A frequency scanning method, system and storage medium
By constructing accurate rational interpolation algorithms and dynamic error analysis, the adaptive frequency scanning algorithm improves accuracy and efficiency in complex frequency responses, solves the problem of low accuracy in the prior art, and achieves more efficient frequency scanning.
Patent Information
- Application Number
- CN202411230729.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-09-04
AI Technical Summary
The existing adaptive frequency scanning algorithms have low accuracy when processing complex frequency responses, making it difficult to adapt to various complex situations, affecting the accuracy and efficiency of frequency scanning results.
By constructing accurate rational interpolation algorithms and dynamic adjustments, combined with interval-based error analysis algorithms, we can determine the error convergence standard value or square root error, insert frequency points in the interval with the largest error, build a brand new sample frequency, optimize frequency response prediction and error evaluation .
The accuracy of frequency response prediction and the accuracy of error evaluation of frequency scanning results are improved, the adaptability and reliability of frequency scanning are enhanced, the number of non-essential frequency points is reduced, and the allocation efficiency of computing resources is improved.
Smart Images

Figure CN118747446B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of frequency scanning, and further to a frequency scanning method, system and storage medium. Background Art
[0002] Frequency scanning is a basic algorithm for analyzing electromagnetic responses. The traditional frequency scanning algorithm performs discrete frequency point tests within a specified frequency range. This algorithm has the problems of large computational complexity and long time consumption. Currently, the most popular frequency scanning algorithm is the adaptive frequency scanning algorithm, which optimizes the test process by intelligently selecting frequency points, reduces the number of required frequency points, and maintains the accuracy of the frequency scanning results. However, the existing adaptive frequency scanning algorithm has low accuracy when processing complex frequency responses. In addition, during the error analysis process, the existing adaptive frequency scanning algorithm adjusts the frequency point by the error between two adjacent frequency points, which is difficult to adapt to various complex situations and will also affect the accuracy of the frequency scanning results. Summary of the invention
[0003] In order to solve the above technical problems, the present application provides a frequency scanning method, system and storage medium, which improve the accuracy and efficiency of the frequency scanning results.
[0004] In a first aspect, the present application provides a frequency scanning method, comprising: determining a first sample frequency and a second sample frequency based on a preset simulation frequency range, wherein the second sample frequency includes the first sample frequency, and the number of frequency points of the second sample frequency is greater than the number of frequency points of the first sample frequency, and the preset simulation frequency range is a wide frequency range; calculating, according to the first sample frequency and the target circuit model, a first frequency response actual value corresponding to each frequency point in the first sample frequency; constructing a first data set according to the first sample frequency and the first frequency response actual value; interpolating the first data set by a centroid rational interpolation algorithm or a minimum rational interpolation algorithm to obtain a first interpolation function; calculating, according to the second sample frequency and the target circuit model, a second frequency response actual value corresponding to each frequency point in the second sample frequency; constructing, according to the second sample frequency and the second frequency response actual value a second data set; interpolating the second data set by the centroid rational interpolation algorithm or the minimum rational interpolation algorithm to obtain a second interpolation function; performing error analysis on the first interpolation function and the second interpolation function by the error analysis algorithm to obtain errors in each interval of the second sample frequency, and determining an error convergence standard value or a root mean square error according to the errors in each interval; when the error convergence standard value or the root mean square error is not less than a preset value, inserting at least one frequency point in the interval with the largest error in the second sample frequency, taking the second sample frequency after the inserted frequency point as the third sample frequency, constructing a third interpolation function for the third sample frequency, performing error analysis and error convergence judgment steps on the second interpolation function and the third interpolation function, and outputting a first result, which is a frequency response estimation value of all frequency points within the simulation frequency range obtained by the third interpolation function.
[0005] The above frequency scanning method improves the accuracy of frequency response prediction and error evaluation of frequency scanning results by constructing an accurate rational interpolation algorithm, dynamically adjusting it, and using an interval-based error analysis algorithm. When the error convergence standard value is not less than the preset value, a frequency point is inserted in the interval with the largest error to construct a new sample frequency. Compared with the method of adjusting the frequency point based on the error of two adjacent frequency points, the adaptability and reliability of frequency scanning are improved, the number of unnecessary frequency points is further reduced, and the efficiency of frequency scanning is improved. In addition, by constructing a first interpolation function and a second interpolation function, and performing error analysis on the interval of the first interpolation function and the second interpolation function, the accuracy of frequency response prediction and error evaluation of frequency scanning results is further improved, and the allocation of computing resources is optimized.
[0006] In one implementation, it also includes: when the error convergence standard value or the root mean square error is less than the preset value, outputting a second result, the second result being a frequency response estimation value of all frequency points within the simulation frequency range obtained by the second interpolation function.
