Dual-processing interference cancelling system and method

• US 6,178,248 B1
• Filed: 04/14/1997
• Issued: 01/23/2001
• Est. Priority Date: 04/14/1997
• Status: Expired due to Term

• Current Assignee: Andrea Electronics Corporation
• Original Assignee: Andrea Electronics Corporation
• Inventors: Marash, Joseph
• Primary Examiner(s): CHIN, VIVIAN C
• Application Number:
  US08/840,159
• Time in Patent Office: 1,380 Days
• Field of Search: 381/94.1, 381/94.2, 381/94.3, 381/94, 381/7, 381/92, 381/71.1, 381/71.14, 381/66, 381/71.11, 381/71.6, 381/71.12
• US Class Current: 381/94.7
• CPC Class Codes: H03H 21/0012 Digital adaptive filters
• CPC Class Codes: H04R 25/407 Circuits for combining signals of a plurality of transducers
• CPC Class Codes: H04R 3/005 for combining the signals of two or more microphones specially adapted for hearing aids H04R25/407

Abstract

A dual-processing interference cancelling system and method for processing a broadband input in a computationally efficient manner. Dual processing divides the input into higher and lower frequency bands and applies adaptive filter processing to the lower frequency band while applying non-adaptive filter processing to the higher frequency band. Various embodiments are shown including those based on sub-bands, broadband processing with band-limited adaptation, and broadband processing with an external main-channel generator.

Claims

1. A dual-processing interference cancelling system for processing an input containing a target signal originating from a target source and interferences originating from interference sources and for producing an output representing the target signal with substantially reduced interferences, comprising:

a main-channel generator capable of receiving signals from such input and for generating therefrom a main channel representing signals received from the target source and having a target signal component and an interference component;

a first splitter, connected to the main-channel generator, for splitting the main channel into lower and upper sub-bands, wherein the lower and upper sub-bands together form the entire main channel;

a reference-channel generator capable of receiving signals from such input and for generating therefrom one or more reference channels representing signals received from the interference sources;

a second splitter, connected to the reference-channel generator, for splitting said one or more reference channels into lower and upper sub-bands, wherein the lower and upper sub-bands for each reference channel together form the entire reference channel;

an adaptive filter, having filter weights, connected to receive the lower sub-bands of said one or more reference channels for generating one or more cancelling signals approximating an interference component of the lower sub-band of the main channel;

a subtracter, connected to the first splitter and the adaptive filter, for generating an output by subtracting said one or more cancelling signals from the lower sub-band of the main channel;

the adaptive filter also being connected to receive the output from the subtracter and said system including filter-weight-updating means for determining updated filter weight values for the adaptive filter such that the differences between the lower sub-band of the main channel and the cancelling signals are substantially minimized; and

a reconstructor, connected to the subtracter and to the first splitter, for reconstructing a broadband output by combining the upper sub-band of the main channel and the output from the subtracter.

2. The system of claim 1, further comprising a first set of one or more sensors for receiving signals from a target source and a second set of one or more sensors for receiving interferences.

3. The system of claim 2, wherein said the sensors in the first set and the second set are microphones.

4. The system of claim 2, wherein one or more sensors of the second set are accelerometers for sensing vibration of a surrounding structure.

5. The system of claim 1, further comprising one or more sensors for receiving signals from a target source and also for receiving signals from interferences sources.

6. The system of claim 1, wherein said main-channel generator is a main-channel matrix which generates a main channel from an array of sensors, the main channel representing signals received in the direction of the target.

7. The system of claim 1, wherein said reference-channel generator is a reference-channel matrix which generates reference channels from an array of sensors, the reference channels representing signals received in the directions other than the direction of the target.

8. The system of claim 1, wherein said adaptive filter comprises a finite-impulse-response filter for generating said one or more cancelling signals.

9. The system of claim 1, wherein said adaptive filter comprises an infinite-impulse-response filter for generating said one or more cancelling signals.

10. The system of claim 1, wherein said filter-weight-updating means uses the least-mean-square algorithm where the mean-square values of the differences between the lower sub-band of the main channel and said one or more cancelling signals are substantially minimized.

