A method for obtaining characteristics of electronic devices and related products

Abstract

The present disclosure discloses a method for acquiring features of an electronic device and related products, wherein the features include a parameterized function of a target variable with respect to a feature variable; the method includes: calculating a target variable simulation value based on feature variable data based on a first parameter; determining an error calculation method based on a value range of the target variable data associated with the feature variable data; calculating the error between the target variable simulation value and the target variable data based on the error calculation method; and updating the first parameter based on the error to obtain a second parameter. The method adaptively selects an error calculation method based on the value range of the target variable, thereby avoiding the problem that a point with a larger order of magnitude has a larger weight when calculating the overall error, thereby improving the accuracy of curve fitting.

Description

Technical Field

[0001] The present disclosure generally relates to the field of electronic information technology. More specifically, the present disclosure relates to a method for acquiring features of an electronic device and related products. Background Art

[0002] Metal-Oxide-Semiconductor Field-Effect Transistor (MOSFET) is a transistor used to control the flow of current in a circuit. It can be used as a switch or amplifier. Because it can handle large currents and high voltages, MOSFET is widely used in small digital circuits and large power systems. In order to facilitate the use of MOSFET, manufacturers will produce data sheets for the MOSFET they produce to explain the various characteristic parameters, electrical performance, application conditions and other detailed information of the product. The MOSFET characteristic curve is important information in the datasheet, including transfer characteristic curve, output characteristic curve, substrate current characteristic curve, gate current characteristic curve, transconductance characteristic curve, drain breakdown characteristic curve, temperature characteristic curve, etc.

[0003] Usually, the characteristic curve is obtained through curve fitting technology. Fitting the characteristic curve of MOSFET is a cyclic iterative process. In each iteration: first, the curve function is used to calculate the target variable simulation value for the current characteristic variable batch data set; then the overall error is calculated based on the target variable simulation value and the current target variable batch data set; then, the overall error and the number of current iterations are used to determine whether the loop termination condition is met. If the termination condition is met, the loop can be terminated. Otherwise, an optimizer (such as a gradient descent algorithm) is used to adjust the function parameters according to the overall error and enter the next loop.

[0004] Calculating the overall error based on the target variable simulation value and the current target variable batch data set usually includes two subdivision steps, namely, first calculating the point error of the target variable simulation value based on each data in the current target variable batch data set, and then using weighted summation, averaging or other statistical methods to calculate the point error to obtain the overall error. In the prior art, a single error calculation method is often used to calculate the point error during the curve fitting process, which results in that when the distribution of the target variable value is wide, the optimization effect of the optimizer on the data points with larger absolute values ​​is not symmetrical with the optimization effect on the data points with smaller absolute values.

[0005] In view of this, there is an urgent need to provide a solution for obtaining characteristic curves for electronic devices, so as to select an error calculation method according to the actual characteristics of the data set and improve the accuracy of curve fitting. Summary of the invention

[0006] In order to at least solve one or more technical problems described in the above background technology section, the present disclosure proposes the following technical solutions and multiple embodiments thereof.

[0007] In a first aspect, the present disclosure discloses a method for acquiring features of an electronic device, wherein the features include a parameterized function of a target variable with respect to a feature variable; the method includes: calculating a target variable simulation value according to feature variable data based on a first parameter; determining an error calculation method according to a value range of the target variable data associated with the feature variable data; calculating an error between the target variable simulation value and the target variable data according to the error calculation method; and updating the first parameter according to the error to obtain a second parameter.

[0008] In a second aspect, the present disclosure discloses a computer-readable storage medium, wherein program instructions are stored, and the program instructions are suitable for being loaded by a processor and executing the method according to the first aspect.

[0009] In a third aspect, the present disclosure discloses an apparatus for acquiring features of an electronic device, comprising: a processor configured to execute program instructions; and a memory configured to store the program instructions, which, when loaded and executed by the processor, enables the apparatus to execute the method described in the first aspect.

