Identifying glitches and levels in mixed-signal waveforms

Patent metadata

Field	Value
Publication number	US-12326466-B2
Application number	US-202218071917-A
Country	US
Kind code	B2
Filing date	Nov 30, 2022
Priority date	Dec 3, 2021
Publication date	Jun 10, 2025
Grant date	Jun 10, 2025

Abstract

Various embodiments disclosed herein provide for a glitch detection and level detection method that use information contained in the signal itself to determine at which resolution or granularity the glitch detection and level detection operates. In particular, the glitch detection method comprises defining a glitch in terms of a change in the area under the waveform which can serve to disambiguate glitches from noises and other transient side effects of level transmissions. Likewise, the level detection method uses an entropy-based metric to identify levels that are significant in context of the entire signal and not in absolute terms.

First claim

What is claimed is:

1. A method to detect a glitch in a signal, comprising:
receiving the signal comprising an analog or mixed-signal waveform;
uniformly sampling the signal to abstract it as a time-series comprising real values;
determining a sequence of rates of change in area under the signal in sequential time periods, wherein the time periods comprise a plurality of time points;
determining that a glitch in the signal has occurred in response to determining that among adjacent time periods the a first absolute value of a rate of change of area between the first and second time periods is above a rate of change threshold, a second absolute value of a rate of change of area between the second and third time periods is above the rate of change threshold, and a third absolute value of a rate of change of area between first and third time periods is below the rate of change threshold; and
performing a signal processing operation based on determining that the glitch in the signal has occurred, wherein the signal processing operation comprises displaying a graphical representation of the waveform and graphically identifying the glitch.

2. The method of claim 1 , wherein the sequence of rates of changes of the values of the characteristic of the signal is based on a function of the values of the characteristic of the signal over the time period.

3. The method of claim 2 , wherein the function represents an area under the analog or mixed-signal waveform associated with the signal.

4. The method of claim 1 , wherein the characteristic of the signal is at least one of voltage, current, power, energy, or impedance or some function thereof.

5. The method of claim 1 , wherein a sum of a first rate of change of the first time period and a second rate of change of the second time period is below the rate of change threshold.

6. The method of claim 5 , wherein the first rate of change of the first time period is in an opposite direction to the second rate of change of the second time period.

7. The method of claim 1 , wherein the signal is uniformly sampled.

8. A method to detect a steady-state levels in a signal, comprising:
receiving the signal comprising an analog or mixed-signal waveform;
sampling the signal to abstract it as a time-series consisting of real values;
determining a width of a time period to detect the levels in the signal, based on a function of entropy associated with the signal, wherein the level indicates that a value of a characteristic of the signal does not change beyond a specified range during the time period;
determining that a level has occurred in response to values of characteristics of the signal remaining within a value range for time points associated with the time period; and
performing a signal processing operation based on determining that the level in the signal has occurred, wherein the signal processing operation comprises displaying a graphical representation of the waveform at a resolution determined based on the analog or mixed signal waveform and graphically identifying the level.

9. The method of claim 8 , further comprising:
determining a value of level-time based on the granularity at which signal is viewed.

10. The method of claim 8 , wherein the entropy associated with the signal comprises a first entropy associated with the series of time points and a second entropy associated with a subset of the series of time points.

11. The method of claim 10 , wherein the determining the time period further comprises identifying the value of the time period that results in a lowest difference between the second entropy and the first entropy.

12. The method of claim 10 , wherein the first entropy is computed based on a distribution of percentage of time points of the time-series across the value range of the time-series.

13. The method of claim 8 , wherein the signal is uniformly sampled.

14. The method of claim 10 , wherein the second entropy is based on a distribution of percentage of time points of a derived time-series across the value range of the derived time-series where,
the derived time-series is obtained by computing the entropies in sequential time periods of the original time-series.

15. The method of claim 8 , wherein the characteristic of the signal is at least one of voltage, current, power, energy, impedance, or some function thereof.

16. A signal processing device, comprising:
a memory that stores computer-executable instructions
a processor that executes the computer-executable instructions to perform operations, comprising:
receive the signal comprising an analog or mixed-signal waveform;
uniformly sample the signal to abstract it as a time-series comprising real values;
determine a sequence of rates of change in area under the signal in sequential time periods, wherein the time periods comprise a plurality of time points;
determine that a glitch in the signal has occurred in response to determining that among adjacent time periods the a first absolute value of a rate of change of area between the first and second time periods is above a rate of change threshold, a second absolute value of a rate of change of area between the second and third time periods is above the rate of change threshold, and a third absolute value of a rate of change of area between first and third time periods is below the rate of change threshold;
determine a width of a time period to detect the levels in the signal, based on a function of entropy associated with the signal, wherein the level indicates that a value of a characteristic of the signal does not change beyond a specified range during the time period;
determine that a level has occurred in response to values of characteristics of the signal remaining within a value range for time points associated with the time period; and
display a graphical representation of the waveform at a resolution determined based on the analog or mixed signal waveform and graphically identifying the glitch and the level.

17. The signal processing device of claim 16 , wherein the characteristic of the signal is at least one of voltage, current, power, energy, or impedance or some function thereof.

18. The signal processing device of claim 16 , wherein the sequence of rates of changes of the values of the characteristic of the signal is based on a function of the values of the characteristic of the signal over the time period.

19. The signal processing device of claim 16 , wherein the function represents an area under the analog or mixed-signal waveform associated with the signal.

20. The signal processing device of claim 16 , wherein the entropy associated with the signal comprises a first entropy associated with the series of time points and a second entropy associated with a subset of the series of time points.

21. The signal processing device of claim 20 , wherein the determining the time period further comprises identifying the value of the time period that results in a lowest difference between the second entropy and the first entropy.

22. The signal processing device of claim 20 , wherein the first entropy is computed based on a distribution of percentage of time points of the time-series across the value range of the time-series, and wherein the second entropy is based on a distribution of percentage of time points of a derived time-series across the value range of the derived time-series where.

Assignees

• Indian Institute Of Tech Kharagpur
• Texas Instruments Inc

Inventors

• Sanyal Sayandeep
• Dasgupta Pallab
• Hazra Aritra
• Morrison Scott
• Surendran Sudhakar
• Balasubramanian Lakshmanan

Classifications

• G01R29/00

  Arrangements for measuring or indicating electric quantities not covered by groups G01R19/00 - G01R27/00 · CPC title

• G01R23/16 Primary

  Spectrum analysis; Fourier analysis · CPC title

• G01R29/027 Primary

  Indicating that a pulse characteristic is either above or below a predetermined value or within or beyond a predetermined range of values · CPC title