: a vector network analyzer 10 and four differential probes 20. The vector network analyzer 10 has four output ports, which correspond to the four differential probes 20 one by one, and the four differential probes 20 can be connected to the corresponding output ports in the vector network analyzer 10 through corresponding coaxial lines 30, respectively. In addition to the coaxial lines 30, the four differential probes 20 can also be connected to the corresponding output ports in the vector network analyzer 10 through other structural connecting lines, which is not limited in this application. This application only takes four differential probes as an example for explanation. In actual applications, there can also be five, six or more differential probes, which is not limited here.
[0062] Typically, the differential probe may include a signal pin and at least one ground pin. The signal pin is used to transmit a test signal, and the ground pin is connected to the ground. Optionally, the differential probe may include one, two, three, four or more ground pins, which are not limited here. This application takes the differential probe including a signal pin and a ground pin as an example.
[0063] When used, the signal pin and the ground pin of the differential probe 20 connected to the vector network analyzer 10 are respectively connected to the device under test 40 (DUT), and the test signal is input to the DUT 40 to obtain the scattering parameter (i.e., S parameter) matrix of the DUT 40. In order to remove the influence of the differential probe itself on the test result, it is necessary to adjust the test reference plane from the output port of the vector network analyzer to the tip of the signal pin of the differential probe to implement the calibration process. Usually, the calibration process is implemented by a calibration piece.
[0064] Figures 3a to 3eThey are schematic diagrams of the structure of an on-wafer calibration component in an embodiment of the present application. Figures 3a to 3e The on-chip calibration piece may include: a substrate 100 and a metal layer 200 disposed on the substrate 100. The metal layer 200 includes: a plurality of conductive parts (e.g., 211a, 211b, 212a, 212b, 213a, 213b, 214a, 214b), and the plurality of conductive parts constitute at least two of a short-circuit calibration structure SG, an open-circuit calibration structure OG, and a load calibration structure RG. This configuration allows a plurality of different calibration structures to be provided on a on-chip calibration piece, thereby improving the accuracy of calibration.
[0065] Reference Figures 3a to 3e , the total number of signal pins and ground pins of the differential probe connected to the vector network analyzer is 8, and the number of conductive parts is 8. Moreover, the 8 conductive parts correspond one-to-one to the signal pins and ground pins of each differential probe 20 connected to the vector network analyzer 10. During calibration, the signal pins and ground pins can be respectively connected to the corresponding conductive parts, and the test signal can be input for testing. It can be understood that when a differential probe 20 includes multiple ground pins, the multiple ground pins can be connected to the same conductive part, or each ground pin can be connected to a conductive part one-to-one. Therefore, the number of conductive parts can be set to be less than or equal to the total number of signal pins and ground pins of the differential probe 20 connected to the vector network analyzer 10.
[0066] In some examples, reference Figures 3a to 3e , the conductive parts in the metal layer 200 can be divided into two conductive parts groups arranged along the first direction F1: the first conductive part group Z1 and the second conductive part group Z2. The first conductive part group Z1 and the second conductive part group Z2 respectively include a plurality of conductive parts arranged along the second direction F2, and the first direction F1 and the second direction F2 are perpendicular. Exemplarily, the first conductive part group Z1 includes the first conductive parts 211a and 211b corresponding to the signal pins and the second conductive parts 212a and 212b corresponding to the ground pins. The second conductive part group Z2 includes the third conductive parts 213a and 213b corresponding to the signal pins and the fourth conductive parts 214a and 214b corresponding to the ground pins. The first conductive portion 211a corresponds to the third conductive portion 213a and is arranged along the first direction F1, the first conductive portion 211b corresponds to the third conductive portion 213b and is arranged along the first direction F1, the second conductive portion 212a corresponds to the fourth conductive portion 214a and is arranged along the first direction F1, and the second conductive portion 212b corresponds to the fourth conductive portion 214b and is arranged along the first direction F1. In addition, the first conductive portions 211a and 211b are arranged adjacent to each other, the second conductive portion 212a is arranged on the side of the first conductive portion 211a away from the first conductive portion 211b, and the second conductive portion 212b is arranged on the side of the first conductive portion 211b away from the first conductive portion 211a.
[0067] Exemplarily, the first conductive part group Z1 and the second conductive part group Z2 can respectively constitute a plurality of different calibration structures. For example, the first conductive parts 211a, 211b and the second conductive parts 212a, 212b can constitute a plurality of first calibration structures, and the third conductive parts 213a, 213b and the fourth conductive parts 214a, 214b can also constitute a plurality of second calibration structures. Moreover, the plurality of first calibration structures correspond one-to-one to the plurality of second calibration structures, and the first calibration structures and the second calibration structures corresponding to each other are the same. It is understandable that the plurality of first calibration structures and the plurality of second calibration structures can respectively include any two or three of the short-circuit calibration structure SG, the open-circuit calibration structure OG, and the load calibration structure RG.
