A semiconductor performance testing method based on data acquisition and analysis

Abstract

The present invention discloses a semiconductor performance test method based on data acquisition and analysis, belonging to the technical field of semiconductor performance test. The present invention inputs high-frequency switching signals of different frequencies to the base of a transistor at each temperature, and collects the output signal of the collector. Then, the amplitude of the corresponding output signal is adjusted according to the amplitude of the high-frequency switching signal to obtain an adjusted signal, so that the adjusted signal is consistent with the high-frequency switching signal in amplitude scale, which is convenient for comparison in the frequency domain, and the response coefficient at each temperature is calculated to improve the accuracy of the calculated response coefficient. The output signals at different temperatures are compared in the frequency domain, the stability coefficient is calculated, and the stability of the transistor in different temperature environments is evaluated. According to the response coefficient at each temperature, based on the compensation of the stability coefficient, the transistor performance score is calculated, and the performance test accuracy is further improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor testing technology, and in particular to a semiconductor performance testing method based on data acquisition and analysis. Background Art

[0002] In the semiconductor field, transistors are a basic electronic component that is widely used in various electronic devices and systems. Transistors are commonly used semiconductor devices in electronic circuits, and their performance directly affects the operating stability and reliability of the entire circuit. Therefore, it is very necessary to test and evaluate the performance of transistors. Traditional transistor performance test methods mainly include static parameter testing and dynamic parameter testing. Static parameter testing mainly evaluates the performance of transistors by measuring the DC characteristics of transistors, such as collector current, base current, emitter current, etc. This method is simple and easy, but it cannot fully reflect the performance of transistors in high-frequency states. Dynamic parameter testing evaluates the performance of transistors by inputting high-frequency signals and observing the response characteristics of transistors. However, traditional dynamic parameter testing methods usually only focus on the response characteristics at a single frequency, and cannot fully reflect the performance differences of transistors at different frequencies, and there is a problem of low performance test accuracy. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a semiconductor performance testing method based on data acquisition and analysis to solve the problem of low performance testing accuracy in the prior art.

[0004] In order to achieve the above-mentioned purpose of the invention, the technical solution adopted by the present invention is: a semiconductor performance testing method based on data acquisition and analysis, comprising the following steps:

[0005] S1. At each temperature, a high-frequency switching signal of different frequencies is input to the base of the transistor, and the output signal of the collector of the transistor is collected to obtain a plurality of output signals, wherein each output signal corresponds to a high-frequency switching signal of a frequency;

[0006] S2. According to the amplitude of the high-frequency switching signal, the amplitude of the corresponding output signal is adjusted to obtain an adjustment signal;

[0007] S3, calculating a response coefficient at a temperature according to the difference between the multiple adjustment signals and the high-frequency switching signal in the frequency domain;

[0008] S4. Calculate the stability coefficient according to the difference in frequency domain between the output signals at different temperatures;

[0009] S5. Calculate the transistor performance score based on the response coefficient at each temperature and the compensation of the stability coefficient.

[0010] Furthermore, the S2 comprises the following sub-steps:

[0011] S21, calculating the average amplitude of the input high-frequency switching signal;

[0012] S22, calculating the average amplitude of the output signal corresponding to the input high-frequency switching signal;

[0013] S23, calculating an adjustment ratio according to the input high-frequency switching signal and the average amplitude of the output signal;

[0014] S24. According to the adjustment ratio, each amplitude in the output signal is adjusted to obtain an adjusted signal.

[0015] Furthermore, the formula adjusted in S24 is:

[0016] , , where A i is to adjust the i-th amplitude of the signal, ε is the adjustment ratio, A out,i is the i-th amplitude of the output signal, A in is the average amplitude of the input high-frequency switching signal, A out is the average amplitude of the output signal.

[0017] Furthermore, S3 includes the following sub-steps:

[0018] S31, at the same temperature, performing time-frequency conversion on high-frequency switching signals and multiple adjustment signals of different frequencies to obtain multiple high-frequency switching frequency domain signals and multiple adjustment frequency domain signals at one temperature;

[0019] S32, extracting the frequency component of each high-frequency switching frequency domain signal as an element to form a set A;

[0020] S33, extracting the frequency component of each adjusted frequency domain signal as an element to form a B set;

[0021] S34, calculating a first frequency difference coefficient between each high-frequency switching frequency domain signal and a corresponding adjustment frequency domain signal according to the A set and the B set;

[0022] S35, calculating a second frequency difference coefficient between each high-frequency switching frequency domain signal and the corresponding adjustment frequency domain signal according to the spectrum concentration value of the high-frequency switching frequency domain signal and the adjustment frequency domain signal;

[0023] S36. Add all first frequency difference coefficients and second frequency difference coefficients at a temperature and take an average to obtain a response coefficient at a temperature.

