CN117590838B - Intelligent detection method and system for aging of microprocessor - Google Patents
Intelligent detection method and system for aging of microprocessor Download PDFInfo
- Publication number
- CN117590838B CN117590838B CN202410081205.0A CN202410081205A CN117590838B CN 117590838 B CN117590838 B CN 117590838B CN 202410081205 A CN202410081205 A CN 202410081205A CN 117590838 B CN117590838 B CN 117590838B
- Authority
- CN
- China
- Prior art keywords
- aging
- microprocessor
- signal
- voltage
- time period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 230000002431 foraging effect Effects 0.000 title description 2
- 230000032683 aging Effects 0.000 claims abstract description 209
- 238000000034 method Methods 0.000 claims abstract description 57
- 239000011159 matrix material Substances 0.000 claims abstract description 35
- 230000007704 transition Effects 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims description 41
- 230000008569 process Effects 0.000 claims description 15
- 230000007306 turnover Effects 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 230000000630 rising effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000015556 catabolic process Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000006731 degradation reaction Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 230000035945 sensitivity Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 4
- 230000009897 systematic effect Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000000739 chaotic effect Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000003247 decreasing effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000004870 electrical engineering Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0221—Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Tests Of Electronic Circuits (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an intelligent detection method and system for microprocessor aging. The method comprises the following steps: acquiring a pin signal, a frequency division clock signal and an ideal waveform; acquiring voltage deviation according to the voltage difference of the pin signals; acquiring a first division signal; acquiring a delay phenomenon observation time period according to an ideal waveform, and acquiring a delay phenomenon significance according to a voltage difference and a time period length of the delay phenomenon observation time period; acquiring a signal jump interval time period, and acquiring the saliency of the burr phenomenon according to the voltage difference and the time length of the signal jump interval time period; acquiring the aging degree significance of the microprocessor based on the aging degree significance; obtaining a Markov transition probability matrix according to the aging degree significance of the microprocessor, and constructing an aging index confusion degree; and acquiring an aging degree index according to the aging index confusion degree to finish intelligent aging detection of the microprocessor. The invention maintains the sensitivity of detection, and makes the detection of microprocessor aging more sensitive and accurate.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an intelligent detection method and system for microprocessor aging.
Background
Microprocessors are the core elements of modern information technology, which have powerful computing functions and are indispensable parts in electronic components. In the use of a microprocessor, the microprocessor inevitably has aging phenomena due to the occurrence of phenomena such as electron migration, breakdown with time and the like. In order to avoid the damage to equipment caused by faults due to the aging of the microprocessor, the aging degree of the microprocessor needs to be detected. The method of microprocessor detection is realized mainly by monitoring the output voltage of the processor.
Microprocessors typically generate signals from internal crystal oscillator structures, with multiple pins outputting digital signals. The general equipment aging detection method directly detects whether equipment is aged or not through the information indicating the aging degree of an output signal, but for a microprocessor, the components of the output digital signal are complex, the signal components indicating the aging fault of the microprocessor are mutually influenced and are difficult to identify, and the detection precision of the fault is influenced, so that the detection precision is lower.
Disclosure of Invention
In order to solve the technical problem of lower fault detection precision, the invention provides an intelligent detection method and system for microprocessor aging, and the adopted technical scheme is as follows:
In a first aspect, the present invention provides a method for intelligently detecting aging of a microprocessor, the method comprising the steps of:
acquiring a pin signal and a frequency division clock signal of a microprocessor;
Acquiring a voltage deviation value according to the difference between the voltage of each time of the pin signal and the high level as well as the low level, and acquiring the voltage deviation according to the sum of the voltage deviation values of all the time within one period; according to the pin signals and the frequency division clock signals, an AD conversion algorithm is adopted to obtain ideal waveforms, and the pin signals are decomposed by using EMD to obtain first division signals;
Acquiring turning time according to an ideal waveform, determining a delay phenomenon observation time period of a first partial signal through the turning time, and acquiring a delay phenomenon significance according to the difference of voltage and high level and low level at each time of the delay phenomenon observation time period and the time length of the delay phenomenon observation time period; acquiring a signal jump interval time period according to adjacent rising edges and falling edges in the first sub-signal, and acquiring the saliency of the burr phenomenon according to the voltage and the high level at each moment of the signal jump interval time period, the low level and the time length of the signal jump interval time period; acquiring the aging degree saliency of the microprocessor of each turnover according to the burr phenomenon saliency, the delay phenomenon saliency and the voltage deviation of each turnover;
counting the aging degree significance of the microprocessor turned over for a plurality of times to form an aging degree statistical vector, and acquiring an aging degree statistical differential vector based on the aging degree statistical vector; obtaining a Markov transition probability matrix according to the ageing degree statistical difference vector, and constructing ageing index confusion according to the Markov transition probability matrix;
And determining the length of an aging window according to the disorder degree of the aging index to obtain an aging degree index, and finishing intelligent aging detection of the microprocessor according to the aging degree index.