[0068] In some examples, reference Figure 3a , the multiple first calibration structures and the multiple second calibration structures respectively include two types: short-circuit calibration structure SG and open-circuit calibration structure OG. Among them, the first conductive part 211a and the second conductive part 212a are connected to each other at one end close to the second conductive part group Z2 through a connecting part, and the first conductive part 211a and the second conductive part 212a are spaced apart from one end of the second conductive part group Z2, forming a short-circuit calibration structure SG. In addition, the first conductive part 211a, the first conductive part 211b and the second conductive part 212b are spaced apart from each other, so that the first conductive part 211b and the second conductive part 212b respectively constitute an open-circuit calibration structure OG. Similarly, the third conductive part 213a and the fourth conductive part 214a are connected to each other at one end close to the first conductive part group Z1 through a connecting part, and the third conductive part 213a and the fourth conductive part 214a are spaced apart from one end of the first conductive part group Z1, forming a short-circuit calibration structure SG. The third conductive portion 213a, the third conductive portion 213b, and the fourth conductive portion 214b are arranged at intervals from each other, so that the third conductive portion 213b and the fourth conductive portion 214b respectively form an open-circuit calibration structure OG.
[0069] In some other examples, reference Figure 3b The first calibration structures and the second calibration structures include two types: a short-circuit calibration structure SG and a load calibration structure RG. The first conductive portion 211a and the second conductive portion 212a constitute the short-circuit calibration structure SG. The first conductive portion 211a and the second conductive portion 212b are also connected to each other at one end close to the second conductive portion group Z2 through a connecting portion. The first conductive portion 211a and the second conductive portion 212b are spaced apart from one end of the second conductive portion group Z2, and the first conductive portion 211b is connected to each other through a load ( Figure 3bThe black block in the figure represents the load) is grounded, so that the first conductive parts 211a, 211b and the second conductive part 212b form a load calibration structure RG. Similarly, the third conductive part 213a and the fourth conductive part 214b form a short-circuit calibration structure SG, and the third conductive part 213b is connected to the load ( Figure 3b The black block in the figure represents a load) is grounded, and the third conductive parts 213a, 213b and the fourth conductive part 214b constitute a load calibration structure RG.
[0070] In some other examples, reference Figure 3c The plurality of first calibration structures and the plurality of second calibration structures respectively include a short-circuit calibration structure SG and an open-circuit calibration structure OG. Among them, the second conductive parts 212a, 212b and the first conductive part 211a constitute the short-circuit calibration structure SG. The first conductive part 211b constitutes the open-circuit calibration structure OG. Similarly, the fourth conductive parts 214a, 214b and the third conductive part 213a constitute the short-circuit calibration structure SG. The third conductive part 213b constitutes the open-circuit calibration structure OG.
[0071] In some other examples, reference Figure 3d The plurality of first calibration structures and the plurality of second calibration structures respectively include a load calibration structure RG and an open circuit calibration structure OG. The second conductive portion 212a constitutes the open circuit calibration structure OG, and the first conductive portion 211b is connected to the load ( Figure 3d The black block in the figure represents the load) is grounded and forms a load calibration structure RG with the first conductive part 211a and the second conductive part 212b. Similarly, the fourth conductive part 214a forms an open circuit calibration structure OG, and the third conductive part 213b is connected to the load ( Figure 3b The black block in the figure represents a load) is grounded and forms a load calibration structure RG with the third conductive portion 213a and the fourth conductive portion 214b.
[0072] In some other examples, reference Figure 3e The plurality of first calibration structures and the plurality of second calibration structures respectively include a short-circuit calibration structure SG and an open-circuit calibration structure OG. The first conductive portion 211a and the second conductive portion 212a constitute the short-circuit calibration structure SG, and the first conductive portion 211b and the second conductive portion 212b respectively constitute the open-circuit calibration structure OG. Similarly, the third conductive portion 213a and the fourth conductive portion 214a respectively constitute the open-circuit calibration structure OG, and the third conductive portion 213b and the fourth conductive portion 214b constitute the short-circuit calibration structure SG.
[0073] Reference Figure 3a to Figure 3d, the first calibration structure and the second calibration structure corresponding to each other are arranged in mirror symmetry, which simplifies the design difficulty. For example, the open-circuit calibration structure OG in the first calibration structure and the second calibration structure is arranged in mirror symmetry, the short-circuit calibration structure SG in the first calibration structure and the second calibration structure is arranged in mirror symmetry, and the load calibration structure RG in the first calibration structure and the second calibration structure is arranged in mirror symmetry. Then, the first conductive part group Z1 and the second conductive part group Z2 are arranged in a non-mirror symmetric manner in the second direction F2, and are arranged in mirror symmetry in the first direction F1. Or, refer to Figure 3e , the first calibration structure and the second calibration structure corresponding to each other are arranged in a central symmetric manner, for example, the open-circuit calibration structure OG in the first calibration structure and the second calibration structure are arranged in a central symmetric manner, and the short-circuit calibration structure SG in the first calibration structure and the second calibration structure are arranged in a central symmetric manner. Then the first conductive portion group Z1 and the second conductive portion group Z2 are arranged in a non-mirror symmetric manner in the second direction F2, and are also arranged in a non-mirror symmetric manner in the first direction F1.