[0024] Furthermore, the formula for calculating the first frequency gap coefficient in S34 is:

[0025] , where γ1 is the first frequency difference coefficient, A is the A set, B is the B set, ∪ is the union, and ∩ is the intersection.

[0026] Furthermore, the formula for calculating the second frequency gap coefficient in S35 is:

[0027] , where γ2 is the second frequency difference coefficient, e is a natural constant, r2 is the spectrum center value of the adjusted frequency domain signal, and r1 is the spectrum center value of the high-frequency switching frequency domain signal.

[0028] Furthermore, the calculation formula of the spectrum concentration value is:

[0029] , where r is the spectrum concentration value, A max is the maximum amplitude of the frequency domain signal, f max is the frequency component corresponding to the maximum amplitude, A j is the amplitude corresponding to the jth frequency component in the frequency domain signal, f j is the jth frequency component, N is the number of frequency components in the frequency domain signal, and j is a positive integer.

[0030] Furthermore, the S4 comprises the following sub-steps:

[0031] S41, performing time-frequency conversion on the output signal to obtain an output frequency domain signal;

[0032] S42, extracting the spectrum concentration value of the output frequency domain signal;

[0033] S43, collecting multiple spectrum values ​​corresponding to the high-frequency switching signal of the same frequency at different temperatures to form a spectrum set at one frequency;

[0034] S44, calculating a fluctuation value in a frequency spectrum set at a frequency;

[0035] S45. Calculate the stability coefficient according to the fluctuation value in the spectrum set at each frequency.

[0036] Furthermore, the formula for calculating the fluctuation value in S44 is:

[0037] , where s m is the fluctuation value in the spectrum set at the mth frequency, x m,n is the nth spectrum concentration value in the spectrum set at the mth frequency, M is the number of spectrum concentration values ​​in the spectrum set, n and m are positive integers, and | | is an absolute value operation;

[0038] The formula for calculating the stability coefficient in S45 is:

[0039] , where θ is the stability factor and L is the number of high-frequency switching signals input at the same temperature.

[0040] Furthermore, the formula for calculating the transistor performance score in S5 is:

[0041] , where y is the transistor performance score, α k is the kth weighting coefficient, c k is the response coefficient at the kth temperature, R is the number of response coefficients, k is a positive integer, and θ is the stability coefficient.

[0042] In summary, the beneficial effects of the present invention are as follows: at each temperature, the present invention inputs a high-frequency switching signal of different frequencies to the base of the transistor, and collects the output signal of the collector. Then, the amplitude of the corresponding output signal is adjusted according to the amplitude of the high-frequency switching signal to obtain an adjustment signal, so that the adjusted signal is consistent with the high-frequency switching signal in amplitude scale, which is convenient for comparison in the frequency domain, and the response coefficient at each temperature is calculated to improve the accuracy of the calculated response coefficient. The output signals at different temperatures are compared in the frequency domain, the stability coefficient is calculated, and the stability of the transistor in different temperature environments is evaluated. According to the response coefficient at each temperature, based on the compensation of the stability coefficient, the performance score of the transistor is calculated, and the performance test accuracy is further improved. The present invention obtains the response coefficient and the stability coefficient by collecting and analyzing the output signals of the transistor at different temperatures and different frequencies, thereby more comprehensively evaluating the performance of the transistor and improving the performance test accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The present invention is a flow chart of a semiconductor performance testing method based on data acquisition and analysis. DETAILED DESCRIPTION

[0044] The specific implementation modes of the present invention are described below to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0045] like Figure 1 As shown, a semiconductor performance testing method based on data acquisition and analysis includes the following steps:

[0046] S1. At each temperature, a high-frequency switching signal of different frequencies is input to the base of the transistor, and the output signal of the collector of the transistor is collected to obtain a plurality of output signals, wherein each output signal corresponds to a high-frequency switching signal of a frequency;

[0047] S2. According to the amplitude of the high-frequency switching signal, the amplitude of the corresponding output signal is adjusted to obtain an adjustment signal;

[0048] S3, calculating a response coefficient at a temperature according to the difference between the multiple adjustment signals and the high-frequency switching signal in the frequency domain;

[0049] S4. Calculate the stability coefficient according to the difference in frequency domain between the output signals at different temperatures;

[0050] S5. Calculate the transistor performance score based on the response coefficient at each temperature and the compensation of the stability coefficient.