Preferably, the method for obtaining the voltage deviation value according to the voltage and the difference of the high level and the low level at each moment of the pin signal comprises the following steps:
Calculating the average value of the high level and the low level and recording the average value as a level average value;
if the voltage at each moment of the pin signal is greater than or equal to the level average value, making the absolute value of the difference value between the voltage at each moment of the pin signal and the high level as a voltage deviation value;
If the voltage at each time of the pin signal is smaller than the level average value, the absolute value of the difference between the voltage at each time of the pin signal and the low level is taken as the voltage deviation value.
Preferably, the method for obtaining the inversion time according to the ideal waveform and determining the delay phenomenon observation time period of the first partial signal through the inversion time includes:
The turning moment is the moment when the signal is from 0 to 1 or from 1 to 0;
For an ideal waveform, a preset time period is taken as a time delay phenomenon observation time period after the turning moment, the preset time period is the product of a period and a decimal, and the period is the time period from a low level to the next low level in the frequency division clock signal.
Preferably, the method for obtaining the significance of the delay phenomenon according to the difference between the voltage and the high level and the low level at each moment in the delay phenomenon observation time period and the time length of the delay phenomenon observation time period comprises the following steps:
If the voltage at each moment in the time period of observing the time delay phenomenon is greater than or equal to the average value of the level, the absolute value of the difference value between the voltage at each moment in the time period of observing the time delay phenomenon and the low level is used as the voltage difference of the time delay phenomenon;
If the voltage at each moment in the time period of observing the time delay phenomenon is smaller than the level average value, making the absolute value of the difference value between the voltage at each moment in the time period of observing the time delay phenomenon and the high level be used as the voltage difference of the time delay phenomenon;
the ratio of the delay phenomenon voltage difference at each moment of the delay phenomenon observation time period to the time length of the delay phenomenon observation time period is recorded as a first ratio, and the accumulated sum of the first ratios at all moments is recorded as the delay phenomenon significance.
Preferably, the method for obtaining the saliency of the glitch according to the voltage and the high level and the low level at each moment in the signal jump interval time period and the time length of the signal jump interval time period includes:
in the method, in the process of the invention, Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>Is the average value of the high level and the low level of the pin signal,/>Voltage value representing mth moment in signal jump interval time period of first partial signal,/>Representing the time length of the signal transition interval period,/>Representing the length of time of one cycle,/>Indicating the significance of the glitch phenomenon.
Preferably, the method for obtaining the microprocessor aging degree saliency of each overturn according to the burr phenomenon saliency, the delay phenomenon saliency and the voltage deviation of each overturn comprises the following steps:
in the method, in the process of the invention, Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>For pin signal (H /)The aging degree of the microprocessor which is turned over again is remarkable; /(I)For pin signal (H /)The time delay phenomenon of the secondary overturn is remarkable; /(I)For pin signal (H /)The saliency of the burr phenomenon of the secondary overturn; /(I)For pin signal (H /)Voltage deviation mean value of secondary flip,/>Is the saliency versus weight.
Preferably, the method for counting the aging degree significance of the microprocessor turned over for several times to form an aging degree statistical vector and obtaining the aging degree statistical differential vector based on the aging degree statistical vector comprises the following steps:
counting the aging degree significance of the microprocessor turned over for S times, and sequencing the aging degree significance of the microprocessor according to a time sequence to form an aging degree statistical vector;
Taking the difference value between the latter element and the former element in the aging degree statistical difference vector as an aging difference value, and obtaining the aging degree statistical difference vector according to time sequence by all the aging difference values, wherein the value of the first element of the aging degree statistical difference vector is 0.
Preferably, the method for obtaining the markov transition probability matrix according to the ageing degree statistical difference vector and constructing the ageing index confusion degree according to the markov transition probability matrix comprises the following steps:
dividing all elements in the aging degree statistical difference vector into three types to form a discrete matrix, and presetting a threshold value Will be less than or equal to/>Elements of (2) are classified into one class; will be greater than/>Less than/>Elements of (2) are classified into one class; will be greater than or equal to/>Wherein each class forms a vector, and three vectors form a discrete matrix;
the discrete matrix and the aging degree statistical difference vector are used as input, the algorithm of the Markov transition probability matrix is used for calculation, the output is the Markov transition probability matrix, and the expression of the aging index chaos degree is as follows:
in the method, in the process of the invention, Is the degree of confusion of the ageing index; /(I)Is the first/>, of the Markov transition probability matrixLine/>Column elements; first/>, markov transition probability matrix Line/>Column elements.