[0074] Reference Figures 3a to 3e , the spacing between adjacent conductive parts in the first conductive part group Z1 can be made the same. Exemplarily, the spacing between adjacent conductive parts in the first conductive part group Z1 can be set to 40um to 80um, for example, 40um, 50um, 55um, 60um, 65um, 70um, 72um, 75um, 78um, 80um.
[0075] Reference Figures 3a to 3e , the spacing between adjacent conductive parts in the second conductive part group Z2 can be made the same. Exemplarily, the spacing between adjacent conductive parts in the second conductive part group Z2 can be set to 40um to 80um, for example, 40um, 50um, 55um, 60um, 65um, 70um, 72um, 75um, 78um, 80um.
[0076] In addition, the distance between adjacent conductive portions in the first conductive portion group Z1 may be the same as the distance between adjacent conductive portions in the second conductive portion group Z2.
[0077] Exemplarily, the length h11 of the first conductive portion 211a in the first direction F1, the length h21 of the third conductive portion 213a in the first direction F1, the length h12 of the first conductive portion 211b in the first direction F1, and the length h22 of the third conductive portion 213b in the first direction F1 can be 100um to 400um, respectively, for example, 100um, 150um, 170um, 190um, 200um, 230um, 250um, 280um, 300um, 320um, 350um, 380um, 400um.
[0078] Reference Figures 3a to 3e , the length h11 of the first conductive portion 211a arranged along the first direction F1 can be made the same as the length h21 of the third conductive portion 213a in the first direction F1, and the length h12 of the first conductive portion 211b arranged along the first direction F1 can be made the same as the length h22 of the third conductive portion 213b in the first direction F1.
[0079] Exemplarily, the length h13 of the second conductive portion 212a in the first direction F1, the length h23 of the fourth conductive portion 214a in the first direction F1, the length h14 of the second conductive portion 212b in the first direction F1, and the length h24 of the fourth conductive portion 214b in the first direction F1 can be 100um to 400um, respectively, for example, 100um, 150um, 170um, 190um, 200um, 230um, 250um, 280um, 300um, 320um, 350um, 380um, and 400um.
[0080] Reference Figures 3a to 3e , the length h13 of the second conductive portion 212a arranged along the first direction F1 in the first direction F1 can be made the same as the length h23 of the fourth conductive portion 214a in the first direction F1, and the length h14 of the second conductive portion 212b arranged along the first direction F1 in the first direction F1 can be made the same as the length h24 of the fourth conductive portion 214b in the first direction F1.
[0081] Reference Figures 3a to 3e In the first conductive portion group Z1, the lengths of the conductive portions corresponding to different signal pins may be different, for example, the length of the first conductive portion 211a in the first direction may be greater than the length of the first conductive portion 211b in the first direction. In the second conductive portion group Z2, the lengths of the conductive portions corresponding to different signal pins may also be different, for example, the length of the third conductive portion 213a in the first direction may be greater than the length of the third conductive portion 213b in the first direction.
[0082] Figures 4a to 4d They are schematic diagrams of the structure of an on-wafer calibration component in another embodiment of the present application. Figures 4a to 4d The calibration piece may include: a substrate 100 and a metal layer 200 disposed on the substrate 100. The metal layer 200 includes: a plurality of conductive parts (e.g., 211a, 211b, 212a, 212b, 213a, 213b, 214a, 214b). This embodiment is a modification of the implementation in the above embodiment. The following only describes the differences between this embodiment and the above embodiment, and the similarities are not repeated here.
[0083] In the embodiments of this application, refer to Figures 4a to 4dThe first conductive parts 211a, 211b and the second conductive parts 212a, 212b can constitute a plurality of third calibration structures, and the third conductive parts 213a, 213b and the fourth conductive parts 214a, 214b can also constitute a plurality of fourth calibration structures. The plurality of third calibration structures include any one of the short-circuit calibration structure SG, the open-circuit calibration structure OG and the load calibration structure RG, and the plurality of fourth calibration structures include any one or more of the remaining calibration structures except the third calibration structure among the short-circuit calibration structure SG, the open-circuit calibration structure OG and the load calibration structure RG. With this arrangement, the third calibration structure and the fourth calibration structure can be different.
[0084] Exemplarily, the first conductive portion group Z1 can be arranged in mirror symmetry in the second direction F2, and the first conductive portion group Z1 and the second conductive portion group Z2 can be arranged in non-mirror symmetry in the first direction F1. When the second conductive portion group Z2 constitutes the same calibration structure, the second conductive portion group Z2 can also be arranged in mirror symmetry in the second direction F2.