[0051] In this embodiment, the high-frequency switching signals of different frequencies input to the base of the transistor include: a 3 MHz high-frequency switching signal, a 1000 MHz high-frequency switching signal, a 5000 MHz high-frequency switching signal, and a 1 GHz high-frequency switching signal.

[0052] In this embodiment, the different temperatures include: -55°C, 0°C, 30°C, and 100°C.

[0053] The S2 comprises the following sub-steps:

[0054] S21, calculating the average amplitude of the input high-frequency switching signal;

[0055] S22, calculating the average amplitude of the output signal corresponding to the input high-frequency switching signal;

[0056] S23, calculating an adjustment ratio according to the input high-frequency switching signal and the average amplitude of the output signal;

[0057] S24. According to the adjustment ratio, each amplitude in the output signal is adjusted to obtain an adjusted signal.

[0058] The formula adjusted in S24 is:

[0059] , , where A i is to adjust the i-th amplitude of the signal, ε is the adjustment ratio, A out,i is the i-th amplitude of the output signal, A in is the average amplitude of the input high-frequency switching signal, A out is the average amplitude of the output signal.

[0060] The present invention takes the average amplitude of the input high-frequency switching signal and the average amplitude of the corresponding output signal, calculates the adjustment ratio, and uses the adjustment ratio to adjust the output signal amplitude to the same scale as the high-frequency switching signal, so as to facilitate frequency domain comparison, thereby detecting whether there is a signal deformation problem when the transistor amplifies the signal.

[0061] The S3 comprises the following sub-steps:

[0062] S31, at the same temperature, performing time-frequency conversion on high-frequency switching signals and multiple adjustment signals of different frequencies to obtain multiple high-frequency switching frequency domain signals and multiple adjustment frequency domain signals at one temperature;

[0063] S32, extracting the frequency component of each high-frequency switching frequency domain signal as an element to form a set A;

[0064] S33, extracting the frequency component of each adjusted frequency domain signal as an element to form a B set;

[0065] S34, calculating a first frequency difference coefficient between each high-frequency switching frequency domain signal and a corresponding adjustment frequency domain signal according to the A set and the B set;

[0066] S35, calculating a second frequency difference coefficient between each high-frequency switching frequency domain signal and the corresponding adjustment frequency domain signal according to the spectrum concentration value of the high-frequency switching frequency domain signal and the adjustment frequency domain signal;

[0067] S36. Add all first frequency difference coefficients and second frequency difference coefficients at a temperature and take an average to obtain a response coefficient at a temperature.

[0068] At the same temperature, high-frequency switching signals of multiple frequencies are input to the triode respectively to obtain multiple output signals. After the amplitude of the output signal is adjusted, multiple adjustment signals are obtained. Therefore, a high-frequency switching signal of one frequency corresponds to one adjustment signal.

[0069] The present invention first calculates the first frequency difference coefficient based on the frequency components of the high-frequency switching frequency domain signal and the adjustment frequency domain signal to reflect the response speed of the transistor in frequency, and then calculates the second frequency difference coefficient based on the spectrum concentration value of the high-frequency switching frequency domain signal and the adjustment frequency domain signal to reflect the frequency response of the transistor and the transmission of the main frequency and signal shape. All the first frequency difference coefficients and second frequency difference coefficients obtained at one temperature are added together and the average is taken to obtain the response coefficient at one temperature.

[0070] The formula for calculating the first frequency gap coefficient in S34 is:

[0071] , where γ1 is the first frequency difference coefficient, A is the A set, B is the B set, ∪ is the union, and ∩ is the intersection.

[0072] In the present invention, the high-frequency switching signal has only one frequency component. Therefore, when the transistor response speed is high, the B set should also have only one frequency component. When there are multiple frequency components in the B set, the transistor response speed is low.

[0073] The formula for calculating the second frequency gap coefficient in S35 is:

[0074] , where γ2 is the second frequency difference coefficient, e is a natural constant, r2 is the spectrum center value of the adjusted frequency domain signal, and r1 is the spectrum center value of the high-frequency switching frequency domain signal.

[0075] The calculation formula of the spectrum concentration value is:

[0076] , where r is the spectrum concentration value, A max is the maximum amplitude of the frequency domain signal, f max is the frequency component corresponding to the maximum amplitude, A j is the amplitude corresponding to the jth frequency component in the frequency domain signal, f j is the jth frequency component, N is the number of frequency components in the frequency domain signal, and j is a positive integer.

