Dynamic threshold methods for filtering noise and restoring attenuated high-frequency components of acoustic signals

Patent metadata

Field	Value
Publication number	US-9684087-B2
Application number	US-201414485562-A
Country	US
Kind code	B2
Filing date	Sep 12, 2014
Priority date	Sep 12, 2013
Publication date	Jun 20, 2017
Grant date	Jun 20, 2017

Abstract

Methods of providing for filtering noise and/or restoring attenuated spectral components in acoustic signals, are provided. An exemplary embodiment of a method includes dynamically filtering each of a plurality of raw FFT data samples of a record to remove or attenuate background noise contained therein to thereby produce a corresponding plurality of cleaned FFT data samples. The sample-specific background noise is removed or attenuated by a record-specific dynamic filter to produce the corresponding cleaned FFT data samples. The method can also include restoring the attenuated high-frequency components of the cleaned data samples through application of a record-specific Restoring Processor at least partially defined by a portion of the cleaned data samples and a Gain Function to thereby produce cleaned and restored data samples, and applying an inverse transformation to convert the cleaned and restored data samples into cleaned and restored data samples in time domain data.

First claim

The invention claimed is:

1. A method of filtering noise and restoring attenuated spectral components in acoustic signals, the method comprising the steps of:
recording, via an acoustic sensor affixed to a metal adapter attaches to a machine, acoustic signal samples, wherein the acoustic signal samples are grouped into one or more acoustic records, and wherein each acoustic record comprises a subset of the acoustic signal samples in time domain format:
performing, for each record of one or more acoustic records, the step of:
transforming the acoustic signal samples of the record from time domain format to frequency domain format to produce a plurality of raw data samples for the record in frequency domain format, wherein each raw data sample of the plurality of raw data samples comprises sample-specific acoustic signal data and sample-specific background noise, and
dynamically filtering each of a plurality of raw data samples in frequency domain format for the respective record to remove or attenuate background noise contained therein to produce a corresponding plurality of cleaned data samples for the respective record, the cleaned data samples for the respective record representing the acoustic signal samples for the respective record with reduced background noise, the dynamic filtering comprising applying, to each raw data sample of the plurality of raw data samples, a record specific dynamic filter to remove or attenuate the sample-specific background noise of the raw data sample to produce a corresponding plurality of cleaned data samples for the raw data sample, wherein the plurality of cleaned data samples for the respective record comprise the plurality of cleaned data samples for the raw data sample,
the record-specific dynamic filter at least partially being defined by a dynamic amplitude noise cutoff (A th ),
the dynamic amplitude noise cutoff (A th ) being defined by a record-specific base noise percentile (P b ) and a record-specific value of a threshold parameter (C th ),
the record-specific base noise percentile (P b ) being the same for each of the plurality of raw data samples for the respective record,
the threshold parameter value (C th ) being the same for each of the plurality of raw data samples for the respective record, and
the dynamic amplitude noise cutoff (A th ) being separately evaluated for and applied to each separate raw data sample of the plurality of raw data samples for the respective record, and

determining, using the cleaned data samples of the one or more acoustic records, a petrophysical property of rock.

2. A method as defined in claim 1 , wherein the sample-specific acoustic signal data contained in each separate one of the plurality of cleaned data samples includes substantially attenuated high-frequency components, the method further comprising the steps of:
restoring the attenuated high-frequency components of the cleaned data samples to thereby produce cleaned and restored data samples being in the frequency domain; and
applying an inverse transformation to convert the cleaned and restored data samples into cleaned and restored data samples in time domain data.

3. A method as defined in claim 1 , wherein the sample-specific background noise is time-varying, and wherein the dynamic filter is a tuned dynamic filter, the method further comprising the step of tuning an initial record-specific dynamic filter to form the tuned dynamic filter, the step of tuning including the steps of:
determining the record-specific Base Noise Percentile for each respective record of the one or more records, the record-specific Base Noise Percentile comprising a K th percentile within a record-specific Specific Frequency Range of an amplitude spectrum of each of the plurality of raw data samples for the respective record, below which each frequency component within the Specific Frequency Range of the respective amplitude spectrum of each of the plurality of raw data samples for the respective record is treated as background noise with substantial certainty; and
determining the record-specific value for the threshold parameter, the record-specific threshold parameter comprising one of the following: a threshold factor to be multiplied with the record-specific base noise percentile and a threshold elevator to be added to the record-specific base noise percentile, to thereby determine a value for the dynamic amplitude noise cutoff to be applied separately to each of the plurality of raw data samples.

4. A method as defined in claim 1 , wherein the dynamic filter is a tuned dynamic filter, the method further comprising the step of tuning an initial record-specific dynamic filter, the step of tuning including the steps of:
receiving or retrieving a subset of the plurality of raw data samples the respective record of the one or more records;
selecting a record-specific Specific Frequency Range for the respective record;
selecting the record-specific Base Noise Percentile for the respective record;
selecting a value of the record-specific Threshold Parameter for the respective record;
determining the Dynamic Amplitude Noise Cutoff for the respective record, the Dynamic Amplitude Noise Cutoff defined by the selected Base Noise Percentile and the selected record-specific value of the Threshold Parameter; and
evaluating results of an initial Dynamic Filter at least partially defined by the Dynamic Amplitude Noise Cutoff, on one or more samples within a set of representative data samples extracted from the plurality of raw data samples to thereby construct the tuned dynamic filter.

5. A method as defined in claim 1 , wherein the sample-specific background noise is time-varying, and wherein the dynamic filter is a tuned dynamic filter, the method further comprising the step of tuning an initial record-specific dynamic filter to form the tuned dynamic filter, the step of tuning including the step of:
receiving or retrieving a subset of the plurality of raw data samples for each respective record of the one or more records, wherein:
if the respective record is a recorded record, performing the step of retrieving a subset of the plurality of raw data samples recorded at substantially different times with different background noise levels to thereby define a set of Representative FFT Data samples; and
if the respective record is an online record to be processed and the raw data samples cannot be selected at substantially different times, performing the step of receiving a subset of the plurality of raw data samples at a beginning of the respective record to thereby define the set of Representative FFT Data samples.

6. A method as defined in claim 1 , wherein the sample-specific background noise is time-varying, and wherein the dynamic filter is a tuned dynamic filter, the method further comprising the step of tuning an initial record-specific dynamic filter to form the tuned dynamic filter, the step of tuning including the step of:
selecting a Specific Frequency Range for a respective record of the one or more records, the Specific Frequency Range defined by a range of frequencies common to each sample of a set of Representative FFT Data samples containing frequency components being dominated by background noise, or if no range of frequencies is dominated by background noise, a range of frequencies common to each of the samples of the set of Representative FFT data samples containing a higher percentage of background noise than other substantial ranges of consecutive frequencies of the set of Representative FFT data samples.

7. A method as defined in claim 6 , wherein the record-specific Base Noise Percentile is an operational record-specific Base Noise Percentile, the method further comprising the step of:
selecting an initial Base Noise Percentile

Assignees

• Saudi Arabian Oil Co

Inventors

• Yang Yunlai

Classifications

• G10L21/0232

  Processing in the frequency domain · CPC title

• G01N29/32

  Arrangements for suppressing undesired influences, e.g. temperature or pressure variations {, compensating for signal noise} · CPC title

• G01V2210/40

  Transforming data representation · CPC title

• G01N29/11 Primary

  by measuring attenuation of acoustic waves · CPC title

• G10L2021/02163

  Only one microphone · CPC title