[0085] In some examples, reference Figure 4a The plurality of third calibration structures include a short-circuit calibration structure SG, and the plurality of fourth calibration structures include an open-circuit calibration structure OG. The first conductive portion 211a, 211b and the second conductive portion 212a, 212b constitute the short-circuit calibration structure SG, and the third conductive portion 213a, 213b and the fourth conductive portion 214a, 214b constitute the open-circuit calibration structure OG.
[0086] In some other examples, reference Figure 4b The plurality of third calibration structures include an open calibration structure OG, and the plurality of fourth calibration structures include a short calibration structure SG. The first conductive parts 211a, 211b and the second conductive parts 212a, 212b constitute the open calibration structure OG, and the third conductive parts 213a, 213b are grounded and constitute the short calibration structure SG with the fourth conductive parts 214a, 214b.
[0087] In some other examples, reference Figure 4c The plurality of third calibration structures include an open-circuit calibration structure OG, and the plurality of fourth calibration structures include a load calibration structure RG. The first conductive portions 211a, 211b and the second conductive portions 212a, 212b constitute the open-circuit calibration structure OG, and the third conductive portions 213a, 213b are respectively connected through loads ( Figure 3b The black block in the figure represents a load) is grounded and forms a load calibration structure RG with the fourth conductive parts 214a and 214b.
[0088] In some other examples, reference Figure 4dThe plurality of third calibration structures include a load calibration structure RG, and the plurality of fourth calibration structures include a short-circuit calibration structure SG. The first conductive portions 211a and 211b are respectively connected by a load ( Figure 3b The black block in the figure represents a load) is grounded and forms a load calibration structure RG with the second conductive parts 212a and 212b, and the third conductive parts 213a and 213b and the fourth conductive parts 214a and 214b form an open circuit calibration structure OG.
[0089] Reference Figure 4a , Figure 4c , Figure 4d , the lengths of the conductive parts in the third alignment structure in the first direction F1 can be made the same. Figure 4b , the lengths (e.g., h21, h22) of the conductive parts corresponding to the signal pins (e.g., the first conductive parts 211a, 211b) in the first direction F1 of the third calibration structure may be the same, the lengths (e.g., h23, h24) of the conductive parts corresponding to the ground pins (e.g., the second conductive parts 212a, 212b) in the first direction F1 of the third calibration structure may be the same, and the lengths (e.g., h21, h22) of the conductive parts corresponding to the signal pins (e.g., the first conductive parts 211a, 211b) in the first direction F1 of the third calibration structure may be different from the lengths (e.g., h23, h24) of the conductive parts corresponding to the ground pins (e.g., the second conductive parts 212a, 212b) in the first direction F1 of the fourth calibration structure may be set in the same manner. Optionally, the lengths of the conductive parts in the first direction F1 of the third calibration structure may be the same as the lengths of the conductive parts in the first direction F1 of the fourth calibration structure. Alternatively, the length of the conductive portion in the third alignment structure in the first direction F1 may be greater than or less than the length of the conductive portion in the fourth alignment structure in the first direction F1.
[0090] Figure 5a to Figure 5c They are schematic diagrams of the structure of an on-wafer calibration component in another embodiment of the present application. Figure 5a to Figure 5c The calibration piece may include: a substrate 100 and a metal layer 200 disposed on the substrate 100. The metal layer 200 includes: a plurality of conductive parts (e.g., 211a, 211b, 212a, 212b, 213a, 213b, 214a, 214b). This embodiment is a modification of the implementation in the above embodiment. The following only describes the differences between this embodiment and the above embodiment, and the similarities are not repeated here.
[0091] In the embodiment of the present application, multiple conductive parts can constitute the same calibration structure. Figure 5a , multiple conductive parts constitute an open circuit calibration structure OG. Figure 5b , multiple conductive parts constitute a short-circuit calibration structure SG. Figure 5c , multiple conductive parts constitute the load calibration structure RG. The present application can make the lengths of different conductive parts corresponding to different signal pins different, so that the calibration piece is arranged in a non-mirror symmetric manner in the first direction F1 and / or the second direction F2.
[0092] For example, refer to Figure 5a to Figure 5c , the length h11 of the first conductive portion 211a along the first direction F1 can be smaller than the length h21 of the corresponding third conductive portion 213a along the first direction F1. Also, the length h12 of the first conductive portion 211b along the first direction F1 can be smaller than the length h22 of the corresponding third conductive portion 213b along the first direction F1.
[0093] Reference Figure 5a to Figure 5c , the lengths of the plurality of first conductive parts arranged along the second direction F2 in the first direction F1 can be increased sequentially, for example, the length h11 of the first conductive part 211a in the first direction F1 is smaller than the length h12 of the first conductive part 211b in the first direction F1. And / or, the lengths of the plurality of third conductive parts arranged along the second direction F2 in the first direction F1 can be increased sequentially, for example, the length h21 of the third conductive part 213a in the first direction F1 is smaller than the length h22 of the third conductive part 213b in the first direction F1.
