Communication system noise cancellation power signal calculation techniques

• US 7,957,965 B2
• Filed: 08/07/2008
• Issued: 06/07/2011
• Est. Priority Date: 03/28/2000
• Status: Expired due to Fees

• Current Assignee: Tellabs Operations Incorporated (Infinera Corp.)
• Original Assignee: Tellabs Operations Incorporated (Infinera Corp.)
• Inventors: Chandran, Ravi, Dunne, Bruce E., Marchok, Daniel J.
• Primary Examiner(s): Jackson, Jakieda R
• Application Number:
  US12/187,581
• Publication Number:
  US 20090024387A1
• Time in Patent Office: 1,034 Days
• Field of Search: 704/226, 704/205
• US Class Current: 704/226
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 21/02 Speech enhancement, e.g. noise reduction or echo cancellation reducing echo effects in line transmission systems H04B3/20; echo suppression in hands-free telephones H04M9/08
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 21/0232 Processing in the frequency domain

Abstract

In order to enhance the quality of a communication signal derived from speech and noise, a filter divides the communication signal into a plurality of frequency band signals. A calculator generates a plurality of power band signals each having a power band value and corresponding to one of the frequency band signals. The power band values are based on estimating, over a time period, the power of one of the frequency band signals. The time period is different for different ones of the frequency band signals. The power band values are used to calculate weighting factors which are used to alter the frequency band signals that are combined to generate an improved communication signal.

Claims

1. A method for enhancing quality of a communication signal in a communication system, the communication signal including a speech component and a noise component, the method comprising:

dividing the communication signal into a plurality of frequency band signals in the communication system;

calculating weighting values as a function of a weighting curve, the weighting curve approximating inverse proportionality to a spectrum shape of the speech component within the plurality of frequency band signals;

altering the frequency band signals in response to the weighting values to generate a plurality of weighted frequency band signals; and

combining the weighted frequency band signals to generate a communication signal with enhanced quality.

2. The method of claim 1 further including calculating the weighting values in response to a control signal and generating the control signal as a function of at least an approximation of the spectrum shape within the plurality of frequency band signals.

3. The method of claim 2 further including deriving the control signal as a function of generating power signals derived from power of the plurality of the frequency band signals.

4. The method of claim 2 further including assigning the weighting values within the plurality of frequency band signals to reduce variations of the power signals across the plurality of frequency band signals.

5. The method of claim 1 further including calculating the weighting values in response to a control signal and generating the control signal as a function at least in part from a pitch of the speech component.

6. The method of claim 1 wherein the weighting values vary monotonically from a first value at a first frequency to a second value at a second higher frequency by at least a factor of two.

7. The method of claim 6 wherein the weighting values vary monotonically from the second value to a third value between the first value and second value at a frequency greater than the second frequency.

8. The method of claim 1 further including assigning a weighting value resulting in a minimum suppression to one of the frequency band signals depending on whether the speech component results from voiced speech or unvoiced speech.

9. The method of claim 8 further including assigning a weighting value resulting in a minimum suppression to one of the frequency band signals as a function of a ratio of properties of the speech component and the noise component within the plurality of frequency band signals.

10. The method of claim 1 further comprising obtaining the weighting values from a table of weighting values.

11. An apparatus to enhance quality of a communication signal including a speech component and a noise component, the apparatus comprising:

a division module to divide the communication signal into a plurality of frequency band signals;

a calculation module to calculate weighting values as a function of a weighting curve, the weighting curve approximating inverse proportionality to a spectrum shape of the speech component within the plurality of frequency band signals;

an alteration module to alter the frequency band signals in response to the weighting values to generate a plurality of weighted frequency band signals; and

a combination module to combine the weighted frequency band signals to generate a communication signal with enhanced quality.

12. The apparatus of claim 11 further including a second calculation module arranged to calculate the weighting values in response to a control signal and generate the control signal as a function of at least an approximation of the spectrum shape within the plurality of frequency band signals.

13. The apparatus of claim 12 further including a module arranged to derive the control signal as a function of generating power signals derived from power of the plurality of the frequency band signals.

14. The apparatus of claim 12 further including an assignment module arranged to assign the weighting values within the plurality of frequency band signals to reduce variations of the power signals across the plurality of frequency band signals.

15. The apparatus of claim 11 further including a second calculation module arranged to calculate the weighting values in response to a control signal and generate the control signal as a function at least in part from a pitch of the speech component.

16. The apparatus of claim 11 wherein the weighting values vary monotonically from a first value at a first frequency to a second value at a second higher frequency by at least a factor of two.

17. The apparatus of claim 16 wherein the weighting values vary monotonically from the second value to a third value between the first value and second value at a frequency greater than the second frequency.

18. The apparatus of claim 11 further including an assignment module to assign a weighting value resulting in a minimum suppression to one of the frequency band signals depending on whether the speech component results from voiced speech or unvoiced speech.

19. The apparatus of claim 18 wherein the assignment module is arranged to assign a weighting value resulting in a minimum suppression to one of the frequency band signals as a function of a ratio of properties of the speech component and the noise component within the plurality of frequency band signals.

20. The apparatus of claim 11 wherein the calculation module is arranged to obtain the weighting values from a table of weighting values.