[0010] The method for acquiring features of electronic devices disclosed in the present disclosure adaptively selects an error calculation method according to the value range of the target variable, thereby avoiding the problem that points with larger magnitudes have larger weights when calculating the overall error, thereby improving the accuracy of curve fitting. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0012] Figure 1 An exemplary schematic diagram of a MOSFET structure in some embodiments of the present disclosure is shown.

[0013] Figure 2 An exemplary schematic diagram of characteristic curves in some embodiments of the present disclosure is shown.

[0014] Figure 3 An exemplary schematic diagram of obtaining a characteristic curve by using a curve fitting technique in some embodiments of the present disclosure is shown.

[0015] Figure 4An exemplary schematic diagram showing target variable values ​​in some embodiments of the present disclosure is shown.

[0016] Figure 5 An exemplary flow chart of a method for acquiring features of an electronic device in some embodiments of the present disclosure is shown.

[0017] Figure 6 An exemplary flow chart of a method for acquiring features of an electronic device in some embodiments of the present disclosure is shown.

[0018] Figure 7 The block diagram shows the hardware configuration of the apparatus that can implement the embodiments of the present disclosure. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] It should be understood that the terms "include" and "comprises" used in the description and claims of the present invention indicate the presence of described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0021] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the claims, the singular forms of "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in the specification of the present invention and the claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0022] As used in this specification and claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0023] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0024] A MOSFET is a transistor used to control the flow of current in a circuit. Figure 1 An exemplary schematic diagram of the MOSFET structure in some embodiments of the present disclosure is shown. MOSFET is built on a substrate; the substrate can be a silicon wafer, which constitutes the physical basis of the MOSFET device; through different weak doping processes, the substrate can be a P-doped substrate or an N-doped substrate. The substrate includes two special regions: a source region and a drain region; further, an N-doped source / drain region can be created in a P-doped substrate by a strong doping process, and a P-doped source / drain region can be created in an N-doped substrate by a strong doping process. A carrier channel, a channel, is formed between the source region and the drain region for the flow of carriers. The P-doped substrate and the N-doped source / drain can form an N-channel for the flow of electrons, and such a MOSFET is also called an N-channel MOSFET. The N-doped substrate and the P-doped source / drain can form a P-channel for the flow of holes, and such a MOSFET is also called a P-channel MOSFET. A layer of oxide (such as silicon dioxide SiO2) can be formed on the surface of the substrate, and the drain and source pins of the MOSFET can be led out through the oxide layer, and the gate pin of the MOSFET can be led out through the oxide layer and the metal (such as aluminum Al) layer thereon. Figure 1 As shown, the gate is connected to the substrate through a metal layer and an oxide layer; the drain is connected to a drain region in the substrate through an oxide layer; and the source is connected to a source region in the substrate through an oxide layer.

[0025] MOSFET can be used as a switch or amplifier; because it can handle high current and high voltage, MOSFET is widely used in small digital circuits and large power systems. In order to facilitate the use of MOSFET, manufacturers will produce data manuals (Datasheet) for the MOSFET they produce, which is used to explain the various characteristic parameters, electrical performance, application conditions and other detailed information of the product; it can be understood that the Datasheet is an important reference for selecting and using MOSFET when designing complex circuits. The MOSFET characteristic curve is important information in the Datasheet; generally speaking, the Datasheet contains a characteristic curve library composed of a series of characteristic curves, including transfer characteristic curves, output characteristic curves, substrate current characteristic curves, gate current characteristic curves, transconductance characteristic curves, drain breakdown characteristic curves, temperature characteristic curves, etc.