Preferably, the method for determining the length of the aging window according to the confusion of the aging index to obtain the aging degree index and finishing the intelligent aging detection of the microprocessor according to the aging degree index comprises the following steps:
the product of the confusion degree of the aging index and S is recorded as the total turnover number, and the aging observation window comprises the turnover number of the total turnover number;
in the length of the aging window, calculating the average value of the aging degree significance of all the overturned microprocessors as the aging degree index of the aging observation window;
Selecting A failed microprocessors, calculating the average value of the aging degree indexes of all aging observation windows of each microprocessor in the previous day, marking the average value as an aging average value, acquiring the average value and the variance of the aging average values of all the microprocessors, marking the average value and the variance as a first aging average value and a first aging variance, and marking the difference value between the first aging average value and three times of the first aging variance as an aging degree index threshold value;
And when the aging degree index of the micro-processing to be detected is larger than the aging degree index threshold, the microprocessor to be detected has faults.
In a second aspect, an embodiment of the present invention further provides a system for intelligently detecting aging of a microprocessor, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of any one of the foregoing methods for intelligently detecting aging of a microprocessor when executing the computer program.
The invention has the following beneficial effects: according to the embodiment of the invention, the signal characteristics representing the aging phenomenon in the output signal of the microprocessor are subjected to characteristic calculation, the voltage deviation value is obtained according to the average value of the difference between the actual voltage and the rated logic voltage aiming at the voltage deviation phenomenon, the delay phenomenon significance is obtained according to the voltage difference in the delay phenomenon observation window, the burr phenomenon significance is obtained according to the voltage difference in the sampling interval aiming at the burr phenomenon, the aging degree significance of the microprocessor is constructed by combining the masking relation of the voltage deviation phenomenon on the delay and the burr phenomenon, the aging degree of the microprocessor is represented, and the extraction of the aging characteristics of the microprocessor is completed; the Markov transition probability matrix is further constructed for the change condition of the aging degree significance of the microprocessor to represent the chaotic degree of the aging degree significance of the microprocessor, and the window size of the aging degree index is adaptively adjusted according to the chaotic degree, so that the systematic error in the aging detection process of the microprocessor is solved, the detection sensitivity is maintained, and the aging detection of the microprocessor is more sensitive and accurate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for intelligently detecting microprocessor aging according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a pin signal and a divided clock signal;
FIG. 3 is a schematic diagram of an ideal waveform;
FIG. 4 is a schematic diagram of a delay phenomenon;
Fig. 5 is a schematic diagram of the burring phenomenon.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to specific embodiments, structures, features and effects of a microprocessor aging intelligent detection method and system according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
An embodiment of a microprocessor aging intelligent detection method and system:
the invention provides a microprocessor aging intelligent detection method and a system specific scheme by combining the drawings.
Referring to fig. 1, a flowchart of a method for intelligently detecting aging of a microprocessor according to an embodiment of the invention is shown, the method includes the following steps:
in step S001, a pin signal and a frequency-divided clock signal are obtained.
Selecting the pin of the output signal from the microprocessor, and recording the pin signal obtained by the monitoring data asAnd knowing the divided clock signal of the pin relative to the clock signal through an oscilloscope, denoted/>The obtained monitoring data are shown in fig. 2, wherein the first square wave signal in fig. 2 is a frequency division clock signal, and the second signal is a pin signal. The pins are wires leading from the internal circuitry of the integrated circuit (chip) to the peripheral circuitry.
Thus, the pin signal and the divided clock signal are obtained.
Step S002, obtaining voltage deviation according to the difference between the pin signal voltage and the high level and the low level; and acquiring an ideal waveform according to the pin signal and the frequency division clock signal, and decomposing the pin signal to acquire a first division signal.
At pin signalIn the method, phenomena representing aging fault information can be observed, namely a time delay phenomenon, a burr phenomenon and a voltage deviation phenomenon. The voltage deviation phenomenon can lead the actual voltage of the pin signal to deviate from the logic voltage, so the voltage deviation phenomenon has covering capability on the delay phenomenon and the burr phenomenon, and when the voltage deviation phenomenon is serious, noise interference can be mistakenly regarded as the delay phenomenon or the burr phenomenon, so the three phenomena can be comprehensively calculated to represent the aging degree of the microprocessor.
The voltage deviation phenomenon refers to the difference value between the average voltage and a high level or a low level when the pin signal represents logic '1' or '0', and the high level refers to the high voltage opposite to the low level, which is a description in electrical engineering, when the input level is higher than the input high voltage, the input level is the high level; when the input level is lower than the low voltage, the input level is low, the high level is 4.5 volts in the embodiment, the low level is 0.2 volts, and the voltage deviation is obtained by the difference between the average voltage and the high and low levels, and the formula is as follows:
in the method, in the process of the invention, Representing the voltage of the pin signal at time t,/>Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>Is the average value of the high level and the low level of the pin signal,/>The voltage deviation value at time t is indicated,Representing the voltage deviation value at the nth period and the t time/>The voltage deviation of the nth cycle is represented, and T represents the time of inclusion of one cycle.