[0077] The present invention selects the maximum amplitude and the corresponding frequency component of the frequency domain signal to reflect the main signal components, and then extracts each amplitude and each frequency component to reflect the overall signal shape. According to the difference between the spectrum concentration value of the high-frequency switching frequency domain signal and the adjustment frequency domain signal, the second frequency difference coefficient is calculated to reflect the signal transmission of the transistor. When the second frequency difference coefficient is larger, the difference before and after the signal transmission is smaller, and the deformation is smaller. When the second frequency difference coefficient is smaller, the difference before and after the signal transmission is larger, and the deformation is larger.

[0078] The present invention adjusts the amplitude of the output signal, so the spectrum concentration value can be used to reflect the difference in form between the high-frequency switching frequency domain signal and the adjustment frequency domain signal.

[0079] The S4 comprises the following sub-steps:

[0080] S41, performing time-frequency conversion on the output signal to obtain an output frequency domain signal;

[0081] S42, extracting the spectrum concentration value of the output frequency domain signal;

[0082] S43, collecting multiple spectrum values ​​corresponding to the high-frequency switching signal of the same frequency at different temperatures to form a spectrum set at one frequency;

[0083] In S43, since the transistor is placed at different temperatures, the frequency of the high-frequency switching signal is kept unchanged, and the output signal of the transistor is collected at each temperature, there are multiple output frequency domain signals;

[0084] S44, calculating a fluctuation value in a frequency spectrum set at a frequency;

[0085] S45. Calculate the stability coefficient according to the fluctuation value in the spectrum set at each frequency.

[0086] In the present invention, the calculation formulas for all spectrum concentration values ​​are the same.

[0087] The present invention performs time-frequency conversion on the output signal to obtain an output frequency domain signal, extracts the spectrum concentration value of the output frequency domain signal, a spectrum concentration value reflects the frequency and amplitude characteristics of the output signal at a temperature and a frequency, and multiple spectrum concentration values ​​corresponding to high-frequency switching signals of the same frequency at different temperatures form a spectrum set, calculates the fluctuation value of the spectrum set, and reflects the output stability of the transistor at different temperatures through the fluctuation value.

[0088] The formula for calculating the fluctuation value in S44 is:

[0089] , where s m is the fluctuation value in the spectrum set at the mth frequency, x m,n is the nth spectrum concentration value in the spectrum set at the mth frequency, M is the number of spectrum concentration values ​​in the spectrum set, n and m are positive integers, and | | is an absolute value operation;

[0090] The formula for calculating the stability coefficient in S45 is:

[0091] , where θ is the stability factor and L is the number of high-frequency switching signals input at the same temperature.

[0092] In the present invention, the high-frequency switching signal is a single-frequency signal, and therefore, each high-frequency switching signal has one frequency component.

[0093] The present invention calculates the absolute value of the difference between each spectrum concentration value and the mean in the spectrum set, reflects the gap between each spectrum concentration value and the mean, and calculates the output fluctuation of the transistor under different temperatures. When the fluctuation value is larger, the transistor is more affected by temperature and is more unstable.

[0094] The formula for calculating the transistor performance score in S5 is:

[0095] , where y is the transistor performance score, α k is the kth weighting coefficient, c k is the response coefficient at the kth temperature, R is the number of response coefficients, k is a positive integer, and θ is the stability coefficient.

[0096] In this embodiment, the weighting coefficient is set according to demand and experience.

[0097] The present invention weights the response coefficients at various temperatures and then compensates them with a stability coefficient, thereby improving the accuracy of transistor test performance.

[0098] The present invention inputs high-frequency switching signals of different frequencies to the base of the triode at each temperature, and collects the output signal of the collector. Then, the amplitude of the corresponding output signal is adjusted according to the amplitude of the high-frequency switching signal to obtain an adjusted signal, so that the adjusted signal is consistent with the high-frequency switching signal in amplitude scale, which is convenient for comparison in the frequency domain, and the response coefficient at each temperature is calculated, and the accuracy of calculating the response coefficient is improved. The output signals at different temperatures are compared in the frequency domain, the stability coefficient is calculated, and the stability of the triode in different temperature environments is evaluated. According to the response coefficient at each temperature, based on the compensation of the stability coefficient, the performance score of the triode is calculated, and the performance test accuracy is further improved. The present invention obtains the response coefficient and the stability coefficient by collecting and analyzing the output signals of the triode at different temperatures and different frequencies, thereby more comprehensively evaluating the performance of the triode and improving the performance test accuracy.