[0007] In one implementation, determining the first sample frequency and the second sample frequency based on a preset simulation frequency range specifically includes: selecting an interval with the largest interval interval in the first sample frequency, inserting frequency points by binary search; and using the first sample frequency with the inserted frequency points as the second sample frequency.
[0008] In one implementation, the first sample frequency includes a minimum frequency point and a maximum frequency point of the simulation frequency range.
[0009] In a second aspect, the present application also provides a frequency scanning system, including: a processor, a data input interface, a data output interface and a frequency response calculation module; the data input interface is used to receive a simulation frequency range and a target circuit model input by a user; the processor determines a first sample frequency and a second sample frequency based on the simulation frequency range input by the user, the second sample frequency includes the first sample frequency, and the number of frequency points of the second sample frequency is greater than the number of frequency points of the first sample frequency; the frequency response calculation module calculates the first frequency response actual value corresponding to each frequency point in the first sample frequency according to the first sample frequency and the target circuit model; the processor constructs a first data set according to the first sample frequency and the first frequency response actual value; the processor interpolates the first data set through a centroid rational interpolation algorithm or a minimum rational interpolation algorithm to obtain the first interpolation function; the frequency response calculation module calculates the second frequency response actual value corresponding to each frequency point in the second sample frequency according to the second sample frequency and the target circuit model. value; the processor constructs a second data set according to the second sample frequency and the actual value of the second frequency response; the processor interpolates the second data set through the centroid rational interpolation algorithm or the minimum rational interpolation algorithm to obtain a second interpolation function; the processor performs error analysis on the first interpolation function and the second interpolation function through the error analysis algorithm to obtain the errors of each interval in the second sample frequency, and determines the error convergence standard value or the root mean square error according to the errors of each interval; when the error convergence standard value or the root mean square error is not less than a preset value, the processor inserts at least one frequency point in the interval with the largest error in the second sample frequency, uses the second sample frequency after the frequency point is inserted as the third sample frequency, constructs a third interpolation function for the third sample frequency, performs error analysis and error convergence judgment steps on the second interpolation function and the third interpolation function, and outputs a first result through the data output interface, the first result being the frequency response estimation value of all frequency points within the simulation frequency range obtained by the third interpolation function.
[0010] In one implementation, when the error convergence standard value or the root mean square error is less than a preset value, the data output interface outputs a second result, where the second result is a frequency response estimate of all frequency points within the simulation frequency range obtained by the second interpolation function.
[0011] In one implementation, the processor further selects an interval with the largest interval interval in the first sample frequency, and inserts a frequency point by binary division; and the processor uses the first sample frequency with the inserted frequency point as the second sample frequency.
[0012] In a third aspect, the present application further provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above steps of implementing the frequency scanning method.
[0013] Compared with the prior art, the present invention has at least one of the following beneficial effects:
[0014] 1. By constructing an accurate rational interpolation algorithm, dynamically adjusting it, and using an interval-based error analysis algorithm, the accuracy of frequency response prediction and error evaluation of frequency scanning results is improved. When the error convergence standard value is not less than the preset value, frequency points are inserted in the interval with the largest error to construct a new sample frequency. Compared with the method of adjusting the frequency point based on the error between two adjacent frequency points, the adaptability and reliability of frequency scanning are improved, the number of unnecessary frequency points is further reduced, and the efficiency of frequency scanning is improved.
[0015] 2. By constructing the first interpolation function and the second interpolation function and performing error analysis on the interval of the first interpolation function and the second interpolation function, the accuracy of frequency response prediction and error evaluation of frequency scanning results is further improved, and the allocation of computing resources is optimized. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The preferred implementation modes will be described below in a clear and understandable manner with reference to the accompanying drawings to further illustrate the above-mentioned characteristics, technical features, advantages and implementation methods of the present invention.
[0017] Figure 1 A flow chart of a frequency scanning method provided by an embodiment of the present application is shown;
[0018] Figure 2 A flowchart of constructing an interpolation function by a centroid rational interpolation algorithm provided in an embodiment of the present application is shown;
[0019] Figure 3 A flow chart of applying a frequency scanning method to an electromagnetic simulation test provided by an embodiment of the present application is shown;
[0020] Figure 4 A microstrip line model diagram provided by an embodiment of the present application is shown;
[0021] Figure 5 A schematic diagram showing a standard frequency response curve and its error provided by an embodiment of the present application is shown;
[0022] Figure 6 A schematic diagram of a frequency response curve and its error based on a minimum rational interpolation algorithm provided in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the specific implementation methods of the present invention will be described below with reference to the accompanying drawings. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings and other implementation methods can be obtained based on these drawings without creative work.
