HEYSER SPIRAL LOW FREQUENCY CORRECTION OF FIR FILTERS

• US 20110317853A1
• Filed: 06/29/2010
• Published: 12/29/2011
• Est. Priority Date: 06/29/2010
• Status: Active Grant

• Current Assignee: Bosch Security Systems Incorporated (Robert Bosch GmbH), Robert Bosch GmbH
• Original Assignee: Bosch Security Systems Incorporated (Robert Bosch GmbH), Robert Bosch GmbH
• Inventors: Donarski, Matthew A.
• Granted Patent: US 8,306,242 B2
• Time in Patent Office: Days
• US Class Current: 381/103
• CPC Class Codes: H03G 5/005 of digital signals
• CPC Class Codes: H03G 5/165 Equalizers; Volume or gain control in limited frequency bands
• CPC Class Codes: H04R 2203/00 Details of circuits for transducers, loudspeakers or microphones covered by H04R3/00 but not provided for in any of its subgroups

Abstract

A method of operating a loudspeaker includes providing a digital audio signal and identifying a target transfer function to be applied to the signal. At least one coefficient of an FIR filter is generated. The generating includes performing Heyser spiral curve fitting, and fitting a three-dimensional curve based on a magnitude and phase of a target transfer function. The digital audio signal is filtered through the FIR filter. The filtered signal is inputted into the loudspeaker.

Claims

1. A method of operating a loudspeaker, the method including:

providing a digital audio signal;

identifying a target transfer function to be applied to the signal;

generating at least one coefficient of an FIR filter, the generating including performing Heyser spiral curve fitting, and fitting a three-dimensional curve based on a magnitude and phase of the target transfer function;

filtering the digital audio signal through the FIR filter; and

inputting the filtered signal into the loudspeaker.

2. The method of claim 1, wherein the generating step includes:

selecting a cutoff frequency below which the FIR filter is to be corrected; and

fitting magnitude and phase of a frequency response of the filter by separate low-order polynomial functions below the selected cutoff frequency, the magnitude and phase of the frequency response of the filter converging to a calculated transfer function above the selected cutoff frequency.

3. The method of claim 2, wherein the filter is defined by a continuous function throughout a transition from below the cutoff frequency to above the cutoff frequency.

4. The method of claim 2, wherein the Heyser spiral curve fitting includes smoothing the target transfer function at frequencies below the cutoff frequency.

5. The method of claim 2, wherein the selecting step includes determining a frequency at which there is a difference between a target transfer function value and an actual transfer function value for the FIR filter that exceeds a threshold value.

6. The method of claim 5, wherein the determining step includes ascertaining a frequency at an upper end of a range of frequencies below the cutoff frequency in which the difference between the target transfer function value and the actual transfer function value for the FIR filter exceeds the threshold value.

7. The method of claim 1, wherein the identifying step includes acquiring information regarding:

local acoustic properties at a listening position within a three-dimensional sound field; and

a level of acoustic power within the three-dimensional sound field.

8. A method of operating a loudspeaker, the method including:

providing a digital audio signal;

selecting a frequency below which an FIR filter is to be corrected;

fitting magnitude and phase of a frequency response of the filter by separate low-order polynomial functions below the selected frequency, the magnitude and phase of the frequency response of the filter converging to a calculated transfer function above the selected frequency;

filtering the digital audio signal through the FIR filter; and

inputting the filtered signal into the loudspeaker.

9. The method of claim 8, comprising the further step of identifying a target transfer function to be applied to the signal, the providing step including generating at least one coefficient of the filter, the generating including performing Heyser spiral curve fitting, and fitting a three-dimensional curve based on a magnitude and phase of a target transfer function.

10. The method of claim 9, wherein the generating step includes selecting a cutoff frequency below which the FIR filter is to be corrected, the filter being defined by a smooth transition between the first polynomial and the second polynomial in the region of the cutoff frequency.

11. The method of claim 10, wherein the Heyser spiral curve fitting includes smoothing the target transfer function at frequencies below the cutoff frequency.

12. The method of claim 9 wherein the identifying step includes acquiring information regarding:

local acoustic properties at a listening position within a three-dimensional sound field; and

a level of acoustic power within the three-dimensional sound field.

13. The method of claim 8, wherein the filter is defined by a continuous function throughout a transition from below the selected frequency to above the selected frequency.

14. The method of claim 8, wherein the selecting step includes determining a frequency at which there is a difference between a target transfer function value and an actual transfer function value for the FIR filter that exceeds a threshold value.

15. The method of claim 14, wherein the determining step includes ascertaining a frequency at an upper end of a range of frequencies in which the difference between the target transfer function value and the actual transfer function value for the FIR filter exceeds the threshold value.

16. A loudspeaker arrangement comprising:

a loudspeaker having an input; and

a digital FIR filter having an input and an output, the input of the filter being configured to receive an audio signal, the output being coupled to the input of the loudspeaker, coefficients of the filter having been created using a Heyser spiral correction method to correct or improve low frequency response of the filter, the frequency response of the FIR filter being defined by a multi-dimensional polynomial at low frequencies and being calculated by conjugation to a target at high frequencies.

17. The arrangement of claim 16, wherein the filter is defined by a smooth transition between the low frequencies and the high frequencies.

18. The arrangement of claim 16, wherein the filter is defined by a continuous function throughout a transition between the low frequencies and the high frequencies.

19. The arrangement of claim 16, further comprising a digital signal processor connected to the input of the filter and configured to transmit the audio signal.

20. The arrangement of claim 16, wherein at least one said coefficient of the filter is calculated by