Adaptive frequency-domain reference noise canceller for multicarrier communications systems

• US 8,605,837 B2
• Filed: 10/10/2008
• Issued: 12/10/2013
• Est. Priority Date: 10/10/2008
• Status: Active Grant

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Broadcom Corporation (Broadcom, Inc.)
• Inventors: Wiese, Brian R., Desjardins, Philip, Iyer, Jayasuryan
• Primary Examiner(s): PHU, PHUONG M
• Application Number:
  US12/287,577
• Publication Number:
  US 20100091827A1
• Time in Patent Office: 1,887 Days
• Field of Search: 375/346, 375/148, 375/231, 375/232, 375/284, 375/285, 375/261, 375/264, 375/260, 375/343, 455/501, 455/63.1, 455/114.2, 455/296
• US Class Current: 375/346
• CPC Class Codes: H04B 1/10 Means associated with receiver for limiting or suppressing noise or interference
• CPC Class Codes: H04L 25/022 of frequency response
• CPC Class Codes: H04L 25/0226 sounding signals per se
• CPC Class Codes: H04L 5/00 Arrangements affording multiple use of the transmission path

Abstract

A method and apparatus to align data blocks in a data signal and a reference signal to increase cross-correlation between the data signal and the reference signal as compared to the unaligned data and reference signals and cancel interference in the data signal in the frequency-domain under changing conditions and in the presence of the data signal.

Claims

1. A method, comprising:

receiving, by a receiver device, a data signal and a reference signal;

aligning, by the receiver device, data blocks in the data signal with data blocks in the reference signal to increase cross-correlation between an aligned data signal and an aligned reference signal as compared to cross-correlation between the data signal and the reference signal; and

cancelling, by the receiver device and in a frequency domain, interference in the aligned data signal.

2. The method of claim 1, wherein aligning the data blocks in the data signal with the data blocks in the reference signal comprises:

introducing a first delay into the data signal; and

introducing a second delay into the reference signal.

3. The method of claim 2, wherein introducing the first delay comprises:

programming a first programmable delay First in, First out (FIFO) buffer, and wherein introducing the second delay comprises;

programming a second programmable delay FIFO buffer.

4. The method of claim 2, wherein introducing the first delay and the second delay comprises:

adjusting a group delay of a first programmable finite impulse response (FIR) filter and a second programmable FIR filter.

5. The method of claim 3, further comprising:

generating an interference estimate by multiplying the aligned reference signal by a single complex tap for each of one or more frequency bins of the reference signal; and

subtracting the interference estimate from the aligned data signal.

6. An apparatus, comprising:

a block aligner configured to align data blocks in a data signal received from a data line with data blocks in a reference signal received from a reference line to increase cross-correlation between the data signal and the reference signal as compared to unaligned data and reference signals; and

a tone-by-tone canceller configured to cancel, in a frequency domain, interference in the aligned data signal.

7. The apparatus of claim 6, wherein the block aligner comprises:

a first delay module configured to introduce a first delay into the data signal; and

a second delay module configured to introduce a second delay into the reference signal.

8. The apparatus of claim 7, wherein the first and second delay modules comprise:

a programmable delay First in, First out (FIFO) buffer.

9. The apparatus of claim 7, wherein the first and second delay modules comprise:

a programmable finite impulse response (FIR) filter.

10. A method, comprising:

generating an interference estimate for a data signal;

adaptively updating an interference canceller with the interference estimate; and

aligning blocks of the data signal with blocks of a reference signal to increase cross-correlation between an aligned data signal and an aligned reference signal as compared to cross-correlation between the data signal and the reference signal.

11. The method of claim 10, wherein generating the interference estimate comprises:

generating an error estimate from the aligned data signal;

adaptively updating canceller taps with the error estimate; and

multiplying the aligned reference signal by a single complex canceller tap for each of one or more frequency bins of the reference signal.

12. The method of claim 11, wherein generating the error estimate comprises:

subtracting an output of a constellation decision logic module from an input of the constellation decision logic module.

13. The method of claim 11, wherein adaptively updating the canceller taps comprises:

using the error estimate in a least-mean-square algorithm.

14. The method of claim 11, wherein adaptively updating the canceller taps comprises:

using the error estimate in a recursive least square algorithm.

15. The method of claim 10, wherein adaptively, updating the interference canceller comprises:

subtracting the interference estimate from the aligned data signal.