[0099] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A semiconductor performance testing method based on data acquisition and analysis, characterized in that: The following steps are involved: S1. At each temperature, a high-frequency switching signal of different frequencies is input to the base of the transistor, and the output signal of the collector of the transistor is collected to obtain a plurality of output signals, wherein each output signal corresponds to a high-frequency switching signal of a frequency; S2. According to the amplitude of the high-frequency switching signal, the amplitude of the corresponding output signal is adjusted to obtain an adjustment signal; S3, calculating a response coefficient at a temperature according to the difference between the multiple adjustment signals and the high-frequency switching signal in the frequency domain; S4. Calculate the stability coefficient according to the difference in frequency domain between the output signals at different temperatures; S5. Calculate the transistor performance score based on the response coefficient at each temperature and the compensation of the stability coefficient: , where y is the transistor performance score, α k is the kth weighting coefficient, c k is the response coefficient at the kth temperature, R is the number of response coefficients, k is a positive integer, and θ is the stability coefficient; The S3 comprises the following sub-steps: S31, at the same temperature, performing time-frequency conversion on high-frequency switching signals and multiple adjustment signals of different frequencies to obtain multiple high-frequency switching frequency domain signals and multiple adjustment frequency domain signals at one temperature; S32, extracting the frequency component of each high-frequency switching frequency domain signal as an element to form a set A; S33, extracting the frequency component of each adjusted frequency domain signal as an element to form a B set; S34, calculating a first frequency difference coefficient between each high-frequency switching frequency domain signal and a corresponding adjustment frequency domain signal according to the A set and the B set; S35, calculating a second frequency difference coefficient between each high-frequency switching frequency domain signal and the corresponding adjustment frequency domain signal according to the spectrum concentration value of the high-frequency switching frequency domain signal and the adjustment frequency domain signal; S36, adding all first frequency gap coefficients and second frequency gap coefficients at a temperature and taking an average value to obtain a response coefficient at a temperature; The S4 comprises the following sub-steps: S41, performing time-frequency conversion on the output signal to obtain an output frequency domain signal; S42, extracting the spectrum concentration value of the output frequency domain signal; S43, collecting multiple spectrum values ​​corresponding to the high-frequency switching signal of the same frequency at different temperatures to form a spectrum set at one frequency; S44, calculating a fluctuation value in a frequency spectrum set at a frequency; S45. Calculate the stability coefficient according to the fluctuation value in the spectrum set at each frequency.

2. The semiconductor performance testing method based on data acquisition and analysis according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21, calculating the average amplitude of the input high-frequency switching signal; S22, calculating the average amplitude of the output signal corresponding to the input high-frequency switching signal; S23, calculating an adjustment ratio according to the input high-frequency switching signal and the average amplitude of the output signal; S24. According to the adjustment ratio, each amplitude in the output signal is adjusted to obtain an adjusted signal.

3. The semiconductor performance testing method based on data acquisition and analysis according to claim 2, characterized in that: The formula adjusted in S24 is: , , where A i is to adjust the i-th amplitude of the signal, ε is the adjustment ratio, A out,i is the i-th amplitude of the output signal, A in is the average amplitude of the input high-frequency switching signal, A out is the average amplitude of the output signal.

4. The semiconductor performance testing method based on data acquisition and analysis according to claim 1, characterized in that: The formula for calculating the first frequency gap coefficient in S34 is: , where γ1 is the first frequency difference coefficient, A is the A set, B is the B set, ∪ is the union, and ∩ is the intersection.

5. The semiconductor performance testing method based on data acquisition and analysis according to claim 1, characterized in that: The formula for calculating the second frequency gap coefficient in S35 is: , where γ2 is the second frequency difference coefficient, e is a natural constant, r2 is the spectrum center value of the adjusted frequency domain signal, and r1 is the spectrum center value of the high-frequency switching frequency domain signal.

6. The semiconductor performance testing method based on data acquisition and analysis according to claim 5, characterized in that: The calculation formula of the spectrum concentration value is: , where r is the spectrum concentration value, A max is the maximum amplitude of the frequency domain signal, f max is the frequency component corresponding to the maximum amplitude, A j is the amplitude corresponding to the jth frequency component in the frequency domain signal, f j is the jth frequency component, N is the number of frequency components in the frequency domain signal, and j is a positive integer.

7. The semiconductor performance testing method based on data acquisition and analysis according to claim 1, characterized in that: The formula for calculating the fluctuation value in S44 is: , where s m is the fluctuation value in the spectrum set at the mth frequency, x m,n is the nth spectrum concentration value in the spectrum set at the mth frequency, M is the number of spectrum concentration values ​​in the spectrum set, n and m are positive integers, and | | is an absolute value operation; The formula for calculating the stability coefficient in S45 is: , where θ is the stability factor and L is the number of high-frequency switching signals input at the same temperature.

Images