Method and device for processing a signal

• US 6,069,975 A
• Filed: 11/16/1998
• Issued: 05/30/2000
• Est. Priority Date: 05/15/1996
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Heilmann, Christian, Lehmann, Klaus, Kartmann, Peter
• Primary Examiner(s): MEHTA, BHAVESH M
• Application Number:
  US09/192,755
• Time in Patent Office: 561 Days
• Field of Search: 382/100, 382/207, 382/128, 382/276, 382/280, 382/281, 702/66, 702/69-71, 702/75-79, 708/400, 708/443, 708/820, 708/822, 324/76.47, 375/272, 375/278, 375/303, 375/324, 375/334, 375/338, 375/346
• US Class Current: 382/207
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06F 2218/08 Feature extraction

Abstract

A signal is processed for recognize an irregularity in the signal. First, an orthogonal imaginary signal corresponding to the signal to be processed is formed. This is followed by a transformation step, which is then followed by a differentiation step. The signal thus processed shows the irregularity particularly clearly, so that the irregularity can be isolated, selected and detected. Preferred fields of application for the present method and corresponding devices are in the medical field, information technology, and signal treatment in network engineering, in particular low, medium, and high voltage engineering.

Claims

1. A method for detecting an irregularity in a signal, which comprises the following steps:

forming an orthogonal imaginary signal corresponding to a signal to be processed;

producing a time-dependent further signal based on the signal and the imaginary signal;

differentiating the further signal to form a differentiated signal; and

subjecting the differentiated signal to detection with regard to an irregularity in the signal.

2. The method according to claim 1, whereby the producing step comprises forming a common mean signal as a weighted function of time from the signal and the imaginary signal.

3. The method according to claim 2, wherein the common mean signal is formed from the relationship ##EQU11## wherein k1 and k2 are weighting factors, s(t) is the signal, and s(t) is the imaginary signal.

4. The method according to claim 3, which comprises setting the factors k1 and k2 as being one of equal and unequal.

5. The method according to claim 1, wherein the producing step comprises forming the further signal as a ratio signal from the imaginary signal and the signal, and which further comprises matching the ratio signal to a finite set of values using a mapping function.

6. The method according to claim 1, which comprises differentiating the further signal at least two times.

7. The method according claim 6, wherein the subjecting step comprises subjecting the multiply differentiated signal to pattern recognition for detecting a signal irregularity.

8. The method according claim 1, wherein the subjecting step comprises subjecting the differentiated signal to pattern recognition for detecting a signal irregularity.

9. The method according to claim 1, which further comprises generating a message signal if an irregularity is recognized in the subjecting step.

10. The method according to claim 9, wherein the message signal is a signal selected from the group consisting of a display signal and a warning signal.

11. The method according to claim 1, wherein the signal to be processed is a signal used in a medical apparatus.

12. The method according to claim 11, wherein the signal to be processed is a signal selected from the group consisting of an on-line phonocardiogram signal, a late potential signal, and an EEG signal.

13. The method according to claim 1, which comprises processing the signal in a demodulation of a frequency-shift keying signal.

14. The method according to claim 1, wherein the signal to be processed is a signal in an electrical power supply.

15. The method according to claim 14, wherein the signal is a signal in one of low, medium, and high voltage engineering.

16. The method according to claim 14, wherein the signal is selected from the group consisting of a current signal and a voltage signal in a power supply network.

17. The method according to claim 16, which further comprises feeding the differentiated signal to a protective device.

18. The method according to claim 16, which further comprises feeding the differentiated signal to a metering instrument.

19. An apparatus for detecting an irregularity in a signal, comprising:

a first device for forming an orthogonal imaginary signal from a signal to be processed;

a second device connected to said first device for producing a time-dependent further signal based on the signal and the imaginary signal;

a differentiating member connected to receive the further signal from said second device for forming a differentiated signal; and

a detector connected to said differentiating member for detecting from the differentiated signal an irregularity in the signal.

20. The device according to claim 19, wherein said second device comprises an averager for producing from the signal and the imaginary signal a common mean signal as a function of time.

21. The device according to claim 19, wherein said second device comprises a divider component for producing a ratio signal from the imaginary signal and the signal, and a mapping component for mapping the ratio signal onto a finite set of values using a mapping function.

22. The device according to claim 19, wherein said differentiating component is a first differentiating component, and including a second differentiating component connected to said first differentiating component for differentiating the differentiated signal output by said first differentiating component.

23. The device according to claim 22, which further comprises a third differentiating component connected to said second differentiating component.

24. The device according to claim 19, wherein said differentiating component has a second or further differentiating function.

25. The device according to claim 19, which further comprises a pattern-recognition module connected to said differentiating component.

26. The device according to claim 19, which comprises a computer, and wherein said first device, said second device, said differentiating member, and said detector are respective program modules being processed in said computer.