[0026] Figure 22 shows an exemplary schematic diagram of characteristic curves in some embodiments of the present disclosure. It can be understood that the carriers flowing from the source through the channel to the drain constitute the channel current, which can also be regarded as the drain current (I D ). When the gate-source voltage changes, the MOSFET can be controlled to be in the cut-off state, the on state, and the saturation state. For example, taking an N-channel enhancement MOSFET as an example: let the voltage between the gate and the source be V GS , the voltage between the drain and source is V DS ; When V DS = 0 and V GS <V TH When MOSFET is in the off state, I D =0, where V TH Indicates the turn-on voltage; when V GS ≥V TH And V DS When the MOSFET is on, I D >0 and I D The size of V DS Linear change; when V GS ≥V TH When V DS When it increases to a certain extent, MOSFET will gradually transition to saturation state. In the transition state, I D The size of V DS Instead of linear growth, the growth slows down and tends to remain constant. Based on the above changes, we can get the following Figure 2 The transfer characteristic curve shown is used to describe I D With V GS and the output characteristic curve, which is used to describe the relationship between I D With V DS The relationship between. Figure 2 The characteristic curve of MOSFET is introduced by taking the output characteristic curve and transfer characteristic curve as examples. It can be understood that the characteristic curve can be used to describe the changing relationship between the target variable and the characteristic variable under certain conditions. The characteristic variable can also be called the independent variable, such as Figure 2 V GS With V DS ; The target variable can also be called the dependent variable, such as Figure 2 I D ; The conditions include the device working environment conditions and the device physical structure parameter conditions.

[0027] Usually, the characteristic curve is obtained by curve fitting technology. Figure 3An exemplary schematic diagram of obtaining a characteristic curve by curve fitting technology in some embodiments of the present disclosure is shown. It can be understood that the characteristic curve can be expressed as a parameterized relationship function y=f of the target variable with respect to the characteristic variable. θ (x), where y represents the target variable, x represents the characteristic variable, and θ represents the parameter to be confirmed. The parameters of the relationship function can be continuously adjusted based on a large amount of actual data and optimization methods. To fit the characteristic curve of MOSFET, a large-scale data set needs to be obtained. Each data in the data set includes characteristic variable measurement data and corresponding target variable measurement data. By dividing the data set, multiple batches of batch data sets can be obtained, each of which contains multiple data. Furthermore, each batch data set can include a characteristic variable batch data set composed of characteristic variable data and a target variable batch data set composed of target variable data. The characteristic curve can be expressed by a polynomial function, and the parameters are initialized based on random numbers or empirical values. Then, an iterative cycle process is entered. In each iteration: first, the curve function is used to calculate the target variable simulation value for the current characteristic variable batch data set; then, the overall error is calculated based on the target variable simulation value and the current target variable batch data set. The overall error is used to measure the fitting effect of the curve function; based on the overall error and the number of current iterations, it can be judged whether the loop termination condition is reached. If the termination condition is met, the loop can be terminated. Otherwise, an optimizer (such as a gradient descent algorithm) is used to adjust the function parameters according to the overall error and enter the next cycle.

[0028] exist Figure 3 In the method of obtaining characteristic curves by curve fitting technology shown, the calculation of the overall error based on the target variable simulation value and the current target variable batch data set usually includes two subdivision steps, namely, firstly calculating the point error of the target variable simulation value according to each data in the current target variable batch data set, and then using weighted summation, averaging or other statistical methods to calculate the point error to obtain the overall error. There are usually two ways to calculate the point error, namely the absolute error calculation method and the relative error calculation method. The absolute error calculation method can be expressed as the following formula (1):

[0029] error=|simu-meas| (1)

[0030] Among them, error represents the point error, meas represents a piece of data in the current target variable batch data set, and simu represents the target variable simulation value. The relative error calculation method can be expressed as the following formula (2):

[0031]

[0032] The inventors of the present disclosure have found that selecting different error calculation methods will lead to different optimization fitting effects. In the problem of MOSFET characteristic curve fitting, the values ​​of the target variables usually have a large order of magnitude difference. Figure 4 An exemplary schematic diagram showing the target variable values ​​in some embodiments of the present disclosure is shown. Figure 4 As shown, the value of the target variable y is distributed in 10 -3 to 0, the maximum and minimum values ​​of y differ by about 1000 times. When performing curve fitting, if the value range of the target variable data is similar to Figure 4 In the case shown in the figure, using absolute error as the point error calculation formula will cause each data point to have a different order of magnitude; for data points with larger absolute values, larger absolute errors will also be calculated after the absolute error is calculated, resulting in that in the calculation process of the overall error, data points with larger absolute values ​​have greater weights than data points with smaller absolute values, which in turn causes the optimizer to focus more on points with larger values; for data points with smaller absolute values, even if the fitting effect is not good, it is difficult to reflect in the final result due to the small absolute value of the error, which ultimately leads to poor fitting of some data points.

