Method for whitening colored noise in a communication system

• US 20020126778A1
• Filed: 07/02/2001
• Published: 09/12/2002
• Est. Priority Date: 01/16/2001
• Status: Active Grant

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.)
• Inventors: Ojard, Eric, Bagchi, Amit Goutam
• Granted Patent: US 6,826,242 B2
• Time in Patent Office: Days
• US Class Current: 375/346
• CPC Class Codes: H04B 1/1036 with automatic suppression of narrow band noise or interference, e.g. by using tuneable notch filters H04B1/123 takes precedence; filter circuits H03H
• CPC Class Codes: H04B 1/123 using adaptive balancing or compensation means adaptive filter circuits and algorithms H03H
• CPC Class Codes: H04L 1/0001 Systems modifying transmission characteristics according to link quality, e.g. power backoff adaptive data allocation for multicarrier modulation H04L5/0044; controlling transmission power for radio systems H04W52/04
• CPC Class Codes: H04L 25/03006 Arrangements for removing intersymbol interference

Abstract

A filter settings generation operation includes sampling colored noise present at the input of a receiver to produce a sampled signal. The sampled signal is spectrally characterized across a frequency band of interest to produce a spectral characterization of the sampled signal. This spectral characterization may not include a signal of interest. The spectral characterization is then modified to produce a modified spectral characterization. Filter settings are then generated based upon the modified spectral characterization. Finally, the input present at the receiver is filtered using the filter settings when the signal of interest is present to whiten colored noise that is present with the signal of interest. In modifying the spectral characterization, pluralities of spectral components of the spectral characterization are independently modified to produce the modified spectral characterization. Modifications to the spectral characterization may be performed in the frequency domain and/or the time domain. Particular modifications include amplifying spectral components, weighting spectral components based upon prior spectral components, and averaging spectral components with prior spectral components.

Claims

1. In a communication network, a method for whitening colored noise present at the input of a receiver, the method comprising:

sampling colored noise present at the input of the receiver during a sampling interval to produce a sampled signal;

spectrally characterizing the sampled signal across a frequency band of interest to produce a spectral characterization of the sampled signal;

modifying the spectral characterization of the sampled signal to produce a modified spectral characterization;

generating filter settings based upon the modified spectral characterization; and

when a signal of interest and colored noise are present at the input of the receiver, filtering the input using the filter settings to whiten the colored noise at the input.

2. The method of claim 1, wherein in modifying the spectral characterization of the sampled signal to produce the modified spectral characterization, a plurality of spectral components of the spectral characterization are modified.

3. The method of claim 2, wherein in modifying the spectral characterization of the sampled signal to produce the modified spectral characterization, each of the plurality of spectral components of the spectral characterization is amplified by one of a plurality of gain factors.

4. The method of claim 3, wherein a gain factor to be applied to a particular spectral component is based upon the magnitude of the spectral component.

5. The method of claim 4, wherein the gain factor to be applied to the particular spectral component is also based upon the magnitudes of other of the spectral components of the spectral characterization.

6. The method of claim 4, wherein the gain factor to be applied to the particular spectral component is also based upon a characterization of the noise power of the spectral characterization.

7. The method of claim 1, wherein in modifying the spectral characterization of the sampled signal to produce the modified spectral characterization, the modified spectral characterization is based upon the spectral characterization and at least one of:

a characterization of the noise power of the spectral characterization;

a prior spectral characterization; and

a characterization of the noise power of the prior spectral characterization.

8. The method of claim 1, wherein:

the modified spectral characterization is based upon the spectral characterization and prior spectral characterizations; and

wherein in modifying the spectral characterization, spectral components of the spectral characterization are averaged with corresponding spectral components of prior spectral characterizations.

9. The method of claim 8, wherein:

a characterization of the noise power of the spectral characterization and characterizations of the noise power of the prior spectral characterizations; and

when the characterization of the noise power of the spectral characterization is above average as compared to characterizations of the noise power of prior spectral characterizations, the spectral components of the spectral characterization are de-weighted in determining the spectral components of the modified spectral characterization.

10. The method of claim 1, wherein in modifying the spectral characterization of the sampled signal to produce a modified spectral characterization, the spectral characterization is folded to decrease the bandwidth of the spectral characterization.

11. The method of claim 10, further comprising modifying the filter settings to compensate for the folding of the spectral characterization.

12. The method of claim 1, wherein the sampling is performed when the signal of interest is not present at the input of the receiver.

13. The method of claim 1, wherein:

the sampling is performed while the signal of interest is present at the input of the receiver; and

the method further comprises removing the signal of interest from the sampled signal.

14. The method of claim 1, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed in the frequency domain.

15. The method of claim 1, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed in the time domain.

16. The method of claim 1, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed both in the frequency domain and in the time domain.

17. The method of claim 1, further comprising discarding the spectral characterization when a characterization of the noise power of the spectral characterization exceeds a threshold.

18. The method of claim 1, wherein the network comprises a home phone line network.

20. The method of claim 19, wherein a gain factor to be applied to a particular spectral component is based upon the magnitude of the spectral component.