[0024] In order to simplify the drawings, only the parts related to the invention are schematically shown in each figure, and they do not represent the actual structure of the product. In addition, in order to simplify the drawings and facilitate understanding, in some figures, only one of the parts with the same structure or function is schematically drawn or marked. In this article, "one" not only means "only one", but also means "more than one".
[0025] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.
[0026] In this document, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0027] In addition, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0028] It should be noted that the above embodiments can be freely combined as needed. The above are only preferred embodiments of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.
[0029] The frequency scanning algorithm has evolved from the traditional fixed interpolation or fixed scanning strategy to the adaptive frequency scanning algorithm. The frequency scanning algorithm using fixed interpolation or fixed scanning strategy achieves the final frequency response by scanning a large number of fixed frequency points in the entire frequency range.
[0030] Adaptive frequency scanning algorithm analyzes the frequency response of the circuit or system in real time and dynamically adjusts the measurement frequency points to optimize the test process and reduce the number of necessary test frequency points, thereby improving the simulation accuracy. Adaptive frequency scanning algorithm has been applied in many fields. For example, in electromagnetic simulation, the adaptive frequency scanning algorithm reduces the number of necessary test points by intelligently selecting frequency points and maintains the accuracy of the frequency scanning results. In signal processing, the adaptive frequency scanning algorithm can more effectively capture the key characteristics of the signal by changing the speed or interval of the frequency scanning and dynamically adjusting the scanning process according to the characteristics of the signal. In electronic testing and measurement, the adaptive frequency scanning algorithm is used to improve the efficiency and accuracy of measurement equipment. For example, in a network analyzer, the adaptive frequency scanning algorithm can automatically adjust the density of frequency points according to the characteristic response of the circuit. In the testing of high-frequency circuits or antennas, the adaptive frequency scanning algorithm can ensure intensive testing at key resonant frequencies or impedance change frequencies, while reducing test points in other frequency ranges, optimizing the test process and improving the test speed. In addition, the adaptive frequency scanning algorithm is particularly important in audio analysis, wireless communication and radar systems. It can automatically adjust the scanning accuracy at different frequency points to adapt to the dynamic changes of the signal.
[0031] The adaptive frequency scanning algorithm performs frequency scanning by intelligently selecting frequency points, and obtains the frequency responses of all frequency points within the frequency range through the interpolation function (or interpolation parameters, hereinafter referred to as the interpolation function) according to the existing frequency response, and performs error analysis and convergence judgment. When the convergence judgment is passed, the frequency responses of all frequency points are output. The adaptive frequency scanning algorithm is applicable to a variety of interpolation functions, a variety of error analysis functions, and a variety of convergence judgment criteria. Moreover, different interpolation functions, different error analysis algorithms, and different convergence judgment criteria will result in different output results. Therefore, the embodiments of the present application can achieve at least one of the following beneficial effects by improving the interpolation function, the error analysis function, and the convergence judgment criterion: dynamically adjusting the interpolation function to more accurately predict the frequency response; or based on the interval-based error analysis algorithm, more accurately evaluating the error of the frequency scanning result, thereby improving the compatibility of the frequency scanning.
[0032] The following is a description with reference to the accompanying drawings:
[0033] Reference Figure 1 , which shows a flow chart of a frequency scanning method provided by an embodiment of the present application, such as Figure 1 As shown, including:
[0034] S100, determining a first sample frequency and a second sample frequency based on a preset simulation frequency range, wherein the second sample frequency includes the first sample frequency, and the number of frequency points of the second sample frequency is greater than the number of frequency points of the first sample frequency;
[0035] S110, constructing a first interpolation function based on an interpolation algorithm, a first sample frequency and a target circuit model, and constructing a second interpolation function based on an interpolation algorithm, a second sample frequency and a target circuit model;
[0036] S120, performing error analysis on the first interpolation function and the second interpolation function by using an error analysis algorithm, obtaining errors in each interval of the second sample frequency, and determining an error convergence standard value or a root mean square error according to the errors in each interval;
[0037] S130, when the error convergence standard value or the root mean square error is not less than a preset value, insert at least one frequency point in the interval with the largest error in the second sample frequency, use the second sample frequency after the inserted frequency point as the third sample frequency, construct a third interpolation function for the third sample frequency, perform error analysis and error convergence judgment steps on the second interpolation function and the third interpolation function, and output a first result, which is a frequency response estimation value of all frequency points within the simulation frequency range obtained by the third interpolation function.