[0094] Reference Figure 5a to Figure 5c , the lengths of the conductive parts of the corresponding ground pins along the first direction F1 can be the same, or different, or partially the same and partially different. For example, the length h13 of the second conductive part 212a along the first direction F1, the length h14 of the second conductive part 212b along the first direction F1, the length h23 of the fourth conductive part 214a along the first direction F1, and the length 24 of the fourth conductive part 214b along the first direction F1 are the same. Of course, the length h13 of the second conductive part 212a along the first direction F1, the length h14 of the second conductive part 212b along the first direction F1, the length h23 of the fourth conductive part 214a along the first direction F1, and the length 24 of the fourth conductive part 214b along the first direction F1 can also be different, or partially the same and partially different.
[0095] Figure 6 This is a flow chart of an error model generation method in one embodiment of the present application. Figure 6 , the error model generation method includes:
[0096] S11. Provide a plurality of calibration pieces, the plurality of calibration pieces comprising: any one of the on-chip calibration pieces provided in the above embodiments of the present application, and Figure 1a The short-circuit calibration kit shown, Figure 1b The open circuit calibration shown, Figure 1c Load calibration shown, Figure 1dThe straight calibration kits shown are 5 in total. Figure 5a As an example, the on-wafer calibration piece shown in the figure is combined with Figure 8 , connect the signal pins 21 of the four differential probes 20 connected to the vector network analyzer 10 to the first conductive parts 211a, 211b and the third conductive parts 213a, 213b respectively, connect the ground pins 22 of the four differential probes to the second conductive parts 212a, 212b and the fourth conductive parts 214a, 214b respectively, output the test to the on-wafer calibration part, and perform the test. The vector network analyzer 10 will test and obtain the measured scattering parameter matrix corresponding to the on-wafer calibration part. Similarly, the vector network analyzer 10 tests and obtains the measured scattering parameter matrix corresponding to each of the remaining four calibration parts. Therefore, the measured scattering parameter matrix corresponding to each calibration part can be obtained from the vector network analyzer 10. Among them, the measured scattering parameter matrix SM_1 corresponding to the on-chip calibration part, the measured scattering parameter matrix SM_2 corresponding to the short-circuit calibration part, the measured scattering parameter matrix SM_3 corresponding to the open-circuit calibration part, the measured scattering parameter matrix SM_4 corresponding to the load calibration part, and the measured scattering parameter matrix SM_5 corresponding to the through calibration part in the embodiment of the present application are tested. Among them, the present application takes N=4 as an example, and the measured scattering parameter matrices SM_1 to SM_5 are matrices with 4 rows and 4 columns respectively.
[0097] S12, the initial 8*8 (i.e. 64) error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix, and the measured scattering parameter matrix and the intrinsic scattering parameter matrix are connected by unknown coefficients. The unknown coefficients in the initial 8*8 (i.e. 64) error model can form four characteristic coefficient matrices TA', TB', TC', and TD' respectively. For example, the initial M*M error model meets the following relationship: SA = (TA'-SM*TC') -1 *(SM*TD'-TB'); wherein SA represents the intrinsic scattering parameter matrix, SM represents the measured scattering parameter matrix, TA' represents the first characteristic coefficient matrix among the four characteristic coefficient matrices with unknown coefficients, TB' represents the second characteristic coefficient matrix among the four characteristic coefficient matrices with unknown coefficients, TC' represents the third characteristic coefficient matrix among the four characteristic coefficient matrices with unknown coefficients, and TD' represents the fourth characteristic coefficient matrix among the four characteristic coefficient matrices with unknown coefficients.
[0098] In addition, the intrinsic scattering parameter matrix corresponding to each calibration component can be obtained in advance through calibration component measurement or simulation, that is, the intrinsic scattering parameter matrix SA_1 corresponding to the on-chip calibration component, the intrinsic scattering parameter matrix SA_2 corresponding to the short-circuit calibration component, the intrinsic scattering parameter matrix SA_3 corresponding to the open-circuit calibration component, the intrinsic scattering parameter matrix SA_4 corresponding to the load calibration component, and the intrinsic scattering parameter matrix SA_5 corresponding to the through calibration component in the embodiment of the present application are all known. Exemplarily, these intrinsic scattering parameter matrices can be stored in the controller or in other devices, and can be directly called when executing step S12.