16. An apparatus, comprising:

a block aligner configured to align data blocks in a data signal and data blocks in a reference signal to increase cross-correlation between an aligned data signal and an aligned reference signal as compared to cross-correlation between the data signal and the reference signal; and

a tone-by-tone canceller operating in a frequency domain, configured to reduce interference in the aligned data signal, the tone-by-tone canceller comprising;

a constellation decision logic module configured to perform constellation decoding,

a first adder configured to subtract an output of the constellation decision logic module from an input of the constellation decision logic module to form an error estimate,

an interference estimator module configured to estimate interference in the aligned reference signal based upon the error estimate to provide an interference estimate, and

a second adder configured to subtract the interference estimate from the aligned data signal.

17. The apparatus of claim 16, wherein the constellation decision logic module comprises:

a trellis decoder.

18. The apparatus of claim 16, wherein the constellation decision logic module comprises:

an un-coded constellation decoder.

19. A method, comprising:

aligning blocks of a data signal with blocks of a reference signal to increase cross-correlation between an aligned data signal and an aligned reference signal as compared to cross-correlation between the data signal and the reference signal; and

cancelling, using a tone-by-tone cancelling device and in a frequency domain, interference in the aligned data signal.

20. The method of claim 19, further comprising:

dropping unneeded samples corresponding to a cyclic extension of the data signal from the aligned data signal and the aligned reference signal.

21. The method of claim 20, further comprising:

transforming the aligned data signal and the aligned reference signal from a representation in a time domain to a representation in the frequency domain.

22. The method of claim 21, further comprising:

normalizing a power of one or more frequency bins of the transformed data signal and the transformed reference signal.

23. The method of claim 22, further comprising:

generating an interference estimate by multiplying the transformed reference signal by a single complex tap for each of the one or more frequency bins of the reference signal.

24. The method of claim 23, wherein cancelling the interference in the aligned data signal comprises subtracting the interference estimate from the transformed data signal.

25. An apparatus, comprising:

a block aligner configured to align data blocks in a data signal and data blocks in a reference signal to increase cross-correlation between an aligned data signal and an aligned reference signal as compared to cross-correlation between the data signal and the reference signal; and

a tone-by-tone canceller, operating in a frequency domain, configured to cancel interference in the aligned data signal.

26. The apparatus of claim 25, wherein the block aligner comprises:

a first delay module configured to introduce a first delay into the data signal; and

a second delay module configured to introduce a second delay into the reference signal.

27. The apparatus of claim 26, wherein the tone-by-tone canceller comprises:

a constellation decision logic module configured to perform constellation decoding;

a first adder configured to subtract an output of the constellation decision logic module from an input of the constellation decision logic module to form an error estimate;

an interference estimator module configured to estimate interference in the aligned reference signal based upon the error estimate to provide an interference estimate; and

a second adder configured to subtract the interference estimate from the data signal.

28. The apparatus of claim 27, further comprising:

a cyclic extension removal and serial-to-parallel module configured to remove a number of redundant samples from the aligned data signal; and

a sample drop and serial-to-parallel module configured to remove a same number of samples from the aligned reference signal as were removed from the aligned data signal.

29. The apparatus of claim 28, further comprising:

a first fast Fourier transform (FFT) module configured to transform the aligned data signal from a representation in a time domain to a representation in the frequency domain; and

a second FFT module configured to transform the aligned reference signal from a second representation in the time domain to a second representation in the frequency domain.

30. The apparatus of claim 29, further comprising:

a channel flattening frequency domain equalizer (CF-FEQ) module configured to receive an output of the first FFT module and provide an output to the second adder;

a gain inversion and constellation scaling (GICS) module configured to receive an output of the second adder; and

a bin power normalization module configured to receive an output of the second FFT module.

31. The apparatus of claim 30, wherein the GICS module is further configured to decouple on-line reconfiguration (OLR), seamless rate adaptation (SRA) and bitswap from the tone-by-tone canceller.

32. An apparatus, comprising:

means for receiving a data signal and a reference signal;

means for aligning data blocks in the data signal with data blocks in the reference signal to maximize cross-correlation between the data signal and the reference signal; and

means for cancelling interference in the data signal in a frequency domain.

33. The apparatus of claim 32, wherein the means for receiving comprises:

means for performing time domain filtering of the data signal and the reference signal.

34. The apparatus of claim 32, further comprising:

means for collecting time domain samples of the data signal and the reference signal.

35. The apparatus of claim 34, further comprising:

means for performing cross-correlation of the data signal and the reference signal and computing a peak offset value of the time domain samples.

36. The apparatus of claim 35, further comprising:

means for introducing a first delay into the data signal and a second delay into the reference signal.

37. The apparatus of claim 36, further comprising:

means for generating an interference estimate for the data signal; and

means for adaptively updating an interference canceller with the interference estimate during data transmission.