[0038] Before frequency scanning, first set the basic parameters of the target circuit model and the simulation frequency range. The basic parameters of the target circuit model can be geometric information and physical parameters. The simulation frequency range (the simulation frequency range is a wide frequency range) can be (in, The value can be set according to the actual situation). In addition, the number of output frequency points and convergence error can also be set.
[0039] A plurality of frequency points are selected from the simulation frequency range to form a first sample frequency, and the first sample frequency includes the minimum frequency point in the simulation frequency range (in the embodiment of the present application, ) and the maximum frequency point (in this embodiment of the application is ). Multiple frequency points in the first sample frequency constitute multiple intervals, select the interval with the largest interval interval, insert at least one frequency point, and use the first sample frequency after adding the frequency point as the second sample frequency. Based on the interpolation algorithm, the basic parameters of the target circuit model, the first sample frequency and the second sample frequency, a first interpolation function and a second interpolation function are constructed respectively. Among them, the interpolation functions mentioned in the embodiments of the present application are all used to obtain the frequency response estimation values of all frequency points within the simulation frequency range.
[0040] The formula for the error analysis algorithm is: , where the second interpolation function is substituted into , the first interpolation function is substituted into , is the error of each interval in the second sample frequency. By substituting the second interpolation function and the first interpolation function into the formula of the error analysis algorithm, the error of each interval in the second sample frequency is obtained, and the error convergence standard value or the root mean square error is determined. It is judged whether the error convergence standard value or the root mean square error is less than the preset value. If the error convergence standard value or the root mean square error is not less than the preset value, then the interval with the largest error is selected from multiple intervals of the second sample frequency and at least one frequency point is inserted to generate a third sample frequency. A third interpolation function is constructed for the third sample frequency, and then the error analysis is performed on the third interpolation function and the second interpolation function (at this time is the third interpolation function, is the second interpolation function, is the error of each interval in the third sample frequency), obtain the error of each interval in the third sample frequency, and re-determine the error convergence standard value or root mean square error. It is judged again whether the error convergence standard value or root mean square error is less than the preset value. If it is less than the preset value, the frequency response estimation value (or transmission scattering parameter (Scattering Parameter)) of all frequency points in the simulation frequency range obtained by the third interpolation function is output. If the error convergence standard value or root mean square error is not less than the preset value, at least one frequency point is inserted in the interval with the largest error in the third sample frequency, and the third sample frequency after the insertion of the frequency point is used as the fourth sample frequency, that is, the aforementioned process of inserting frequency points in the maximum error interval to construct a new sample frequency, constructing an interpolation function, error analysis, and comparing the error convergence standard value with the preset value is continuously iterated until the error convergence standard value or the root mean square error is less than the preset value.
[0041] It is conceivable that in the process of inserting frequency points in the embodiment of the present application, there may be a situation where the number of intervals with the largest interval intervals is not unique, and / or the number of intervals with the largest error is not unique. In either case, one of the intervals can be selected to insert at least one frequency point, or multiple intervals can be selected to insert at least one frequency point. There are many ways to insert frequency points in the embodiment of the present application, for example, including but not limited to dichotomy, golden section method, model-based method, uniform distribution method, statistical-based method, etc.
[0042] The embodiment of the present application improves the accuracy of frequency response prediction and error evaluation of frequency scanning results by constructing an accurate rational interpolation algorithm, dynamically adjusting it, and performing an interval-based error analysis algorithm. When the error convergence standard value or the root mean square error is not less than the preset value, a frequency point is inserted in the interval with the largest error to construct a new sample frequency. Compared with the method of adjusting the frequency point based on the error of two adjacent frequency points, the adaptability and reliability of frequency scanning are improved, the number of unnecessary frequency points is further reduced, and the efficiency of frequency scanning is improved.
[0043] The embodiments of the present application improve the accuracy of frequency response prediction and error evaluation of frequency scanning results by dynamically adjusting the interpolation function and the interval-based error analysis algorithm. Among them, all algorithms that can construct interpolation functions can be adopted by the embodiments of the present application, including interpolation algorithms based on rational functions, such as centroid rational interpolation algorithm and minimum rational interpolation algorithm. Other interpolation algorithms include but are not limited to model-based parameter estimation algorithms. Take the centroid rational interpolation algorithm and the minimum rational interpolation algorithm as examples, refer to the attached Figure 2 , which illustrates a flowchart of constructing an interpolation function provided in an embodiment of the present application, such as Figure 2 As shown, including:
[0044] S200, calculating a first frequency response actual value corresponding to each frequency point in the first sample frequency according to the first sample frequency and the target circuit model.
[0045] S210: Construct a first data set according to the first sample frequency and the first frequency response actual value.
[0046] S220, performing interpolation processing on the first data set by using a centroid rational interpolation algorithm or a minimum rational interpolation algorithm to obtain a first interpolation function.