[0033] In the prior art, a single error calculation method is often used to calculate point errors during curve fitting, which results in that when the distribution of target variable values ​​is wide, the optimization effect of the optimizer on data points with larger absolute values ​​is not symmetrical with the optimization effect on data points with smaller absolute values. In view of this, the present disclosure discloses a method for acquiring features of electronic devices, so as to select an error calculation method according to the actual characteristics of a data set and improve the accuracy of curve fitting.

[0034] Figure 5 An exemplary flow chart of a method for acquiring features of an electronic device in some embodiments of the present disclosure is shown, wherein the features of the electronic device include a parameterized function of a target variable with respect to a feature variable; Figure 5 As shown, the method includes: step 501, calculating the target variable simulation value according to the characteristic variable data based on the first parameter; step 502, determining the error calculation method according to the value range of the target variable data associated with the characteristic variable data; step 503, calculating the error between the target variable simulation value and the target variable data according to the error calculation method; step 504, updating the first parameter according to the error to obtain the second parameter.

[0035] It is understood that the characteristics of the electronic device can be characteristic curves of the electronic device, including but not limited to the transfer characteristic curve, output characteristic curve, substrate current characteristic curve, gate current characteristic curve, transconductance characteristic curve, drain breakdown characteristic curve, and temperature characteristic curve of MOSFET; these curves can be expressed as parameterized relationship functions y=f of the target variable with respect to the characteristic variable θ (x). As for the type of the relational function, in some embodiments, a polynomial function is selected as the relational function. In these embodiments, f θ (x) = a0 + a1x + a2x 2 +…+a n x n , where θ=[a0,a1,a2,…,a n ], n can be 3, 5, 7 or other integers.

[0036] It can be understood that the characteristic variable data may be a characteristic variable batch data set (x) in a data set actually collected from electronic devices; for each characteristic variable data (x) in the characteristic variable batch data set i ) can be calculated based on f θ1 (x i ) calculate the corresponding Thus, based on the first parameter, the target variable simulation value is calculated according to the feature variable batch data set x The target variable data associated with the characteristic variable data may be a target variable batch data set (y) in a data set collected from electronic devices. The value range of y may be determined based on the maximum value, minimum value or discreteness of the data in y, thereby determining an applicable error calculation formula. Based on the selected optimizer, the first parameter may be updated based on the error to obtain the second parameter, thereby converting the relationship function from Updated to

[0037] The method for acquiring features of electronic devices disclosed in the present disclosure adaptively selects an error calculation method according to the value range of the target variable, thereby avoiding the problem that points with larger magnitudes have larger weights when calculating the overall error, thereby improving the accuracy of curve fitting.

[0038] In some embodiments, the characteristic variable includes voltage; and the target variable includes current.

[0039] It is understandable that the method for obtaining characteristics of an electronic device disclosed in the present disclosure can be used to obtain the corresponding transfer characteristic curve of a MOSFET, where the characteristic variable is the voltage between the gate and the source, V GS , the target variable is the drain current I DIn addition, the method for obtaining characteristics of an electronic device disclosed in the present disclosure can also be used to obtain the output characteristic curve corresponding to MOSFET, where the characteristic variable is the voltage between the drain and the source, V DS , the target variable is the drain current I D .

[0040] In some embodiments, the feature variable data includes measured data of the feature variable; and the target variable data includes measured data of the target variable.