[0099] Exemplarily, the intrinsic scattering parameter matrices SA_1 to SA_5 and the characteristic coefficient matrix are matrices with 4 rows and 4 columns respectively. For example, taking the measured scattering parameter matrix SM_1\intrinsic scattering parameter matrix SA_1 and the characteristic coefficient matrix TA' as examples, SM_1 is as follows:
[0100]
[0101] SA_1 is as follows:
[0102]
[0103] TA' is as follows:
[0104]
[0105] Since the measured scattering parameter matrices SM_1~SM_5 and the intrinsic scattering parameter matrices SA_1~SA_5 are all known, the measured scattering parameter matrix SM_1 and the intrinsic scattering parameter matrix SA_1, the measured scattering parameter matrix SM_2 and the intrinsic scattering parameter matrix SA_2, the measured scattering parameter matrix SM_3 and the intrinsic scattering parameter matrix SA_3, the measured scattering parameter matrix SM_4 and the intrinsic scattering parameter matrix SA_4, and the measured scattering parameter matrix SM_5 and the intrinsic scattering parameter matrix SA_5 are input into the relationship that the initial 8*8 (i.e., 64) error model conforms to to obtain five relationship equations: SA_1=(TA'-SM_1*TC') -1 *(SM_1*TD'-TB'), SA_2=(TA'-SM_2*TC') -1 *(SM_2*TD'-TB'), SA_3=(TA'-SM_3*TC') -1 *(SM_3*TD'-TB'), SA_4=(TA'-SM_4*TC') -1 *(SM_4*TD'-TB'), SA_5=(TA'-SM_5*TC') -1*(SM_5*TD'-TB'), by fitting these five equations, the unknown coefficients in the initial 8*8 (i.e. 64) error model can be obtained, that is, the unknown characteristic coefficient matrices TA', TB', TC', and TD' can be solved respectively to obtain the solved characteristic coefficient matrices TA, TB, TC, and TD.
[0106] S13, the four characteristic coefficient matrices TA, TB, TC, TD are solved and substituted into the initial 8*8 (i.e. 64) error model, and the characteristic coefficient matrices TA', TB', TC', TD' in the initial 8*8 (i.e. 64) error model are replaced to obtain the target 8*8 (i.e. 64) error model, and the target 8*8 (i.e. 64) error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix. For example, the target 8*8 (i.e. 64) error model meets the following relationship: SA = (TA-SM*TC) -1 *(SM*TD-TB); TA represents the first characteristic coefficient matrix among the four characteristic coefficient matrices, TB represents the second characteristic coefficient matrix among the four characteristic coefficient matrices, TC represents the third characteristic coefficient matrix among the four characteristic coefficient matrices, and TD represents the fourth characteristic coefficient matrix among the four characteristic coefficient matrices.
[0107] The embodiment of the present application also provides an error model generation system, which includes: multiple calibration parts and a controller. The controller is used to obtain the measured scattering parameter matrix corresponding to each calibration part when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to multiple calibration parts for testing. The measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration part are input into the initial M*M error model for fitting to obtain the fitted target M*M error model, where M=2N, and N represents the total number of signal pins of all the differential probes; the target M*M error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix. The process of the method implemented by the controller can refer to the process of the above-mentioned error model generation method, which will not be repeated here.
[0108] In addition, the plurality of calibration pieces include: any one or more of the on-chip calibration pieces provided in the embodiments of the present application, and any one or more of the on-chip calibration pieces in the prior art. Figure 1a The short-circuit calibration kit shown, Figure 1b The open circuit calibration shown, Figure 1c Load calibration shown, Figure 1d Thru calibration kit shown.
[0109] Figure 7 This is a flow chart of a testing method in one embodiment of the present application. Figure 7 , the test methods include:
[0110] S21, connect the signal pin and ground pin of the differential probe 20 connected to the vector network analyzer 10 to the device under test respectively, output the test signal, and perform the test. The vector network analyzer 10 will test and obtain the measured scattering parameter matrix Sdutm corresponding to the device under test. This application takes N=4 as an example, then Sdutm is a matrix of 4 rows and 4 columns.
[0111] S22, input the measured scattering parameter matrix Sdutm of the device under test into the target 8*8 (i.e. 64) error model, and the calibrated intrinsic scattering parameter matrix Sdut of the device under test can be obtained. Among them, the corresponding relationship between the intrinsic scattering parameter matrix Sdut and the measured scattering parameter matrix Sdutm input into the target 8*8 (i.e. 64) error model is: Sdut = (TA-Sdutm*TC) -1 *(Sdutm*TD-TB).
[0112] The embodiment of the present application also provides a test system, which may include: a vector network analyzer, four differential probes and a controller. The controller is used to: when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to the device under test for testing, obtain the measured scattering parameter matrix of the device under test, input the measured scattering parameter matrix of the device under test into the target M*M term error model, and obtain the calibrated intrinsic scattering parameter matrix of the device under test. The process implemented by the controller can refer to the process of the above-mentioned test method, which will not be repeated here.
[0113] In addition, in the test system provided in the embodiment of the present application, a plurality of calibration parts may also be included, and the controller is also used to obtain the measured scattering parameter matrix corresponding to each calibration part when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to different multiple calibration parts for testing. The measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration part are input into the initial M*M error model for fitting to obtain the fitted target M*M error model, where M=2N, and N represents the total number of signal pins of all differential probes; the target M*M error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix. With this arrangement, the test system can realize not only the test function, but also the calibration function.