[0047] S230, calculating a second frequency response actual value corresponding to each frequency point in the second sample frequency according to the second sample frequency and the target circuit model.
[0048] S240: Construct a second data set according to the second sample frequency and the second frequency response actual value.
[0049] S250, performing interpolation processing on the second data set by using a centroid rational interpolation algorithm or a minimum rational interpolation algorithm to obtain a second interpolation function.
[0050] Taking the centroid rational interpolation algorithm as an example, after determining the first sample frequency, select the interval with the largest interval in the first sample frequency, and insert a frequency point through binary division to obtain the second sample frequency. Input the frequency points in the first sample frequency into the target circuit model to obtain the accurate electromagnetic response of the target circuit model, that is, to obtain the actual value of the first frequency response corresponding to each frequency point. Through the formula The first sample frequency and the first frequency response actual values corresponding to each frequency point in the first sample frequency are constructed as a first data set. can be a frequency point in the first sample frequency, It can be the actual value of the first frequency response corresponding to each frequency point in the first sample frequency. , for the constructed dataset Points: Among them, the highest order is The interpolation polynomial is Represents. Taking the weighted average of all points in the data set and the interpolation polynomials corresponding to each point, the centroid rational interpolation function It can be expressed as Thus, by taking a weighted average of all the points in the first data set and the interpolation polynomials corresponding to each point, the first interpolation function can be obtained. Similarly, the second interpolation function can be obtained through the same process . Get the first interpolation function and the second interpolation function Then, substitute it into the formula Perform error analysis to obtain the error of each interval of the second sample frequency and determine the error convergence standard value. When the error convergence standard value is less than the preset value, the output is passed through the second interpolation function The frequency response estimate of the frequency point within the simulation frequency range is obtained. If the error convergence standard value is not less than the preset value, then the formula , determine the interval with the largest error in the second sample frequency, where Indicates the interval corresponding to the maximum error. The third sample frequency is obtained by inserting a frequency point in the interval with the largest error in the second sample frequency through the dichotomy method, and the processes of constructing the interpolation function, error analysis, and comparing the error convergence standard value with the preset value are repeated until the error convergence standard value is less than the preset value.
[0051] Taking the minimum rational interpolation algorithm as an example, after determining the first sample frequency, select the interval with the largest interval in the first sample frequency, and insert a frequency point through binary search to obtain the second sample frequency. Input the frequency points in the first sample frequency into the target circuit model to obtain the accurate electromagnetic response of the target circuit model, that is, to obtain the actual value of the first frequency response corresponding to each frequency point. Through the formula The first sample frequency and the first frequency response actual values corresponding to each frequency point in the first sample frequency are constructed as a first data set. can be the Laplace frequency point in the first sample frequency, It can be the actual value of the first frequency response corresponding to each frequency point in the first sample frequency. Indicates that is the Laplace frequency point:
[0052]
[0053] According to the rational function model:
[0054]
[0055] Then we can get the system of equations:
[0056]
[0057]
[0058] make:
[0059]
[0060]
[0061]
[0062]
[0063] So as to construct the matrix equation:
[0064]
[0065] in:
[0066]
[0067] Using the optimized minimum interpolation algorithm to solve equation (3), we can obtain the lowest MacMillan order and :
[0068]
[0069]
[0070] The polynomial So that:
[0071]
[0072] Therefore, the first interpolation function can be obtained through formulas (1), (2), and (3): Similarly, the second interpolation function can be obtained through the same process . Get the first interpolation function and the second interpolation function Then, substitute it into the formula Perform error analysis to obtain the error of each interval of the second sample frequency, where the second interpolation function is substituted into , the first interpolation function is substituted into , is the error of each interval in the second sample frequency. After obtaining the error of each interval of the second sample frequency, the formula Determine the RMS error and the interval with the largest error in the second sample frequency, where is the root mean square error, is the average error of each interval of the second sample frequency.
[0073] When the RMS error is less than the preset value, the output is passed through the second interpolation function Obtain the frequency response estimate of the frequency point within the simulation frequency range. If the RMS error is not less than the preset value, insert a frequency point in the interval with the largest error in the second sample frequency by bisection to obtain the third sample frequency, and repeat the process of constructing the interpolation function, error analysis, and comparing the RMS error with the preset value until the RMS error is less than the preset value.
[0074] In the embodiment of the present application, the centroid rational interpolation algorithm can be used to construct the interpolation function, and the frequency points can be added by the root mean square error to combine; the minimum rational interpolation algorithm can also be used to construct the interpolation function, and the frequency points can be added by the error convergence standard value to combine. In addition, the preset values used for judging the root mean square error and the error convergence standard value in the embodiment of the present application can be set according to the actual situation and are not limited here.