[0041] It is understandable that the characteristic variable data and the target variable data may be data obtained by actually measuring the electronic device. In some embodiments, after obtaining the actual measurement data, the actual measurement data is cleaned, including removing invalid data and abnormal data in the measurement data.

[0042] In some embodiments, an error calculation method is determined based on a value range of target variable data associated with feature variable data, including: obtaining the order of magnitude corresponding to the maximum value of the target variable data to obtain a first order of magnitude; obtaining the order of magnitude corresponding to the minimum value of the target variable data to obtain a second order of magnitude; in response to a ratio of the first order of magnitude to the second order of magnitude exceeding a set ratio threshold, using a first method to calculate the error; in response to a ratio of the first order of magnitude to the second order of magnitude not exceeding a set ratio threshold, using a second method to calculate the error.

[0043] It is understood that in order to obtain the order of magnitude of the data, scientific notation can be used to represent the data. For example, for data T, T can be represented as m×10 n , where 1≤m<10, so 10 n The symbol represents the order of magnitude of T. In some embodiments, the maximum value T1 of the target variable data is represented as m1×10 n1 , the minimum value T2 of the target variable data is expressed as m2×10 n2 ; When 10 n1-n2 >s, the first method is used to calculate the error; when 10 n1-n2 ≤s, the second method is used to calculate the error; where s can be 10 3 , 10 4 , 10 5 or other integer.

[0044] In some embodiments, the first method is used to calculate relative errors; the second method is used to calculate absolute errors.

[0045] It can be understood that the calculation of relative error requires more computational effort than the calculation of absolute error; the present disclosure calculates the error between the target variable simulation value and the target variable data based on absolute error when the value range of the target variable data is small, and calculates the error between the target variable simulation value and the target variable data based on relative error when the value range of the target variable data is large, thereby achieving a better balance between the accuracy of the curve fitting results and the computational effort of the simulation process.

[0046] In some embodiments, the error between the target variable simulation value and the target variable data is calculated according to an error calculation method, including: calculating the error between the target variable simulation value and the target variable data point by point to obtain a point error; and calculating the overall error between the target variable simulation value and the target variable data based on the point error.

[0047] It can be understood that calculating the error between the target variable simulation value and the target variable point by point is to calculate the point error of the target variable simulation value according to each data in the target variable batch data set. k ], target variable simulation value Where k represents the number of data contained in the target variable batch data set, then the point error includes: In these embodiments, calculating the overall error (Err) between the target variable simulation value and the target variable data based on the point errors includes: summing the point errors to obtain the overall error, that is, Alternatively, the point errors are averaged to get the overall error, i.e.

[0048] In some embodiments, updating the first parameter according to the error to obtain the second parameter includes: calculating a parameter gradient according to the error; and updating the first parameter according to the parameter gradient to obtain the second parameter.

[0049] In these embodiments, the relationship function y=f θ (x) to determine the gradient calculation method of parameter θ Where Err represents the overall error between the target variable simulation value and the target variable data; then according to To update the first parameter θ1 to obtain the second parameter θ2. For example, f θ (x) = a0 + a1x + a2x 2 +…+a n x n , but in

[0050] In some embodiments, the method for acquiring characteristics of an electronic device disclosed in the present disclosure further includes: cyclically updating the parameters of the function according to the characteristic variable data and the target variable data until a preset termination condition is met.

[0051] It can be understood that the data set actually collected for the electronic device can be divided into batch data sets by batch, and each batch data set includes a characteristic variable batch data set and a target variable batch data set; thus, the optimal parameters in the parameterized function of the characteristic curve can be obtained through a loop iteration process. In each iteration: first, the curve function is used to calculate the target variable simulation value for the current characteristic variable batch data set; then the overall error is calculated based on the target variable simulation value and the current target variable batch data set; then, based on the overall error and the number of current iterations, it can be determined whether the loop termination condition is met. If the termination condition is met, the loop can be ended, otherwise, an optimizer (such as a gradient descent algorithm) is used to adjust the function parameters according to the overall error and enter the next loop. In some embodiments, the termination condition is that the number of cycles reaches a set cycle number threshold or the overall error is less than a set error threshold. The cycle number threshold can be 100 or other integers, and the error threshold can be 10 -3 , 10 -4 , 10 -5 or other values.