[0114] Furthermore, the plurality of calibration pieces include: any one or more of the on-chip calibration pieces in the above-mentioned embodiments of the present application, and any one or more of the on-chip calibration pieces in the existing embodiments. Figure 1a The short-circuit calibration kit shown, Figure 1b The open circuit calibration shown, Figure 1c Load calibration shown, Figure 1d Thru calibration shown.
[0115] An embodiment of the present application also provides a computer storage medium, which stores executable instructions. When the executable instructions are executed by a computer, the error model generation method provided in the above embodiment of the present application is executed, or the test method provided in the above embodiment of the present application is executed.
[0116] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.
Claims
1. An on-wafer calibration component, characterized in that: include: substrate; A metal layer is disposed on the substrate; The metal layer comprises: a plurality of conductive parts, the plurality of conductive parts forming at least two of a short circuit calibration structure, an open circuit calibration structure and a load calibration structure; The number of the multiple conductive parts is not greater than the total number of signal pins and ground pins of the differential probes connected to the vector network analyzer, the signal pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of the multiple conductive parts, and the ground pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one or more conductive parts of the multiple conductive parts, and the vector network analyzer is connected to at least four differential probes.
2. The on-wafer calibration component according to claim 1, characterized in that: The plurality of conductive parts include: a first conductive part group and a second conductive part group arranged along a first direction, the first conductive part group includes a plurality of first conductive parts corresponding to the signal pins and a plurality of second conductive parts corresponding to the ground pins; the second conductive part group includes: a plurality of third conductive parts corresponding to the signal pins and a plurality of fourth conductive parts corresponding to the ground pins; The plurality of first conductive parts and the plurality of second conductive parts constitute a plurality of first calibration structures, wherein the plurality of first calibration structures include at least two of the short-circuit calibration structure, the open-circuit calibration structure, and the load calibration structure; The plurality of third conductive parts and the plurality of fourth conductive parts constitute a plurality of second calibration structures, and the plurality of second calibration structures include at least two of the short-circuit calibration structure, the open-circuit calibration structure, and the load calibration structure.
3. The on-wafer calibration component according to claim 2, characterized in that: The plurality of first conductive portions correspond to the plurality of third conductive portions one by one, and the first conductive portions and the third conductive portions corresponding to each other are arranged along the first direction; The plurality of second conductive portions correspond to the plurality of fourth conductive portions one by one, and the second conductive portions and the fourth conductive portions corresponding to each other are arranged along the set direction.
4. The on-wafer calibration piece according to claim 2 or 3, characterized in that: The plurality of first calibration structures correspond to the plurality of second calibration structures one by one, and the first calibration structures and second calibration structures corresponding to each other are the same.
5. The on-wafer calibration component according to claim 4, characterized in that: The first calibration structure and the second calibration structure corresponding to each other are arranged in a mirror-symmetrical manner or a center-symmetrical manner.
6. The on-wafer calibration component according to claim 1, characterized in that: The plurality of conductive parts include: a first conductive part group and a second conductive part group arranged along a first direction, the first conductive part group includes a plurality of first conductive parts corresponding to the signal pins and a plurality of second conductive parts corresponding to the ground pins; the second conductive part group includes: a plurality of third conductive parts corresponding to the signal pins and a plurality of fourth conductive parts corresponding to the ground pins; The plurality of first conductive parts and the plurality of second conductive parts constitute a plurality of third calibration structures, wherein the plurality of third calibration structures include one of the short-circuit calibration structure, the open-circuit calibration structure and the load calibration structure; The plurality of third conductive parts and the plurality of fourth conductive parts constitute a plurality of fourth calibration structures, and the plurality of fourth calibration structures include at least one of the remaining calibration structures among the short-circuit calibration structure, the open-circuit calibration structure, and the load calibration structure except the third calibration structure.
7. An on-wafer calibration component, characterized in that: include: substrate; A metal layer is disposed on the substrate; The metal layer comprises: a plurality of conductive parts, the plurality of conductive parts constitute a same calibration structure, and the calibration structure comprises one of a short-circuit calibration structure, an open-circuit calibration structure and a load calibration structure; The number of the multiple conductive parts is not greater than the total number of signal pins and ground pins of the differential probes connected to the vector network analyzer, the signal pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of the multiple conductive parts, and the ground pin of each differential probe connected to the vector network analyzer corresponds one-to-one to one of one or more conductive parts of the multiple conductive parts, and the lengths of different conductive parts corresponding to different signal pins are different; the vector network analyzer is connected to at least four differential probes.
8. The on-wafer calibration component according to claim 7, characterized in that: The plurality of conductive parts include: a first conductive part group and a second conductive part group arranged along a first direction, the first conductive part group includes a plurality of first conductive parts corresponding to the signal pins and a plurality of second conductive parts corresponding to the ground pins; the second conductive part group includes: a plurality of third conductive parts corresponding to the signal pins and a plurality of fourth conductive parts corresponding to the ground pins; The plurality of first conductive portions correspond to the plurality of third conductive portions one by one, and the first conductive portions and the third conductive portions corresponding to each other are arranged along the first direction; The plurality of second conductive portions correspond to the plurality of fourth conductive portions one by one, and the second conductive portions and the fourth conductive portions corresponding to each other are arranged along the first direction.