[0075] In the embodiment of the present application, by constructing a first interpolation function and a second interpolation function, an error analysis is performed on the intervals of the first interpolation function and the second interpolation function, thereby further improving the accuracy of frequency response prediction and error evaluation of frequency scanning results, and optimizing the allocation of computing resources.
[0076] In some embodiments of the present application, it also includes: when the error convergence standard value or the root mean square error is less than a preset value, outputting a second result, the second result being a frequency response estimate of a frequency point within the simulation frequency range obtained by a second interpolation function.
[0077] In some embodiments of the present application, determining the first sample frequency and the second sample frequency based on a preset simulation frequency range specifically includes: selecting the interval with the largest interval in the first sample frequency, inserting frequency points by binary division, and using the first sample frequency with the inserted frequency points as the second sample frequency.
[0078] In some embodiments of the present application, the first sample frequency includes a minimum frequency point and a maximum frequency point of a simulation frequency range.
[0079] Reference Figure 3 , which shows a flow chart of applying a frequency scanning method to an electromagnetic simulation test provided by an embodiment of the present application, such as Figure 3 As shown, including:
[0080] S300, configure the system to test frequency response.
[0081] The system includes an integrated circuit simulation platform equipped with a processor, a memory, and an interface for data input and result output. In addition, the system has a built-in software module for solving the target circuit model, which is used to calculate the actual value of the frequency response (or precise electromagnetic response) of the target circuit model: for each input frequency point, the corresponding actual value of the frequency response (or S11 parameter) is output. , where the S11 parameter, also known as the scattering parameter or reflection coefficient, is one of the key parameters used in RF and microwave engineering to describe the signal reflection and transmission characteristics between circuit or system ports). All frequency points and their corresponding actual values of frequency response constitute the data set.
[0082] S310, setting basic parameters of the target circuit model and defining a preset range of simulation frequencies.
[0083] Before the electromagnetic simulation test begins, the basic parameters of the target circuit model and the expected test frequency range are set. The target circuit model parameters include the geometric information and physical parameters of the model. Other parameters that control the frequency sweep can also be set, such as the sweep mode selection (interpolation sweep or discrete sweep), the number of output frequency points, the convergence error, etc. These parameters can be set according to user needs. The preset simulation frequency range is .
[0084] S320, constructing an interpolation function through a centroid rational interpolation algorithm.
[0085] The system selects the two endpoints of the preset simulation frequency range and or preset the minimum and maximum frequency points of the simulation frequency range), and and Insert three frequency points between , , (The three inserted frequency points can be selected according to user needs), thus forming the first sample frequency. and Insert frequency points between , thereby forming a second sample frequency. The built-in software module of the system calculates the actual value of the first frequency response of the target circuit model corresponding to the first sample frequency and the actual value of the second frequency response of the target circuit model corresponding to the second sample frequency, respectively, thereby constructing the first data set and the second data set. The system interpolates between the first data set and the second data set using the centroid rational interpolation algorithm to obtain the first interpolation function and the second interpolation function .
[0086] S330, performing error analysis and error convergence judgment on the interpolation function based on an interval integration error analysis algorithm.
[0087] The first interpolation function and the second interpolation function The error and error convergence standard value of each interval of the second sample frequency are obtained. When the error convergence standard value is less than the preset value, the output is passed through the second interpolation function Obtain the estimated frequency response value of the frequency point within the simulation frequency range (actually output the frequency response curve). When the error convergence standard value is not less than the preset value, the system inserts a new frequency point in the interval with the largest error in the second sample frequency by dichotomy to form the third sample frequency. The system will repeat the process of constructing the interpolation function and error analysis, comparing the error convergence standard value with the preset value, etc. for the third sample frequency until the error convergence standard value is less than the preset value and outputs a frequency response curve that meets the requirements.
[0088] In the embodiment of the present application, the process of using the minimum rational interpolation algorithm is the same as the process of using the centroid rational interpolation algorithm, except that the algorithm for constructing the interpolation function and the standard for error convergence judgment are different. The differences have been explained in detail in the previous embodiments and will not be repeated here.
[0089] The model diagram of the embodiment of this application is shown in the attached Figure 4 , which illustrates a longer microstrip line model.
[0090] Reference Figure 5 , which shows a schematic diagram of a standard frequency response curve and its error. The embodiment of the present application takes the return loss S11 parameter as an example. The solid line represents the simulation result of the discrete frequency sweep of the industry benchmark commercial software, and the dotted line represents the simulation result of its interpolation frequency sweep. The ordinate on the right represents the absolute value of the absolute error between the two, represented by the dotted line. The absolute error represented by the dotted line parallel to the horizontal axis is 0.03. The red dots represent the inserted frequency points. Under this model, the number of frequency points inserted by the commercial software is 64.