The difference between the pin signal and the high level or the low level in one period is averaged to obtain a voltage deviation, the voltage deviation is represented in the period, the larger the voltage deviation is, the more serious the microprocessor is aged, the period is shown in fig. 2, and the two grids are taken as one period, namely, the time period from the low level to the next low level in the frequency division clock signal.
For the delay phenomenon and the burr phenomenon representing the aging degree of the microprocessor, when the characteristic calculation is carried out, the pin signals are obtainedThe square wave component characterizing the digital signal information affects it, so it is necessary to know the pin signal/>Because the digital information causes a trip point for voltage generation.
By pin signalsAnd frequency-divided clock signal/>As input, AD conversion algorithm is adopted, and output is a binary digital code which represents pin signal/>Digital signal information of the (b). Further encoded and divided clock signal/>, with the resulting binary numbersThe input is a DA conversion algorithm, and the output is a standard square wave signal which is recorded as ideal waveform/>As shown in fig. 3, the trip point is the moment when the signal turns over, and for ideal waveforms, only "1" and "0" exist, so that signal turning over refers to turning over from "1" to "0" or from "0" to "1", and the ideal waveform is recorded as one turn over from "0" to "1" to "0" or from "1" to "0" to "1".
Since the noise exists in the pin signals, the pin signals are used as input, and the EMD signal decomposition algorithm is used for outputting a plurality of EMD partial signals, and is not described in detail herein. Since the glitch phenomenon and the delay phenomenon are represented as high-amplitude value portions in the pin signal, the signal with the lowest frequency among the EMD partial signals is represented as a noise signal. Therefore, the signal with the lowest frequency in all the decomposed partial signals is eliminated, the rest signals are reconstructed by using the mean value filtering, and the reconstructed signals are recorded as the first partial signals, so that noise attenuation is completed.
Step S003, a time delay phenomenon observation time period of the first partial signal is obtained according to an ideal waveform, and a time delay phenomenon significance is obtained according to a voltage difference and a time period length of the time delay phenomenon observation time period; acquiring a signal jump interval time period, and acquiring the saliency of the burr phenomenon according to the voltage difference and the time length of the signal jump interval time period; based on this, a degree of significance of the microprocessor burn-in is obtained.
Delay refers to the inversion time and ideal waveform of the logic voltage of the output signalThe larger the time difference is, the more serious the delay phenomenon is. In the first partial signal, the first peak or trough appearing after the logic voltage turning moment can be observed, and the longer the duration of the wave band where the peak or trough is located, the larger the time difference between the pin signal turning moment and the ideal waveform signal turning moment is, the more serious the delay phenomenon is, and the delay phenomenon is shown in fig. 4.
To calculate the characteristics of the delay phenomenon, it is necessary to calculate the characteristics according to the ideal waveformThe first partial signal is divided in the event of a flip. Ideal waveform/>The moment when the overturn occurs is recorded as the overturn starting moment, and the overturn starting moment and the moment/>, are empirically recordedThe time period included at the time after the second is recorded as a time period for observing the time delay phenomenon, and the voltage value of the first partial signal is considered to be characteristic information of the time delay phenomenon in the time period for observing the time delay phenomenon. /(I)Is the time length of one period of the pin signal,/>For the size of the observation window of the time delay phenomenon, the empirical value is 0.15.
The method comprises the steps of obtaining a delay phenomenon voltage difference at each moment according to the voltage at each moment and the difference of high and low levels in a delay phenomenon observation time period, and obtaining a delay phenomenon significance according to the delay phenomenon voltage difference at each moment and the time length of the delay phenomenon observation time period, wherein the formula is as follows:
in the method, in the process of the invention, Voltage value representing the kth time in the observation period of the time delay phenomenon of the first partial signal,/>Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>Is the average value of the high level and the low level of the pin signal,/>Representing the delay phenomenon voltage difference at the kth moment in the delay phenomenon observation time period of the first partial signal,/>Time length representing time period of observation of time delay phenomenon,/>Indicating the significance of the delay phenomenon.
Voltage value of first partial signal at end time of time delay observation periodThe voltage value after the inversion can be used for judging whether the voltage value after the inversion belongs to a high voltage state corresponding to logic '1' or a low voltage state corresponding to logic '0'; the delay phenomenon voltage difference represents the difference value between the pin signal voltage and the ideal waveform voltage in the delay phenomenon observation time period, which is caused by the delay phenomenon, when the difference value is large, the delay phenomenon is indicated to be happening, and when the difference value is small, the delay phenomenon is indicated to be ended; therefore, in the time period of observing the delay phenomenon, the voltage difference of the delay phenomenon is averaged, the remarkable degree of the delay phenomenon is represented, and the larger the value is, the more remarkable the delay phenomenon is.