[0052] Figure 6 An exemplary flow chart of a method for acquiring features of an electronic device in some embodiments of the present disclosure is shown. Figure 6 As shown, in the iterative process of obtaining the optimal parameters in the parameterized function of the characteristic curve, each iteration can first use the curve function to calculate the target variable simulation value for the current characteristic variable batch data set, and before calculating the error based on the target variable simulation value and the current target variable batch data set, pre-analyze the data, that is, analyze the value range of the current target variable batch data set; if the ratio of the magnitude of the maximum value to the minimum value in the target variable batch data set exceeds 10 3 , it is considered that the value range of the current target variable batch data set is relatively large. At this time, the point error between the target variable simulation value and the current target variable batch data set is calculated based on the relative error calculation method, otherwise the point error between the target variable simulation value and the current target variable batch data set is calculated based on the absolute error calculation method; then the overall error is calculated based on the point error, and the optimizer is started to update the function parameters.

[0053] Furthermore, the present disclosure also discloses a computer-readable storage medium, in which program instructions are stored, and the program instructions are suitable for being loaded by a processor and executing the methods described in the above embodiments of the present disclosure.

[0054] The present disclosure also discloses a device for acquiring features of an electronic device, comprising: a processor configured to execute program instructions; and a memory configured to store program instructions, so that when the program instructions are loaded and executed by the processor, the device executes the methods described in the previous embodiments of the present disclosure.

[0055] Figure 7 FIG. 7 is a block diagram showing a hardware configuration of an apparatus 700 that can implement an embodiment of the present disclosure. Figure 7 As shown, the apparatus 700 may include a processor 701 and a memory 702. The processor is configured to execute program instructions, and the memory is configured to store program instructions. When the program instructions are loaded and executed by the processor, the apparatus executes the method for acquiring features of an electronic device according to any one of the above embodiments. Figure 7 In the device 700, only the components related to this embodiment are shown. Therefore, it is obvious to those skilled in the art that the device 700 may also include Figure 7 The specific functions implemented by the memory 702 and the processor 701 of the device 700 provided in the embodiment of this specification can be explained in comparison with the aforementioned implementations in this specification, and can achieve the technical effects of the aforementioned implementations, so they will not be repeated here.

[0056] The apparatus 700 may correspond to a computing device having various processing functions. For example, the apparatus 700 may be implemented as various types of devices, such as a personal computer (PC), a server device, a mobile device, and the like.

[0057] The processor 701 can control the operation of the device 700. For example, the processor 701 can be implemented by a central processing unit (CPU), a graphics processing unit (GPU), an application processor (AP), an artificial intelligence processor chip (IPU), etc. provided in the device 700. However, the present invention is not limited to this. In this embodiment, the processor 701 can be implemented in any appropriate manner. For example, the processor 701 can take the form of a computer-readable medium, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, etc., such as a microprocessor or a processor and a computer-readable program code (such as software or firmware) that can be executed by the (micro) processor.

[0058] The memory 702 can be used to store various data and instructions processed in the device 700. For example, the memory 702 can store processed data and data to be processed in the device 700. The memory 702 can store data that has been processed or to be processed by the processor 701, for example, feature variable data and target variable data. In addition, the memory 702 can store applications, drivers, etc. to be driven by the device 700. For example: the memory 702 can store various programs related to the method for obtaining features of electronic devices to be executed by the processor 701. The memory 702 can be a DRAM, but the present invention is not limited thereto. The memory 702 can include at least one of a volatile memory or a non-volatile memory. The non-volatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a phase change RAM (PRAM), a magnetic RAM (MRAM), a resistive RAM (RRAM), a ferroelectric RAM (FRAM), etc. The volatile memory may include dynamic RAM (DRAM), static RAM (SRAM), synchronous DRAM (SDRAM), PRAM, MRAM, RRAM, ferroelectric RAM (FeRAM), etc. In some embodiments, the memory 702 may include at least one of a hard disk drive (HDD), a solid state drive (SSD), a high-density flash memory (CF), a secure digital (SD) card, a micro secure digital (Micro-SD) card, a mini secure digital (Mini-SD) card, an extreme digital (xD) card, caches, or a memory stick.