[0091] Reference Figure 6 , which shows a frequency response (S11 parameter) curve and its error diagram based on the minimum rational interpolation algorithm provided in the embodiment of the present application, in which the discrete frequency sweep result is the same as the simulation result of the discrete frequency sweep of the commercial software. It can be seen that in the embodiment of the present application, the area where the interpolation frequency sweep absolute error is greater than 0.03 is much smaller than the standard result, and the overall absolute error is mainly concentrated in arrive The absolute error of the standard result is mainly concentrated in arrive This shows that the simulation results of the embodiment of the present application are more accurate than the standard results. In addition, the number of frequency points sampled by the embodiment of the present application under this model is 56, which is less than the standard number of frequency points of 64, and the speed is improved by about 12.5%. It can be seen that although the frequency response curve varies, the performance of the frequency scanning method provided by the embodiment of the present application is still excellent, and the accuracy is also improved within the wide frequency range of 0.05GHz to 140GHz while improving the speed. This shows that the frequency scanning method provided by the embodiment of the present application is more compatible, sufficient to be applicable to a wide frequency range, and can cope with different models and other complex conditions.
[0092] The embodiment of the present application also provides a frequency scanning system for executing any of the above frequency scanning methods, including: a processor, a data input interface and a data output interface; the data input interface is used to receive a simulation frequency range and a target circuit model input by a user; the processor determines a first sample frequency and a second sample frequency based on the simulation frequency range input by the user, the second sample frequency includes the first sample frequency, and the number of frequency points of the second sample frequency is greater than the number of frequency points of the first sample frequency; the processor constructs a first interpolation function based on a centroid rational interpolation algorithm, the first sample frequency and the target circuit model, and constructs a second interpolation function based on a centroid rational interpolation algorithm, the second sample frequency and the target circuit model; the processor An error analysis algorithm is used to perform error analysis on the first interpolation function and the second interpolation function to obtain errors in each interval of the second sample frequency, and an error convergence standard value is determined based on the errors in each interval; when the error convergence standard value is not less than a preset value, the processor inserts at least one frequency point in the interval with the largest error in the second sample frequency, uses the second sample frequency after the inserted frequency point as the third sample frequency, constructs a third interpolation function for the third sample frequency, performs error analysis and error convergence judgment steps on the second interpolation function and the third interpolation function, and outputs a first result through a data output interface, the first result being an estimated value of the frequency response of all frequency points within the simulation frequency range obtained by the third interpolation function.
[0093] In some embodiments of the present application, it also includes: a frequency response calculation module; the frequency response calculation module calculates the first frequency response actual value corresponding to each frequency point in the first sample frequency according to the first sample frequency and the target circuit model; the processor constructs a first data set according to the first sample frequency and the first frequency response actual value; the processor interpolates the first data set through the centroid rational interpolation algorithm to obtain a first interpolation function; the frequency response calculation module calculates the second frequency response actual value corresponding to each frequency point in the second sample frequency according to the second sample frequency and the target circuit model; the processor constructs a second data set according to the second sample frequency and the second frequency response actual value; the processor interpolates the second data set through the centroid rational interpolation algorithm to obtain a second interpolation function.
[0094] In some embodiments of the present application, when the error convergence criterion value is less than a preset value, the data output interface outputs a second result, where the second result is a frequency response estimation value of all frequency points within the simulation frequency range obtained by a second interpolation function.
[0095] In some embodiments of the present application, the processor further selects an interval with the largest interval interval in the first sample frequency, and inserts a frequency point by binary division; the processor uses the first sample frequency with the inserted frequency point as the second sample frequency.
[0096] A processor is a circuit with signal processing capabilities. In one example, a processor may be a circuit with instruction reading and execution capabilities; for example, a central processing unit (CPU), a microcontroller (MCU), a microprocessor (MPU), a graphics processing unit (GPU), or a digital signal processor (DSP). In another example, a processor may realize its processing capabilities through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable; for example, a processor may be a dedicated processor, such as a processor implemented by an application specific integrated circuit (ASIC), which realizes processing capabilities by designing the logical relationship of components in the circuit; for another example, a processor implemented by a programmable logic device (PLD) configures the logical relationship between logic devices through a configuration file to realize processing capabilities; for example, a processor implemented by a field programmable gate array (FPGA). In another example, a processor may be a hardware circuit designed for artificial intelligence, which may be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc. This application is not limited by the type of processor.
[0097] Based on similar technical concepts, an embodiment of the present application further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned frequency scanning methods are performed.