The time period from the adjacent rising edge to the falling edge or the time period from the adjacent falling edge to the rising edge in the first sub-signal is called a signal jump interval time period, and the time point from the rising edge to the high level is the time point from the high level to the low level. And the glitch phenomenon refers to any transition of the pin signal more than once across the logic threshold in the signal transition interval period, as shown in fig. 5. If no glitch occurs in the first partial signal, the voltage value should be kept stable during the signal transition interval period, and if a glitch occurs, a peak or a trough occurs during the signal transition interval period.
The burr phenomenon significance can be calculated through the analysis, and the formula is as follows:
in the method, in the process of the invention, Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>Is the average value of the high level and the low level of the pin signal,/>Voltage value representing mth moment in signal jump interval time period of first partial signal,/>Representing the time length of the signal transition interval period,/>Representing the length of time of one cycle,/>Indicating the significance of the glitch phenomenon. /(I)The number of the periods of the half pin signal contained in the signal jump interval period is the time length of the signal jump interval period divided by the time length of the half pin signal period and rounded up.
In the method, in the process of the invention,Is the difference between the voltage of the pin signal and the median voltage, and when no glitch occurs, the difference should be equal to half of the difference between the high level and the low level of the pin signal, which means that the voltage of the pin signal is always equal to the high level or the low level; when the burr phenomenon occurs, the pin signal is lower than the high level or higher than the low level in certain time periods in the time period of the signal jump interval, so that the difference value between the voltage of the pin signal and the median voltage is less than half of the difference value between the high level and the low level of the pin signal; thus the burr phenomenon saliency/>The larger the value, the more severe the glitch occurrence during the signal transition interval.
Because the delay phenomenon saliency and the burr phenomenon saliency exist in each turn in the pin signal, in one turn, a plurality of periods possibly exist, and each period corresponds to one voltage deviation, the voltage deviation average value of each turn can be obtained, and the microprocessor aging degree saliency of each turn is obtained according to the voltage deviation average value, the delay phenomenon saliency and the burr phenomenon saliency of each turn, wherein the formula is as follows:
in the method, in the process of the invention, Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>For pin signal (H /)The aging degree of the microprocessor which is turned over again is remarkable; /(I)For pin signal (H /)The time delay phenomenon of the secondary overturn is remarkable; /(I)For pin signal (H /)The saliency of the burr phenomenon of the secondary overturn; /(I)For pin signal (H /)Voltage deviation mean value of secondary flip,/>Is significance versus weight, in this example let/>。
The larger the delay phenomenon significance, the burr phenomenon significance and the voltage deviation mean value are, the more remarkable the aging degree of the microprocessor is represented. When voltage deviation occurs in the microprocessor, the higher the difference between the high level and the low level in the pin signal and the high voltage and the low voltage on the ideal waveform is, the less remarkable the waveform caused by the burr or the delay phenomenon is when the first divided signal is observed, and the more easily the waveform is covered by noise and the like. Therefore, the larger the voltage deviation mean value is, the lower the reliability of the delay phenomenon significance and the burr phenomenon significance is, the more unreliable.
Therefore, the difference between the delay phenomenon significance and the burr phenomenon significance and the voltage deviation mean value is taken as a voltage deviation covering judgment formula, compared with the comparison weight of the high and low level difference value multiplied by the significance, the obtained ratio is taken as a weight when the judgment formula is met, the delay phenomenon significance and the burr phenomenon significance are taken as weights for influencing the microprocessor aging degree significance, otherwise, only the voltage deviation mean value is taken as a part for influencing the microprocessor aging degree significance. When the microprocessor is aged to a significant degreeThe greater the degree of aging of the microprocessor is considered to be more significant.
Thus, the degree of aging significance of the microprocessor during each overturn is obtained.
Step S004, obtaining a Markov transition probability matrix according to the aging degree significance of the microprocessor, and constructing the aging index confusion.
The microprocessor aging degree significance during each overturn is obtained through the steps, however, due to the existence of systematic errors, a certain time period is selected as an aging observation window, and the pin signals in the aging observation window are usedAnd calculating the aging degree significance of the microprocessor, and calculating an average value to judge the aging degree of the microprocessor so as to eliminate statistical errors caused by systematic errors. However, the selected aging observation window is too small, so that systematic errors cannot be eliminated, and judgment errors are easy to occur; the excessively large aging observation window is selected to cause excessively large time interval for judging the aging degree each time, and the judgment of the aging degree of the microprocessor is not sensitive enough.
Therefore, according to the embodiment, the size of the aging observation window is adaptively adjusted according to the change condition of the aging degree significance of the microprocessor in a certain time, and the specific method is as follows:
Counting the remarkable degree of the aging degree of the microprocessor generated in the S times of overturning process to form an aging degree statistical vector . In this embodiment, the number of times is counted/>。
Calculating a differential vector for the aging degree statistical vector to obtain the aging degree statistical differential vector. The s-th element/>, in the aging degree statistical difference vectorFor the s-th element/>, in the aging statistics vectorMinus the s-1 st element/>Of (3), wherein/>At 0, the difference characterizes the/>, of the aging statistics vectorThe state of change of the individual elements, when/>When the number is positive, the aging degree significance of the characterization microprocessor is rising; when/>When negative, the significance of the degree of aging characterizing the microprocessor is decreasing.