11. A dual-processing interference cancelling system for processing an input containing a target signal as well as interferences and for producing an output representing the target signal with substantially reduced interferences, comprising:

a main-channel matrix capable of receiving signals from such input and for generating therefrom a main channel representing signals received from the direction of the target source and having a target signal component and an interference component;

a first splitter, connected to the main-channel matrix, for splitting the main channel into lower and upper sub-bands, wherein the lower and upper sub-bands together form the entire main channel;

a reference-channel matrix capable of receiving signals from signal sources and for generating therefrom reference channels representing signals received from the directions other than the direction of the target source;

a second splitter, connected to the reference-channel matrix, for splitting the reference channels into lower and upper sub-bands, wherein the lower and upper sub-bands of each reference channel together form the entire reference channel;

an adaptive filter, having filter weights, connected to receive the lower sub-band reference channels for generating cancelling signals approximating an interference component of the lower sub-band main channel;

a subtracter, connected to the main channel splitter and the adaptive filter, for generating a sub-band output by subtracting the cancelling signals from the lower sub-band main channel;

the adaptive filter also being connected to receive the output from the subtracter and said system including filter weight updating means for determining updated filter weight values for the adaptive filter such that the differences between the lower sub-band of the main channel and the cancelling signals are substantially minimized; and

a reconstructor, connected to the subtracter and to the main-channel matrix, for reconstructing a broadband output by combining the upper sub-band of the main channel and the sub-band output from the subtracter.

12. The system of claim 11, wherein said adaptive filter comprises a finite-impulse-response filter for generating the cancelling signals.

13. The system of claim 11, wherein said adaptive filter comprises an infinite-impulse-response filter for generating the cancelling signals.

14. The system of claim 11, wherein said filter-weight-updating means uses the least-mean-square algorithm.

15. The system of claim 11, further comprising:

a sensor array of spatially distributed sensors, each for receiving the target signal and interferences;

a sampling unit, connected to receive signals from the sensor array, for converting such signals to digital form and for sending them to the main-channel matrix and the reference-channel matrix; and

an output digital-to-analog converter, connected to the subtracter, for converting said broadband output to analog form.

16. The system of claim 15, wherein the sensors are microphones.

17. The system of claim 15 wherein the first splitter comprises a down-sampler.

18. The system of claim 15 wherein the second splitter comprises a down-sampler.

19. The system of claim 15, wherein the reconstructor comprises an interpolator.

20. A method for processing an input containing a target signal originating from a target source and interferences originating from interference sources and for producing an output representing the target signal with substantially reduced interferences, comprising the steps of:

generating a broadband main channel from such input, the broadband main channel representing signals received from the target source and having a target signal component and an interference component;

splitting the broadband main channel into lower and higher sub-band main channels;

generating one or more broadband reference channels representing signals received from interference sources;

splitting said one or more broadband reference channels into one or more lower and upper sub-band reference channels;

generating one or more lower sub-band cancelling signal approximating the interference component in the main channel by filtering said one or more lower sub-band reference channels using an adaptive filter;

generating a lower sub-band output by subtracting said one or more lower sub-band cancelling signals from the lower sub-band main channel;

reconstructing a broadband output by combining the lower sub-band output and the upper sub-band main channel; and

adaptively adjusting filter weights of the adaptive filter so that the differences between the lower sub-band main channel and said one or more lower sub-band cancelling signals are substantially minimized.

21. The method of claim 20, wherein the step of generating said one or more lower sub-band cancelling signals uses a finite-impulse-response filter.

22. The method of claim 20, wherein the step of generating said one or more lower sub-band cancelling signals uses an infinite-impulse-response filter.

23. The method of claim 20, wherein the step of adaptively updating filter weights uses the least-mean-square algorithm.

24. The method of claim 20, further comprising the steps of:

generating input signals from a sensor array of spatially distributed sensors, each for receiving the target signal and interferences;

sampling the inputs signal and converting them to digital form before the step of generating a broadband main channel and the step of generating one or more broadband reference channels; and

converting the broadband output to analog form after the combining step.

25. The method of claim 24, wherein the sensors are microphones.