[0098] It should be noted that the above embodiments can be freely combined as needed. The above are only preferred embodiments of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be regarded as the protection scope of the present invention.
Claims
1. A frequency scanning method, characterized in that: include: Determine a first sample frequency and a second sample frequency based on a preset simulation frequency range, wherein the second sample frequency includes the first sample frequency, and the number of frequency points of the second sample frequency is greater than the number of frequency points of the first sample frequency, and the preset simulation frequency range is a wide frequency range; Calculating a first frequency response actual value corresponding to each frequency point in the first sample frequency according to the first sample frequency and the target circuit model; Constructing a first data set according to the first sample frequency and the first frequency response actual value; Performing interpolation processing on the first data set by using a centroid rational interpolation algorithm or a minimum rational interpolation algorithm to obtain a first interpolation function; Calculating a second frequency response actual value corresponding to each frequency point in the second sample frequency according to the second sample frequency and the target circuit model; constructing a second data set according to the second sample frequency and the second frequency response actual value; Performing interpolation processing on the second data set by using the centroid rational interpolation algorithm or the minimum rational interpolation algorithm to obtain a second interpolation function; Performing error analysis on the first interpolation function and the second interpolation function by using an error analysis algorithm to obtain errors in each interval of the second sample frequency, and determining an error convergence standard value or a root mean square error according to the errors in each interval; When the error convergence standard value or the root mean square error is not less than a preset value, at least one frequency point is inserted in the interval with the largest error in the second sample frequency, the second sample frequency after the inserted frequency point is used as the third sample frequency, a third interpolation function is constructed for the third sample frequency, error analysis and error convergence judgment steps are performed on the second interpolation function and the third interpolation function, and a first result is output, where the first result is a frequency response estimation value of all frequency points within the simulation frequency range obtained by the third interpolation function.
2. The frequency scanning method according to claim 1, characterized in that: Also includes: When the error convergence standard value or the root mean square error is less than the preset value, a second result is output, where the second result is a frequency response estimation value of all frequency points within the simulation frequency range obtained by the second interpolation function.
3. The frequency scanning method according to claim 1, characterized in that: The determining of the first sample frequency and the second sample frequency based on the preset simulation frequency range specifically includes: Select the interval with the largest interval in the first sample frequency, and insert the frequency point by binary division; The first sample frequency of the inserted frequency point is used as the second sample frequency.
4. The frequency scanning method according to any one of claims 1 to 3, characterized in that: The first sample frequency includes a minimum frequency point and a maximum frequency point of the simulation frequency range.
5. A frequency scanning system, characterized in that: include: A processor, a data input interface, a data output interface and a frequency response calculation module; The data input interface is used to receive the simulation frequency range and target circuit model input by the user; The processor determines a first sample frequency and a second sample frequency based on a simulation frequency range input by a user, wherein the second sample frequency includes the first sample frequency, and the number of frequency points of the second sample frequency is greater than the number of frequency points of the first sample frequency; The frequency response calculation module calculates the first frequency response actual value corresponding to each frequency point in the first sample frequency according to the first sample frequency and the target circuit model; The processor constructs a first data set according to the first sample frequency and the first frequency response actual value; The processor performs interpolation processing on the first data set by using a centroid rational interpolation algorithm or a minimum rational interpolation algorithm to obtain a first interpolation function; The frequency response calculation module calculates the second frequency response actual value corresponding to each frequency point in the second sample frequency according to the second sample frequency and the target circuit model; The processor constructs a second data set according to the second sample frequency and the second frequency response actual value; The processor performs interpolation processing on the second data set by using the centroid rational interpolation algorithm or the minimum rational interpolation algorithm to obtain a second interpolation function; The processor performs error analysis on the first interpolation function and the second interpolation function by using an error analysis algorithm, obtains errors in each interval of the second sample frequency, and determines an error convergence standard value or a root mean square error according to the errors in each interval; When the error convergence standard value or the root mean square error is not less than a preset value, the processor inserts at least one frequency point in the interval with the largest error in the second sample frequency, uses the second sample frequency after the inserted frequency point as the third sample frequency, constructs a third interpolation function for the third sample frequency, performs error analysis and error convergence judgment steps on the second interpolation function and the third interpolation function, and outputs a first result through the data output interface, where the first result is a frequency response estimation value of all frequency points within the simulation frequency range obtained by the third interpolation function.
6. The frequency scanning system according to claim 5, characterized in that: When the error convergence standard value or the root mean square error is less than a preset value, the data output interface outputs a second result, where the second result is a frequency response estimation value of all frequency points within the simulation frequency range obtained by the second interpolation function.
7. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the frequency scanning method according to any one of claims 1 to 4 are implemented.
