Method to extract target signals of a known type from raw data containing an unknown number of target signals, interference, and noise

• US 8,073,634 B2
• Filed: 09/22/2008
• Issued: 12/06/2011
• Est. Priority Date: 09/22/2008
• Status: Expired due to Fees

• Current Assignee: University of Ottawa
• Original Assignee: University of Ottawa
• Inventors: Yeap, Tet Hin, Liang, Ming, Fan, Xianfeng, Bozchalooi, Iman Soltani
• Primary Examiner(s): Tsai, Carol
• Application Number:
  US12/235,008
• Publication Number:
  US 20100076693A1
• Time in Patent Office: 1,170 Days
• Field of Search: 706/15, 702/22, 702/193, 367/87, 367/93, 367/135, 367/100, 367/90, 367/91, 367/131, 367/136, 736/02, 342/90, 342/27, 342/89, 342/159, 342/175, 342/195
• US Class Current: 702/22
• CPC Class Codes: G05B 23/0229 knowledge based, e.g. expert systems; genetic algorithms
• CPC Class Codes: G06F 2218/06 by applying a scale-space analysis, e.g. using wavelet analysis

Abstract

A signal analysis method extracts transient target signals of known type from a raw data source signal that contains an unknown number of target signals. The method can enhance the analysis of data obtained from in-line oil-debris sensors. The method comprises steps of: defining signatures of the known target signal, and of at least one of the intrinsic noise and interfering signals; performing a mathematical transform that decomposes the raw data into distinct data sets; using the signal signatures to identify and nullify the data sets containing noise and interfering signal signatures; using the target signal signatures to identify the data sets containing target signal components, or may further use a thresholding rule to remove intrinsic noise from said data sets, and finally applying the inverse transform to the processed data sets in order to reconstruct an enhanced output signal.

Claims

1. A method for extracting intermittent, randomly occurring transient target signals of known type from a raw data source signal containing an unknown number of the target signals in addition to interfering signals or intrinsic noise signals, the method comprising:

a) obtaining the raw data source signal from a sensor for detecting the passage of particles or bubbles in a fluid flow, wherein a target signal is generated by the sensor when detecting a particle or bubble;

b) defining quantifiable signatures or characteristics that respectively and independently represent the target signals, the interfering signals or the intrinsic noise signals, wherein the signatures comprise a frequency range of the target signals and a kurtosis value range of wavelet coefficients of respective wavelet scales of interest that reflect a transient nature of the target signals;

c) applying a Time-Invariant Wavelet Transform (TIWT) to decompose the raw data source signal into distinct data sets in a form of wavelet coefficients of the respective wavelet scales;

d) processing said data sets to identify a first group of data sets which display the signatures or characteristics representing the interfering signals or intrinsic noise signals and a second group of data sets which display the signatures or characteristics representing the target signals;

e) setting the data sets of the first group to zero; and

f) applying an inverse transform to the processed data sets in order to reconstruct a processed output signal.

2. The method as defined in claim 1 wherein step (d) further comprises removing any intrinsic noise signals which remain in the data sets of the second group.

3. The method as defined in claim 1 wherein the signatures comprise two kurtosis value ranges of the wavelet coefficients of each wavelet scale that correspond to the interfering signals and correspond to the intrinsic noise signals, respectively.

4. The method as defined in claim 2 comprising employing a thresholding rule to process the data sets of the second group in order to reduce the intrinsic noise signals.

5. The method as defined in claim 1 wherein an Inverse Time-Invariant Wavelet Transform (ITIWT) is applied to the processed data sets for reconstructing the processed output signal.

6. The method as defined in claim 1 wherein the raw data source signal comprises acoustic signals.

7. The method as defined in claim 1 wherein the raw data source signal comprises optical signals.

8. The method of claim 1 in which the raw data source signal is obtained from sonar or radar apparatus.

9. The method as defined in claim 1 wherein the raw data source signals comprise electronic signals.

10. The method as defined in claim 1 wherein the frequency range of the target signals is determined from information relating to a physical dimension of the sensor and a velocity of the fluid flow.

11. The method as defined in claim 1 wherein the raw data source signal obtained from the sensor is decomposed into a number of scales so that a frequency band width of a highest scale is smaller than a lowest frequency of the target signals.

12. The method as defined in claim 1 wherein the identifying process of the data sets comprises calculating a kurtosis value for each wavelet scale of the raw data source signal and then comparing the same with the kurtosis value range of the wavelet coefficients associated with the defined signatures of the target signals, thereby identifying the respective first and second group of data sets.

13. The method as defined in claim 12 wherein wavelet coefficients in respective wavelet scales are down-sampled prior to an inverse transform according to a decomposition level, following a kurtosis value calculation in the respective scales of said raw data source signal, in order to reduce memory requirement in hardware implementation.

14. The method as defined in claim 13 wherein the inverse transform is an Inverse Discrete Wavelet Transform (IDWT).

15. A method of enhancing the capability of a sensor for sensing a particle or bubble in a fluid flow, the sensor being adapted for generating transient target signals when particles or bubbles in the fluid pass through the sensor, the method comprising steps of:

a) defining quantifiable signatures of the target signals, having a known frequency range of the target signals and a known kurtosis value range of wavelet coefficients of respective wavelet scales of interest that reflect a transient nature of the target signals;

b) defining quantifiable signatures of interfering signals and intrinsic noise signals, having two known kurtosis value ranges of wavelet coefficients of each wavelet scale that correspond to the interfering signals and correspond to the intrinsic noise signals, respectively;

c) obtaining a raw data sample signal in a time domain from the sensor in use for monitoring the fluid flow;

d) applying a Time-Invariant Wavelet Transform (TIWT) to said raw data sample signal to effect a decomposition of said raw data sample signal into a plurality of data sets in a form of wavelet coefficients of respective wavelet scales;

e) processing the data sets to calculate a kurtosis value for each data set to compare the kurtosis value for each data set with the kurtosis value range defined in step (a) or the kurtosis value ranges defined in step (b), in order to identify a first group of data sets which display the signatures representing the interfering signals and the intrinsic noise signals and a second group of data sets which display the signatures representing the target signals, and then setting the data sets of the first group to zero;

f) applying a thresholding rule to process the data sets of the second group in order to reduce any intrinsic noise signals which remain in the data sets of the second group; and

g) constructing a processed output signal by applying an inverse transform to the data sets processed in steps (e) and (f).