Further, discretizing the aging degree statistical difference vector artificially into three types to construct a discrete matrixWhen/>When (1)The degree of degradation of the microprocessor with the secondary flip is decreasing significantly; /(I)When (1)The aging degree significance of the microprocessor which is turned over again is kept stable; when/>At the time of the firstThe degree of microprocessor burn-in significance of the secondary flip is rising. Wherein/>, of the present embodimentSet to 0.2.
The method comprises the steps of taking an aging degree statistical difference vector and a discrete matrix as input, adopting an algorithm of a Markov transition probability matrix to calculate, outputting a 3*3 Markov transition probability matrix, and constructing an aging index chaos degree according to the Markov transition probability matrix, wherein the formula is as follows:
in the method, in the process of the invention, Is the degree of confusion of the ageing index; /(I)Is the first/>, of the Markov transition probability matrixLine/>Column elements; first/>, markov transition probability matrix Line/>Column elements.
In the formula, the mean value of non-main diagonal elements of the Markov transition probability matrix is compared with the mean value of main diagonal elements, the non-main diagonal elements represent that states in the Markov transition probability matrix are changed, and the non-main diagonal elements represent that the states are stable and unchanged, so that the obtained ratio represents the degree of confusion of the states in the Markov transition probability matrix, and the larger the value, the more chaotic the states. Thus, the confusion of the ageing indexThe larger the value, the more confusing the microprocessor with a significant degree of burn-in, the longer the burn-in window should be.
And step S005, obtaining an aging degree index according to the disorder degree of the aging index, and finishing intelligent detection of the aging of the microprocessor.
The length of the aging observation window isAnd turning over again, and obtaining an aging degree index under the aging observation window by averaging the aging degree saliency of the microprocessor in the aging observation window.
Selecting A failed microprocessors, calculating the aging degree indexes in all aging observation windows of the day before failure for the A microprocessors, calculating the average value of all the aging degree indexes in the day before each micro-processing as an aging average valueThe mean value and the variance of the aging mean values of the A micro-processes are obtained and respectively recorded as a first aging mean value/>And first aging variance/>Obtaining an aging degree index threshold value through a first aging mean value and a first aging variance. In this embodiment, a is 3000.
When the aging degree index of the microprocessor to be detected is larger than the aging degree index threshold, the microprocessor to be detected is considered to have higher aging fault risk, namely, faults exist, so that intelligent aging detection of the microprocessor is completed.
The embodiment provides a microprocessor aging intelligent detection system, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to realize the methods of the steps S001 to S005.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (7)
1. The intelligent detection method for the aging of the microprocessor is characterized by comprising the following steps of:
acquiring a pin signal and a frequency division clock signal of a microprocessor;
Acquiring a voltage deviation value according to the difference between the voltage of each time of the pin signal and the high level as well as the low level, and acquiring the voltage deviation according to the accumulated sum of the voltage deviation values of all the time within one period; according to the pin signals and the frequency division clock signals, an AD conversion algorithm is adopted to obtain ideal waveforms, and the pin signals are decomposed by using EMD to obtain first division signals;
Acquiring turning time according to an ideal waveform, determining a delay phenomenon observation time period of a first partial signal through the turning time, and acquiring a delay phenomenon significance according to the difference of voltage and high level and low level at each time of the delay phenomenon observation time period and the time length of the delay phenomenon observation time period; acquiring a signal jump interval time period according to adjacent rising edges and falling edges in the first sub-signal, and acquiring the saliency of the burr phenomenon according to the voltage and the high level at each moment of the signal jump interval time period, the low level and the time length of the signal jump interval time period; acquiring the aging degree saliency of the microprocessor of each turnover according to the burr phenomenon saliency, the delay phenomenon saliency and the voltage deviation of each turnover;
counting the aging degree significance of the microprocessor turned over for a plurality of times to form an aging degree statistical vector, and acquiring an aging degree statistical differential vector based on the aging degree statistical vector; obtaining a Markov transition probability matrix according to the ageing degree statistical difference vector, and constructing ageing index confusion according to the Markov transition probability matrix;
determining the length of an aging window according to the disorder degree of the aging index to obtain an aging degree index, and finishing intelligent aging detection of the microprocessor according to the aging degree index;
the method for obtaining the burr phenomenon significance according to the voltage and the high level and the low level of each moment of the signal jump interval time period and the time length of the signal jump interval time period comprises the following steps:
in the method, in the process of the invention, Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>Is the average value of the high level and the low level of the pin signal,/>Voltage value representing mth moment in signal jump interval time period of first partial signal,/>Representing the time length of the signal transition interval period,/>Representing the length of time of one cycle,/>Representing the significance of the burr phenomenon;
