Method and system for channel estimation in a data transmission system

• US 7,561,626 B2
• Filed: 06/09/2004
• Issued: 07/14/2009
• Est. Priority Date: 06/09/2004
• Status: Active Grant

• Current Assignee: Marvell Asia Pte Limited (Marvell Technology Group Limited)
• Original Assignee: Marvell World Trade Limited (Marvell Technology Group Limited)
• Inventors: Leung, Pak Hei Matthew, Nedic, Slobodan, Tang, Eugene Yuk-Yin, Chow, Jacky S., Luk, King-Fuk
• Primary Examiner(s): Payne; David C
• Assistant Examiner(s): Flores; Leon
• Application Number:
  US10/865,093
• Publication Number:
  US 20050276355A1
• Time in Patent Office: 1,861 Days
• Field of Search: 375/260, 375/340, 370/210, 370/480, 379/93.01
• US Class Current: 375/260
• CPC Class Codes: H04L 25/0204 of multiple channels
• CPC Class Codes: H04L 25/0212 of impulse response
• CPC Class Codes: H04L 25/0226 sounding signals per se
• CPC Class Codes: H04L 25/023 with extension to other symbols
• CPC Class Codes: H04L 25/025 using least-mean-square [LMS] method
• CPC Class Codes: H04L 27/2613 Structure of the reference signals
• CPC Class Codes: H04L 27/2647 Arrangements specific to the receiver only equalisation H04L27/01

Abstract

Improved techniques for estimating a channel (e.g., channel impulse response) in the time domain are disclosed. The improved techniques can be used to estimate time-domain channel taps in a digital communication system. An improved training sequence enables time-domain estimation and eliminates the need for correct frequency-domain channel estimation at every frequency. By utilizing the improved training sequence, the estimation of each of the time-domain channel taps can be decoupled from each other. This enables a channel estimate to be performed with not only a higher convergence speed but also lower complexity.

Claims

1. A method for determining a channel impulse response at a receiver, said method comprising:

delaying a received first block of samples at the receiver via a channel;

subsequently receiving a second block of samples at the receiver via the channel;

producing an estimated channel response of the channel using M channel taps;

producing an estimate of the delayed first block of samples based on the second block of samples received via the channel and the estimated channel response, the estimate having M components;

determining a difference estimate of the estimated delayed first block of samples and the delayed first block of samples received via the channel, the difference estimate having M components; and

updating the estimated channel response by individually updating one or more of the M channel taps based on the difference estimate.

2. A method as recited in claim 1, wherein the estimated first block is produced without any convolution operations.

3. A method as recited in claim 1, wherein the estimated first block is produced using addition and multiplication operations without involving any convolution operations.

4. A method as recited in claim 1, wherein the estimated first block is produced using in accordance with the following equation:

rec_—

sym₂(M . . . 2M −

1)+(p−

q)≅

hh(1 . . . M)

wherein rec_sym₂ is a received symbol, p and q are the different samples in an otherwise identical blocks of samples being transmitted, and hh is the channel response.

5. A method as recited in claim 1, wherein each of the M components of the difference estimate correspond to one of the M channel taps of the estimated channel response.

6. A method as recited in claim 1, wherein said updating comprises:

independently updating one of the M channel taps based on the corresponding one of the M components of the difference estimate.

7. A method as recited in claim 1, wherein said method is performed entirely in the time domain.

8. A method as recited in claim 1, wherein the first and second blocks of samples received at the receiver are respectively associated with first and second block of samples of N samples that were transmitted from a transmitter.

9. A method as recited in claim 8, wherein the estimated channel response is of length M, where N is equal to at least 2M.

10. A method as recited in claim 1, wherein the receiver is part of a multicarrier data transmission system.

11. A method as recited in claim 1, wherein the channel is at least one of a direct channel, an echo channel and a crosstalk channel.

12. A method as recited in claim 1,

wherein the first block of samples and the second block of samples result from a transmitter transmitting a first block and a second block of training samples to the receiver through the channel, and

wherein the second block of training samples is the same as the first block of training samples except that one of the training samples is different.

13. A method as recited in claim 1,

wherein the first block of samples and the second block of samples result from a transmitter transmitting a group of training samples to the receiver through the channel, and

wherein the training samples in the group are periodic except that one of the training samples is different.

14. A method as recited in claim 1, wherein said producing provides the estimate of the received samples in both a time domain and a frequency domain.

15. A method as recited in claim 1, wherein the training sequence is used to train-up a near-end channel and a far-end channel.

16. A method as recited in claim 15, wherein the training sequence is used to train-up two or more of direct, echo, far-end crosstalk and near-end crosstalk channels.

17. A method as recited in claim 1, wherein said method further comprises:

subsequently tracking the estimated channel response during showtime operation.