9. The on-wafer calibration component according to claim 8, characterized in that: A length of the first conductive portion along the first direction is smaller than a length of the corresponding third conductive portion along the first direction.
10. The on-wafer calibration piece according to claim 8 or 9, characterized in that: The plurality of first conductive portions are arranged along the second direction, and the lengths of the plurality of first conductive portions arranged along the second direction in the first direction are increased sequentially; and / or, The plurality of third conductive portions are arranged along the second direction, and lengths of the plurality of third conductive portions arranged along the second direction are sequentially increased in the first direction.
11. A method for generating an error model, characterized in that: include: When the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to a plurality of calibration pieces for testing, a measured scattering parameter matrix corresponding to each of the calibration pieces is obtained; The measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each of the calibration components are input into the initial M*M-term error model for fitting to obtain a fitted target M*M-term error model, where M=2N, and N represents the total number of signal needles of all the differential probes; the target M*M-term error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix.
12. The error model generation method according to claim 11, characterized in that: The step of inputting the measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration piece into the initial M*M error model for fitting to obtain the fitted target M*M error model includes: Inputting the measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each calibration piece into the initial M*M error model for fitting, and solving N characteristic coefficient matrices; Substitute the N characteristic coefficient matrices into the initial M*M error model to obtain the target M*M error model.
13. The error model generation method according to claim 12, characterized in that: The target M*M error model conforms to the following relationship: SA(TA-SM*TC) -1 *(SM*TD-TB) Among them, the SA represents the intrinsic scattering parameter matrix, the SM represents the measured scattering parameter matrix, the TA represents the first characteristic coefficient matrix among the N characteristic coefficient matrices, the TB represents the second characteristic coefficient matrix among the N characteristic coefficient matrices, the TC represents the third characteristic coefficient matrix among the N characteristic coefficient matrices, and the TD represents the fourth characteristic coefficient matrix among the N characteristic coefficient matrices.
14. The error model generation method according to any one of claims 11 to 13, characterized in that: The multiple calibration parts include: one or more of the on-chip calibration parts as described in any one of claims 1-11, as well as a short-circuit calibration part, an open-circuit calibration part, a load calibration part, and a through calibration part.
15. A testing method, characterized in that: include: When the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to the device under test for testing, a measured scattering parameter matrix of the device under test is obtained; Inputting the measured scattering parameter matrix of the device under test into the target M*M-term error model to obtain a calibrated intrinsic scattering parameter matrix of the device under test; The target M*M error model is obtained based on the error model generation method according to any one of claims 11-13.
16. An error model generation system, characterized in that: include: Multiple calibration kits, Controller for: When the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to the plurality of calibration pieces for testing, a measured scattering parameter matrix corresponding to each of the calibration pieces is obtained; The measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each of the calibration components are input into the initial M*M-term error model for fitting to obtain a fitted target M*M-term error model, where M=2N, and N represents the total number of signal needles of all the differential probes; the target M*M-term error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix.
17. The error model generation system according to claim 16, characterized in that: The multiple calibration parts include: one or more of the on-chip calibration parts as described in any one of claims 1 to 10, as well as a short-circuit calibration part, an open-circuit calibration part, a load calibration part, and a through calibration part.
18. A testing system, characterized in that: include: Vector network analyzer; At least four differential probes, wherein the at least four differential probes are respectively connected to the vector network analyzer; The controller is used to obtain a measured scattering parameter matrix of the device under test when the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to the device under test for testing, and input the measured scattering parameter matrix of the device under test into a target M*M-term error model to obtain a calibrated intrinsic scattering parameter matrix of the device under test.
19. The test system according to claim 18, characterized in that: Also includes: A plurality of calibration parts, the plurality of calibration parts comprising: one or more of the on-chip calibration parts according to any one of claims 1 to 10, as well as a short-circuit calibration part, an open-circuit calibration part, a load calibration part, and a through-calibration part; The controller is further used for: When the signal pin and the ground pin of the differential probe connected to the vector network analyzer are respectively connected to different calibration pieces for testing, obtaining a measured scattering parameter matrix corresponding to each calibration piece; The measured scattering parameter matrix and the intrinsic scattering parameter matrix corresponding to each of the calibration components are input into the initial M*M-term error model for fitting to obtain a fitted target M*M-term error model, where M=2N, and N represents the total number of signal needles of all the differential probes; the target M*M-term error model is used to describe the relationship between the measured scattering parameter matrix and the intrinsic scattering parameter matrix.
20. A computer storage medium, characterized in that The computer storage medium stores executable instructions, and when the executable instructions are executed by a computer, the error model generation method according to any one of claims 11 to 14 is executed, or the testing method according to claim 15 is executed.