[0059] In summary, the specific functions implemented by the computer-readable storage medium and the device for acquiring features of electronic devices provided in the embodiments of this specification can be interpreted in comparison with the aforementioned embodiments in this specification, and can achieve the technical effects of the aforementioned embodiments, so they will not be repeated here.

[0060] It should be noted that, for the purpose of simplicity, the present invention describes some methods and embodiments thereof as a series of actions and combinations thereof, but those skilled in the art will appreciate that the scheme of the present invention is not limited by the order of the described actions. Therefore, based on the disclosure or teaching of the present invention, those skilled in the art will appreciate that some of the steps therein may be performed in other orders or simultaneously. Further, those skilled in the art will appreciate that the embodiments described in the present invention may be regarded as optional embodiments, i.e., the actions or modules involved therein are not necessarily necessary for the implementation of one or some schemes of the present invention. In addition, depending on the different schemes, the present invention also has different emphases on the description of some embodiments. In view of this, those skilled in the art will appreciate that the parts not described in detail in a certain embodiment of the present invention may also refer to the relevant descriptions of other embodiments.

Claims

1. A method for acquiring characteristics of an electronic device, wherein: The feature includes a parameterized function of a target variable with respect to a feature variable; the method includes: Calculating a target variable simulation value from the characteristic variable data based on the first parameter; Determining an error calculation method according to a value range of the target variable data associated with the characteristic variable data; Calculate the error between the target variable simulation value and the target variable data according to the error calculation method; The first parameter is updated according to the error to obtain a second parameter.

2. The method according to claim 1, wherein: The characteristic variable includes voltage; the target variable includes current.

3. The method according to claim 1, wherein: The characteristic variable data includes the measured data of the characteristic variable; the target variable data includes the measured data of the target variable.

4. The method according to claim 1, wherein: Determining an error calculation method according to a value range of the target variable data associated with the characteristic variable data includes: Obtaining the order of magnitude corresponding to the maximum value of the target variable data to obtain a first order of magnitude; Obtaining the order of magnitude corresponding to the minimum value of the target variable data to obtain a second order of magnitude; In response to a ratio of the first order of magnitude to the second order of magnitude exceeding a set ratio threshold, selecting a first method to calculate an error; In response to the ratio of the first order of magnitude to the second order of magnitude not exceeding a set ratio threshold, a second method is selected to calculate the error.

5. The method according to claim 4, wherein: The first method is used to calculate relative errors; the second method is used to calculate absolute errors.

6. The method according to claim 1, wherein: Calculating the error between the target variable simulation value and the target variable data according to the error calculation method includes: Calculating the error between the target variable simulation value and the target variable data point by point to obtain a point error; Based on the point error, the overall error between the target variable simulation value and the target variable data is calculated.

7. The method according to claim 1, wherein: Updating the first parameter according to the error to obtain a second parameter comprises: Calculating a parameter gradient based on the error; The first parameter is updated according to the parameter gradient to obtain a second parameter.

8. The method according to claim 1, further comprising: Until a preset termination condition is met, the parameters of the function are cyclically updated according to the characteristic variable data and the target variable data.

9. A computer-readable storage medium storing program instructions, wherein the program instructions are suitable for being loaded by a processor and executing the method according to any one of claims 1 to 8.

10. An apparatus for acquiring characteristics of an electronic device, comprising: a processor configured to execute program instructions; as well as A memory configured to store the program instructions, which, when loaded and executed by the processor, causes the apparatus to perform the method according to any one of claims 1 to 8.

Images