The method for counting the aging degree significance of the microprocessor turned for a plurality of times to form an aging degree statistical vector and obtaining the aging degree statistical differential vector based on the aging degree statistical vector comprises the following steps:
counting the aging degree significance of the microprocessor turned over for S times, and sequencing the aging degree significance of the microprocessor according to a time sequence to form an aging degree statistical vector;
Taking the difference value between the latter element and the former element in the aging degree statistical difference vector as an aging difference value, and forming an aging degree statistical difference vector by all the aging difference values according to time sequence, wherein the value of the first element of the aging degree statistical difference vector is 0;
The method for determining the length of the aging window according to the disorder degree of the aging index to obtain the aging degree index and finishing the intelligent aging detection of the microprocessor according to the aging degree index comprises the following steps:
The product of the disorder degree of the aging index and the S is recorded as the total turnover number, and the aging observation window comprises the turnover number of the total turnover number;
in the length of the aging window, calculating the average value of the aging degree significance of all the overturned microprocessors as the aging degree index of the aging observation window;
Selecting A failed microprocessors, calculating the average value of the aging degree indexes of all aging observation windows of each microprocessor in the previous day, marking the average value as an aging average value, acquiring the average value and the variance of the aging average values of all the microprocessors, marking the average value and the variance as a first aging average value and a first aging variance, and marking the difference value between the first aging average value and three times of the first aging variance as an aging degree index threshold value;
When the aging degree index of the micro-processing to be detected is larger than the aging degree index threshold, the microprocessor to be detected has faults.
2. The intelligent detection method for microprocessor aging according to claim 1, wherein the method for obtaining the voltage deviation value according to the voltage and the difference of the high level and the low level at each moment of the pin signal is as follows:
Calculating the average value of the high level and the low level and recording the average value as a level average value;
if the voltage at each moment of the pin signal is greater than or equal to the level average value, making the absolute value of the difference value between the voltage at each moment of the pin signal and the high level as a voltage deviation value;
If the voltage at each time of the pin signal is smaller than the level average value, the absolute value of the difference between the voltage at each time of the pin signal and the low level is taken as the voltage deviation value.
3. The intelligent detection method for microprocessor aging according to claim 1, wherein the method for acquiring the inversion time according to the ideal waveform and determining the observation time period of the delay phenomenon of the first partial signal according to the inversion time comprises the following steps:
The turning moment is the moment when the signal is from 0 to 1 or from 1 to 0;
For an ideal waveform, a preset time period is taken as a time delay phenomenon observation time period after the turning moment, the preset time period is the product of a period and a decimal, and the period is the time period from a low level to the next low level in the frequency division clock signal.
4. The intelligent detection method for microprocessor aging according to claim 2, wherein the method for obtaining the significance of the delay phenomenon according to the difference between the voltage and the high level and the low level at each moment in the observation period of the delay phenomenon and the time length of the observation period of the delay phenomenon is as follows:
If the voltage at each moment in the time period of observing the time delay phenomenon is greater than or equal to the average value of the level, the absolute value of the difference value between the voltage at each moment in the time period of observing the time delay phenomenon and the low level is used as the voltage difference of the time delay phenomenon;
If the voltage at each moment in the time period of observing the time delay phenomenon is smaller than the level average value, making the absolute value of the difference value between the voltage at each moment in the time period of observing the time delay phenomenon and the high level be used as the voltage difference of the time delay phenomenon;
the ratio of the delay phenomenon voltage difference at each moment of the delay phenomenon observation time period to the time length of the delay phenomenon observation time period is recorded as a first ratio, and the accumulated sum of the first ratios at all moments is recorded as the delay phenomenon significance.
5. The intelligent detection method for microprocessor aging according to claim 1, wherein the method for obtaining the microprocessor aging degree saliency of each flip according to the burr saliency, the delay saliency and the voltage deviation of each flip comprises the following steps:
in the method, in the process of the invention, Representing a high level of the pin signal,/>Representing a low level of the pin signal,/>For pin signal (H /)The aging degree of the microprocessor which is turned over again is remarkable; /(I)For pin signal (H /)The time delay phenomenon of the secondary overturn is remarkable; /(I)For pin signal (H /)The saliency of the burr phenomenon of the secondary overturn; /(I)For pin signal (H /)Voltage deviation mean value of secondary flip,/>Is the saliency versus weight.
