# US 8,788,554 B2 · Sub-band adaptive FIR-filtering

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 8,788,554 B2
- Document ID: US8788554B2
- Status: Granted
- Kind: B2
- Application number: 13/038,964
- Filed: 2011-03-02
- Published: 2011-09-08
- Granted: 2014-07-22
- Earliest claimed: 2010-03-02
- Assignees: Harman Becker Automotive Systems GmbH
- Applicants: Markus Christoph
- Inventors: Markus Christoph
- Web page: https://patentgrep.com/patent/US8788554B2

## CPC

- H03H 21/0012: {Digital adaptive filters}
- G10L 19/0204: {using subband decomposition}
- H03H: IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS (waveguides, resonators, lines or other devices of the waveguide type H01P)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A method for designing a set of sub-band FIR filters, where each FIR filter has a number of filter coefficients and is connected to an adjustable delay line. The method includes dividing an input signal into a number of sub-band signals, where a spectrum of the input signal comprises spectra of the sub-band signals; providing a respective goal sub-band signal for each sub-band dependent on a goal signal; filtering and delaying each sub-band signal using a corresponding FIR filter and delay line to provide filtered signals; providing error signals for each sub-band dependent on the filtered signals and the corresponding goal signals; adapting the filter coefficients of each sub-band FIR filter such that the respective filtered signal approximately matches a corresponding goal sub-band signal; and changing a respective delay of the delay line for each sub-band to reduce or increase a first quality criterion.

## Claims

### Claim 1

A method for designing a set of sub-band FIR filters, where each FIR filter has a number of filter coefficients and is connected to an adjustable delay line, the method comprising:
dividing an input signal into a number of sub-band signals, where a spectrum of the input signal comprises spectra of the sub-band signals;
providing a respective goal sub-band signal for each sub-band dependent on a goal signal;
filtering and delaying each sub-band signal using a corresponding FIR filter and delay line to provide filtered signals;
providing error signals for each sub-band dependent on the filtered signals and the corresponding goal signals;
adapting the filter coefficients of each sub-band FIR filter in response to the error signals; and
changing a respective delay of the delay line for each sub-band in response to the error signals.

### Claim 2 (depends on 1)

The method of claim 1, where lengths of impulse responses of the sub-band FIR filters differ from each other in accordance with a psycho-acoustically motivated frequency scale.

### Claim 3 (depends on 2)

The method of claim 2, where the psycho-acoustically motivated frequency scale comprises one of a Bark scale, a Mel scale and an ERB scale.

### Claim 4 (depends on 1)

The method of claim 1, further comprising changing the number of filter coefficients of the respective FIR filter for each sub-band to increase or decrease a second quality criterion.

### Claim 5 (depends on 4)

The method of claim 4, where the second quality criterion depends on at least one of energy of the sub-band error signal, energy of the sub-band input signal, and energy of the sub-band FIR filter coefficients.

### Claim 6 (depends on 5)

The method of claim 5, where the second quality criterion depends on energy of a subset of consecutive filter coefficients including an end filter coefficient.

### Claim 7 (depends on 5)

The method of claim 5, where the second quality criterion depends on the energy of the sub-band error signal given a number of consecutive samples of the respective error signal, which include a latest sample of the respective error signal.

### Claim 8 (depends on 4)

The method of claim 4, where the second quality criterion is weighted in each sub-band by a weighting factor that represents a frequency resolution of a human auditory system.

### Claim 9 (depends on 1)

The method of claim 1, where the first quality criterion depends on energy of the filter coefficients of the respective sub-band FIR filter for each sub-band.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
defining a first sub-set of filter coefficients of the respective sub-band FIR filter for each sub-band;
defining a second sub-set of filter coefficients of the respective sub-band FIR filter for each sub-band; and
assuming a minimum for the first quality criterion when, in the respective sub-band, the energy of a first sub-set of filter coefficients substantially equals the energy of a second sub-set of filter coefficients.

### Claim 11 (depends on 1)

The method of claim 1, further comprising:
calculating a norm of a sum of the error signals of each sub-band; and
assuming a minimum for the first quality criterion when the norm becomes minimal.

### Claim 12 (depends on 1)

The method of claim 1, where the dividing of the input signal into the sub-band signals comprises filtering the input signal with an evenly stacked generalized discrete Fourier transform filter bank.

### Claim 13 (depends on 12)

The method of claim 12, where the sub-band input signal and corresponding FIR filter coefficients are real-valued in lowest and highest sub-bands and complex-valued in other sub-bands.

### Claim 14

A method for filtering an input signal with a set of sub-band filters, where each filter has a number of filter coefficients and is connected to an adjustable delay line, the method comprising:
dividing the input signal into a number of sub-band signals, where a spectrum of the input signal comprises spectra of the sub-band signals;
providing a respective goal sub-band signal for each sub-band dependent on a goal signal;
filtering and delaying each sub-band signal using a corresponding filter and delay line to provide filtered signals;
providing error signals for each sub-band dependent on the filtered signals and the corresponding goal signals;
adapting the filter coefficients of each sub-band filter in response to the error signals; and
changing a respective delay of the delay line for each sub-band in response to the error signals.