21. The method of claim 20, wherein the gain factor to be applied to the particular spectral component is also based upon the magnitudes of other of the spectral components of the spectral characterization.

22. The method of claim 20, wherein the gain factor to be applied to the particular spectral component is also based upon a characterization of the noise power of the spectral characterization.

23. The method of claim 19, wherein in modifying the spectral characterization of the sampled signal to produce the modified spectral characterization, the modified spectral characterization is also based upon a characterization of the noise power of the spectral characterization.

24. The method of claim 23, wherein:

the modified spectral characterization is also based upon characterizations of the noise power of prior spectral characterizations; and

when the characterization of the noise power of the spectral characterization is above average as compared to the characterizations of the noise power of prior spectral characterizations, the spectral components of the spectral characterization are de-weighted in determining the spectral components of the modified spectral characterization.

25. The method of claim 19, wherein in modifying the spectral characterization of the sampled signal to produce a modified spectral characterization, the spectral characterization is folded to decrease the bandwidth of the spectral characterization.

26. The method of claim 25, further comprising modifying the filter settings to compensate for the folding of the spectral characterization.

27. The method of claim 19, wherein the sampling is performed when the signal of interest is not present at the input of the receiver.

28. The method of claim 19, wherein:

the sampling is performed while the signal of interest is present at the input of the receiver; and

the method further comprises removing the signal of interest from the sampled signal.

29. The method of claim 19, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed in the frequency domain.

30. The method of claim 19, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed in the time domain.

31. The method of claim 19, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed both in the frequency domain and in the time domain.

32. The method of claim 19, further comprising discarding the spectral characterization when a characterization of the noise power of the spectral characterization exceeds a threshold.

33. The method of claim 19, wherein the network comprises a home phone line network.

35. The method of claim 34, wherein a gain factor to be applied to a particular spectral component is based upon the magnitude of the spectral component.

36. The method of claim 35, wherein the gain factor to be applied to the particular spectral component is also based upon the magnitudes of other of the spectral components of the spectral characterization.

37. The method of claim 35, wherein the gain factor to be applied to the particular spectral component is also based upon a characterization of the noise power of the spectral characterization.

38. The method of claim 34, wherein when the characterization of the noise power of the spectral characterization is above average as compared to the characterizations of the noise power of prior spectral characterizations, the spectral components of the spectral characterization are de-weighted in determining the spectral components of the modified spectral characterization.

39. The method of claim 34, wherein in modifying the spectral characterization of the sampled signal to produce a modified spectral characterization, the spectral characterization is folded to decrease the bandwidth of the spectral characterization.

40. The method of claim 39, further comprising modifying the filter settings to compensate for the folding of the spectral characterization.

41. The method of claim 34, wherein the sampling is performed when the signal of interest is not present at the input of the receiver.

42. The method of claim 34, wherein:

the sampling is performed while the signal of interest is present at the input of the receiver; and

the method further comprises removing the signal of interest from the sampled signal.

43. The method of claim 34, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed in the frequency domain.

44. The method of claim 34, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed in the time domain.

45. The method of claim 34, wherein modifying the spectral characterization of the sampled signal to produce a modified spectral characterization is performed both in the frequency domain and in the time domain.

46. The method of claim 34, further comprising discarding the spectral characterization when a characterization of the noise power of the spectral characterization exceeds a threshold.

47. The method of claim 34, wherein the network comprises a home phone line network.

19. In a communication network, a method for whitening colored noise present at the input of a receiver, the method comprising:

sampling colored noise present at the input of the receiver during a sampling interval to produce a sampled signal;

spectrally characterizing the sampled signal across a frequency band of interest to produce a spectral characterization of the sampled signal;

modifying the spectral characterization of the sampled signal to produce a modified spectral characterization by;

amplifying each of a plurality of spectral components of the spectral characterization by one of a plurality of gain factors; and

averaging each of the plurality of spectral components with corresponding spectral components of prior spectral characterizations;

generating filter settings based upon the modified spectral characterization; and

when a signal of interest and colored noise are present at the input of the receiver, filtering the input using the filter settings to whiten the colored noise at the input.

34. In a communication network, a method for whitening colored noise present at the input of a receiver, the method comprising:

sampling colored noise present at the input of the receiver during a sampling interval to produce a sampled signal;

spectrally characterizing the sampled signal across a frequency band of interest to produce a spectral characterization of the sampled signal;

determining a characterization of the noise power of the spectral characterization;

modifying the spectral characterization of the sampled signal to produce a modified spectral characterization by;

amplifying each of a plurality of spectral components of the spectral characterization by one of a plurality of gain factors;

weighting each of the plurality of spectral components based upon the spectral component, the characterization of the noise power of the spectral characterization, and characterizations of the noise power of prior spectral characterizations; and

averaging each of the plurality of spectral components with corresponding spectral components of prior spectral characterizations;

generating filter settings based upon the modified spectral characterization; and

when a signal of interest and colored noise are present at the input of the receiver, filtering the input using the filter settings to whiten the colored noise at the input.