18. A method as recited in claim 17, wherein said tracking is performed in the frequency domain.

19. A method are recited in claim 18, wherein said tracking updates at most only a subset of the M channel taps.

20. A method as recited in claim 18, wherein said tracking converts frequency components of the tracked estimated channel response and other components in the frequency domain into values for one or more of the M channel taps in the time domain.

21. A computer readable medium encoded with computer program code for determining a channel impulse response at a receiver, said computer readable medium comprising:

computer program code for delaying a received first block of samples at the receiver via a channel;

computer program code for subsequently receiving a second block of samples at the receiver via the channel;

computer program code for producing an estimated channel response of the channel using M channel taps;

computer program code for producing an estimate of the delayed first block of samples based on the second block of samples received via the channel and the estimated channel response, the estimate having M components;

computer program code for determining a difference estimate of the estimated delayed first block of samples and the delayed first block of samples received via the channel, the difference estimate having M components; and

computer program code for updating the estimated channel response by individually updating one or more of the M channel taps based on the difference estimate.

22. A computer readable medium as recited in claim 21, wherein said computer program code for producing the estimate of the first block is achieved without any convolution operations.

23. A computer readable medium as recited in claim 22, wherein each of the M components of the difference estimate correspond to one of the M channel taps of the estimated channel response.

24. A computer readable medium as recited in claim 23, wherein the receiver is part of a multicarrier data transmission system.

25. A system for determining a channel impulse response, said system comprising:

receiver circuitry operative to;

delay a received first block of samples via a channel;

subsequently receive a second block of samples via the channel;

estimator circuitry operative to produce an estimated channel response of the channel using M channel taps;

multiplier circuitry operative to produce an estimate of the delayed first block of samples based on the second block of samples received via the channel and the estimated channel response, the estimate having M components;

adder circuitry operative to determine a difference estimate of the delayed estimated first block of samples and the delayed first block of samples received via the channel, the difference estimate having M components; and

controller circuitry operative to update the estimated channel response by individually updating one or more of the M channel taps based on the difference estimate.

26. The system as recited in claim 25, wherein the estimated first block is produced without any convolution operations.

27. The system as recited in claim 25, wherein the estimated first block is produced using addition and multiplication operations without involving any convolution operations.

28. The system as recited in claim 25, wherein the estimated first block is produced in accordance with the following equation:

rec_sym₂(M . . . 2M−

1)+(p−

q)≅

hh(1 . . . M)

wherein rec_sym₂ is a received symbol, p and q are the different samples in an otherwise identical blocks of samples being transmitted, and hh is the channel response.

29. The system as recited in claim 25, wherein each of the M components of the difference estimate correspond to one of the M channel taps of the estimated channel response.

30. The system as recited in claim 25, wherein said controller circuitry is further operative to update:

one of the M channel taps independently based on the corresponding one of the M components of the difference estimate.

31. The system as recited in claim 25, wherein said system is operative to perform entirely in the time domain.

32. The system as recited in claim 25, wherein the first and second blocks of samples received at the receiver are respectively associated with first and second block of samples of N samples that were transmitted from a transmitter.

33. The system as recited in claim 32, wherein the estimated channel response is of length M, where N is equal to at least 2M.

34. The system as recited in claim 25, wherein the receiver is part of a multicarrier data transmission system.

35. The system as recited in claim 25, wherein the channel is at least one of a direct channel, an echo channel and a crosstalk channel.

36. The system as recited in claim 25,

wherein the first block of samples and the second block of samples result from a transmitter operative to transmit a first block and a second block of training samples to the receiver through the channel, and

wherein the second block of training samples is the same as the first block of training samples except that one of the training samples is different.

37. The system as recited in claim 25,

wherein the first block of samples and the second block of samples result from a transmitter operative to transmit a group of training samples to the receiver through the channel, and

wherein the training samples in the group are periodic except that one of the training samples is different.

38. The system as recited in claim 25, wherein said receiver operative to produce provides the estimate of the received samples in both a time domain and a frequency domain.

39. The system as recited in claim 25, wherein the training sequence is used to train-up a near-end channel and a far-end channel.

40. The system as recited in claim 39, wherein the training sequence is used to train-up two or more of direct, echo, far-end crosstalk and near-end crosstalk channels.

41. The system as recited in claim 25, further comprising:

FFT/IFFT circuitry operative to subsequently track the estimated channel response during showtime operation.

42. The system as recited in claim 41, wherein said FFT/IFFT circuitry is operative to track in the frequency domain.

43. The system as recited in claim 42, wherein said FFT/IFFT circuitry operative to track, updates at most only a subset of the M channel taps.

44. The system as recited in claim 42, wherein said FFT/IFFT circuitry operative to track, converts frequency components of the tracked estimated channel response and other components in the frequency domain into values for one or more of the M channel taps in the time domain.