PREVENTION OF ANC INSTABILITY IN THE PRESENCE OF LOW FREQUENCY NOISE

• US 20140072135A1
• Filed: 12/10/2012
• Published: 03/13/2014
• Est. Priority Date: 09/10/2012
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Inc.
• Inventors: Jensen, Thomas M., Bajic, Vladan
• Granted Patent: US 9,129,586 B2
• Time in Patent Office: Days
• US Class Current: 381/71.11
• CPC Class Codes: G10K 11/002 Devices for damping, suppressing, obstructing or conducting sound in acoustic devices G10K1/06 - G10K1/10 take precedence; for electro-mechanical transducers for communication H04R3/002
• CPC Class Codes: G10K 11/17823 Reference signals, e.g. ambient acoustic environment
• CPC Class Codes: G10K 11/17825 Error signals
• CPC Class Codes: G10K 11/17835 using detection of abnormal input signals
• CPC Class Codes: G10K 11/17853 of the filter
• CPC Class Codes: G10K 11/17854 the filter being an adaptive filter
• CPC Class Codes: G10K 11/17881 the reference signal being an acoustic signal, e.g. recorded with a microphone
• CPC Class Codes: G10K 11/17885 additionally using a desired external signal, e.g. pass-through audio such as music or speech
• CPC Class Codes: G10K 2210/1081 Earphones, e.g. for telephones, ear protectors or headsets
• CPC Class Codes: G10K 2210/30231 Sources, e.g. identifying noisy processes or components
• CPC Class Codes: G10K 2210/3028 Filtering, e.g. Kalman filters or special analogue or digital filters
• CPC Class Codes: G10K 2210/511 Narrow band, e.g. implementations for single frequency cancellation

Abstract

An active noise control (ANC) processor has an adaptive filter that uses a reference signal to produce an anti-noise signal, and an error signal to evaluate cancellation performance. An adaptive filter algorithm engine configures the filter coefficients of the adaptive filter, in accordance with pre-shaped versions of the error and reference signals. The pre-shaping filter has a high-pass transfer function and enables the adaptive algorithm engine to increase noise cancellation performance in a high frequency band during the presence of focused or narrow-band noise energy in a low frequency band. Other embodiments are also described and claimed.

Claims

1. An audio apparatus comprising:

a reference microphone to produce a reference signal;

an error microphone to produce an error signal; and

an active noise control (ANC) processor having

an adaptive filter to use the reference signal and produce an anti-noise signal,

an adaptive filter algorithm engine to configure digital filter coefficients of the adaptive filter in accordance with signals at a first input and a second input,

a first pre-shaping filter to filter the reference signal at the first input,

a second pre-shaping filter to filter the error signal at the second input,

wherein each of the first and second pre-shaping filters has a high pass transfer function that is designed to suppress energy in a low frequency band, relative to a high frequency band, wherein the first and second pre-shaping filters together enable the engine to adapt the adaptive filter to thereby produce substantial anti-noise in the high frequency band during the presence of focused or narrowband noise energy in the low frequency band.

2. The audio apparatus of claim 1 wherein the low frequency band is 5 Hz to 400 Hz, and the high frequency band is 400 Hz to 1 kHz.

3. The audio apparatus of claim 1 wherein the focused noise energy is one of the group consisting of car noise, as picked up by the reference mic while inside a car that has a running combustion engine and that is being driven, bus noise, train noise, and any other noise environment that has a focused or emphasized low frequency content.

4. The audio apparatus of claim 1 wherein the high pass transfer function has a knee between 100 Hz and 400 Hz.

5. The audio apparatus of claim 4 wherein the high pass transfer function has a roll-off slope between 10 dB/octave and 25 dB/octave.

6. The audio apparatus of claim 1 wherein without the first and second pre-shaping filters, the ANC processor is substantially less effective across 400 Hz to 2,000 Hz, in the presence of focused noise energy that lies below 400 Hz, than with the first and second pre-shaping filters.

7. The audio apparatus of claim 1 wherein the low frequency band is defined as the smallest frequency band that contains substantially all of the energy of passenger car noise as it is picked up by the reference microphone inside a driven passenger car.

8. The audio apparatus of claim 1 wherein the ANC processor further comprises:

a first high pass filter having a knee at less than 200 Hz that is to filter the reference signal, upstream of the first pre-shaping filter; and

a second high pass filter having a knee at less than 200 Hz that is to filter the error signal, upstream of the second pre-shaping filter.

9. The audio apparatus of claim 8 wherein the ANC processor further comprises a third high pass filter having a knee no higher than 4 Hz, and wherein the output of the third high pass filter feeds both the first high pass filter and an input of the adaptive filter.

10. The audio apparatus of claim 1 wherein the first and second pre-shaping filters are one of a) infinite impulse response (IIR) filters and b) finite impulse response (FIR) filters.

11. A method for active noise control (ANC), comprising:

pre-shaping a reference microphone signal in accordance with a transfer function with a high-pass section having a knee between 100 Hz and 500 Hz;

pre-shaping an error microphone signal in accordance with a transfer function with a high-pass section having a knee between 100 Hz and 500 Hz; and

performing an adaptive filter algorithm to configure digital filter coefficients of an adaptive filter in accordance with the pre-shaped ref mic signal and the pre-shaped error mic signal.

12. The method of claim 11 wherein each of the transfer functions has a roll off slope between 10 dB/octave and 25 dB/octave.

13. The method of claim 11 further comprising high pass filtering the ref mic signal in accordance with a transfer function having a knee below 4 Hz, and using the 4 Hz filtered ref mic signal as an input signal of the adaptive filter.

14. The method of claim 13 wherein the adaptive filter models a primary acoustic path, the method further comprising:

high pass filtering the ref mic signal in accordance with a transfer function having a knee below 200 Hz, and using the 200 Hz filtered ref mic signal at an input of a further adaptive filter that models a secondary acoustic path.

15. A mobile phone comprising:

a mobile phone handset housing having therein

an earpiece speaker;

an audio source to produce an audio user content signal;

a reference microphone to produce a reference signal;

an error microphone positioned closer to the earpiece speaker than the reference microphone to produce an error signal; and

an active noise control (ANC) processor having an adaptive filter to use the reference signal to produce an anti-noise signal that is combined with the audio user content signal to drive the earpiece speaker, a first pre-shaping filter to filter the reference signal, a second pre-shaping filter to filter the error signal, an adaptive filter algorithm engine to configure filter coefficients of the adaptive filter in accordance with the filtered reference and error signals, wherein each of the first and second pre-shaping filters has a high pass transfer function that is designed to suppress energy in a low frequency band, relative to a high frequency band, to prevent instability of the ANC processor in the high frequency band during the presence of narrowband noise energy in the low frequency band.

16. The mobile phone device of claim 15 wherein the low frequency band is 5 Hz to 400 Hz, and the high frequency band is 400 Hz to 1 kHz.

17. The mobile phone device of claim 15 wherein the high pass transfer function has a knee between 100 Hz and 400 Hz.

18. The mobile phone device of claim 17 wherein the high pass transfer function has a roll-off slope between 10 dB/octave and 25 dB/octave.