6. The intelligent detection method for microprocessor aging according to claim 1, wherein the method for obtaining a markov transition probability matrix according to the aging degree statistical difference vector and constructing the aging index confusion according to the markov transition probability matrix comprises the following steps:
dividing all elements in the aging degree statistical difference vector into three types to form a discrete matrix, and presetting a threshold value Will be less than or equal to/>Elements of (2) are classified into one class; will be greater than/>Less than/>Elements of (2) are classified into one class; will be greater than or equal to/>Wherein each class forms a vector, and three vectors form a discrete matrix;
the discrete matrix and the aging degree statistical difference vector are used as input, the algorithm of the Markov transition probability matrix is used for calculation, the output is the Markov transition probability matrix, and the expression of the aging index chaos degree is as follows:
in the method, in the process of the invention, Is the degree of confusion of the ageing index; /(I)Is the first/>, of the Markov transition probability matrixLine/>Column elements; /(I)First/>, markov transition probability matrixLine/>Column elements.
7. A microprocessor-based intelligent degradation detection system comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor, when executing the computer program, implements the steps of a microprocessor-based intelligent degradation detection method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410081205.0A CN117590838B (en) | 2024-01-19 | 2024-01-19 | Intelligent detection method and system for aging of microprocessor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410081205.0A CN117590838B (en) | 2024-01-19 | 2024-01-19 | Intelligent detection method and system for aging of microprocessor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117590838A CN117590838A (en) | 2024-02-23 |
| CN117590838B true CN117590838B (en) | 2024-05-03 |
Family
ID=89920633
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410081205.0A Active CN117590838B (en) | 2024-01-19 | 2024-01-19 | Intelligent detection method and system for aging of microprocessor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117590838B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN85107568A (en) * | 1984-10-17 | 1986-04-10 | 菲利浦光灯制造公司 | The circuit of control picture tube |
| CN101852839A (en) * | 2010-05-19 | 2010-10-06 | 中国科学院计算技术研究所 | Ageing predetermination and overspeed delay testing bifunctional system and method thereof |
| CN110287537A (en) * | 2019-05-27 | 2019-09-27 | 西北大学 | Anti- outlier method for adaptive kalman filtering for frequency marking output transition detection |
| CN117269734A (en) * | 2023-11-17 | 2023-12-22 | 沈阳安姆迅电子有限公司 | Electrified ageing detection system of mainboard |
-
2024
- 2024-01-19 CN CN202410081205.0A patent/CN117590838B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN85107568A (en) * | 1984-10-17 | 1986-04-10 | 菲利浦光灯制造公司 | The circuit of control picture tube |
| CN101852839A (en) * | 2010-05-19 | 2010-10-06 | 中国科学院计算技术研究所 | Ageing predetermination and overspeed delay testing bifunctional system and method thereof |
| CN110287537A (en) * | 2019-05-27 | 2019-09-27 | 西北大学 | Anti- outlier method for adaptive kalman filtering for frequency marking output transition detection |
| CN117269734A (en) * | 2023-11-17 | 2023-12-22 | 沈阳安姆迅电子有限公司 | Electrified ageing detection system of mainboard |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117590838A (en) | 2024-02-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7479908B2 (en) | Semiconductor device including A/D converter | |
| US8624631B2 (en) | Programmable pulse width discriminator | |
| US10516407B2 (en) | Signal processing device | |
| US8378707B2 (en) | Evaluation of an output signal of a device under test | |
| US9634680B1 (en) | Large-error detection and correction of digital sample sequence from analog-to-digital converter | |
| US9281832B1 (en) | Circuits and methods for bandwidth estimation optimization of analog to digital converters | |
| US20220382322A1 (en) | Glitch detector | |
| CN117590838B (en) | Intelligent detection method and system for aging of microprocessor | |
| US4025768A (en) | Method and apparatus for testing and diagnosing data processing circuitry | |
| US20220404859A1 (en) | Clock generating circuit and method for trimming period of oscillator clock signal | |
| CN111092611A (en) | Signal processing device and method with small edge slope | |
| US20220187423A1 (en) | Feedback system monitoring | |
| US6098195A (en) | Multiple recent event age tracking method and apparatus | |
| US10698018B2 (en) | Noise detection circuit capable of recognizing noise event occurring in device under test, and related system and method for testing device under test | |
| CN113283316A (en) | Switch mechanical fault diagnosis method, device and equipment based on sound signals | |
| CN114083987A (en) | Battery monitoring parameter correction method and device and computer equipment | |
| CN117851414B (en) | Lightning arrester aging test data storage method and system | |
| CN110187167A (en) | A kind of detection method and device of the load switch event based on manifold classification | |
| CN111887894B (en) | Normalization software processing method and system for fetal heart monitor Doppler signals | |
| CN103345945A (en) | Memory testing device with frequency testing function, as well as memory testing method | |
| US12047087B2 (en) | Ad converter | |
| US20120056645A1 (en) | Analog to Digital Acquisition Eliminating Uncertainty of Level Test in High Noise Environments | |
| CN117668445A (en) | Calculation method of pulse waveform parameters | |
| CN116192142B (en) | Sampling monitoring circuit, method, chip, electronic device and storage medium | |
| CN115882823A (en) | Software filtering method of fA-level weak current signal, electronic device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |