Method for reallocating data in a discrete multi-tone communication system

• US 6,122,247 A
• Filed: 11/24/1997
• Issued: 09/19/2000
• Est. Priority Date: 11/24/1997
• Status: Expired due to Term

• Current Assignee: Apple Inc.
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Levin, Howard E., Teitler, Nicole D.
• Primary Examiner(s): Nguyen, Chau
• Assistant Examiner(s): HYUN, SOON D
• Application Number:
  US08/976,472
• Time in Patent Office: 1,030 Days
• Field of Search: 370/480, 370/482, 370/484, 370/203, 370/210, 370/252, 370/204, 370/205, 370/209, 375/260, 375/224, 375/225, 375/227
• US Class Current: 370/210
• CPC Class Codes: H04L 1/0059 Convolutional codes
• CPC Class Codes: H04L 5/0007 the frequencies being orthogonal, e.g. OFDM(A), DMT
• CPC Class Codes: H04L 5/0046 Determination of how many bits are transmitted on different sub-channels
• CPC Class Codes: H04L 5/006 Quality of the received signal, e.g. BER, SNR, water filling

Abstract

In the present invention, an ADSL system (10) determines if a bit swap is to occur based upon a projected margin of a best carrier and a current margin of a worst carrier. If a proposed swap is not beneficial (320, 512, 604, 612) no swap occurs. In addition, the present invention determines the effects of a bit swap on an error correction scheme (510, 611). If the swap effects the scheme, adjustments to system parameters are made to assure continued error correction. In addition, the present invention allows for bit swapping to and from bins containing zero or two bits respectively (331, 332, 403).

Claims

1. A method for dynamically reallocating data in a discrete multi-tone communication system having a plurality of bins for sending data, the method comprising the steps of:

determining a signal-to-noise value for each of the plurality of bins;

defining a current reference signal-to-noise value;

defining a projected reference signal-to-noise value;

calculating a current signal-to-noise margin for each of the plurality of bins based on the signal-to-noise value and the current reference signal-to-noise value;

calculating a projected signal-to-noise margin for each of the plurality of bins based on the signal-to-noise value and the projected reference signal-to-noise value;

identifying a first portion of the plurality of bins which have a maximum projected signal-to-noise margin;

identifying a second portion of the plurality of bins which have a minimum current signal-to-noise margin; and

dynamically determining a first number of bits to be reallocated from the second portion of bins to the first portion of bins.

2. The method of claim 1, wherein the step of dynamically determining further comprises:

dynamically determining the first number of bits from a set comprising zero, one, two, and three.

3. The method of claim 2, further comprising the step of:

providing a fine gain adjustment value if the first number is not three.

4. The method of claim 1, further comprising the step of:

storing the current reference signal-to-noise value and the projected reference signal-to-noise value in a table in a memory.

5. The method of claim 1, further comprising the step of:

allocating an error correction bit to at least one of the plurality of bins.

6. The method of claim 1, further comprising the steps of:

identifying a best bin within the first portion of the plurality of bins; and

identifying a worst bin within the second portion of the plurality of bins.

7. The method of claim 6, further comprising the step of:

determining if a projected signal-to-noise margin of the best bin is greater than a sum of a current signal-to-noise margin of the worst bin plus a first threshold value.

8. The method of claim 7, wherein the first threshold value is non-zero.

9. The method of claim 6, wherein said step of dynamically reallocating comprises the steps of:

dynamically reallocating the first number of bits from the worst bin to the best bin if a current number of bits allocated to the best bin is zero and a current number of bits allocated to the worst bin is two; and

dynamically reallocating the first number of bits from the worst bin to the best bin if the current number of bits allocated to the best bin is greater than zero and the current number of bits allocated to the worst bin is greater than two.

10. The method of claim 9, wherein if the current number of bits allocated to the best bin is zero, then the first number of bits is two.

11. The method of claim 9, wherein said step of dynamically reallocating further comprises the steps of:

if the current number of bits allocated to the best bin is greater than zero, then the first number of bits is one.

12. The method of claim 6, further comprising the steps of:

determining if a current number of bits allocated to the best bin is zero.

13. The method of claim 12, wherein said step of dynamically reallocating comprises the steps of:

if a predetermined number of used bins are used, dynamically allocating a new error correction bit to the best bin.

14. The method of claim 13, wherein the step of dynamically allocating if a predetermined number of bins are used further includes dynamically allocating when Trellis error is used.

15. The method of claim 13, wherein said step of dynamically allocating comprises the step of:

if a predetermined number of bins are not used, identifying a second worst bin within the second portion of the plurality of bins.

16. The method of claim 15, wherein said step of dynamically reallocating comprises the steps of:

determining if a current number of bits allocated to the second worst bin is greater than two.

17. The method of claim 16, wherein said step of dynamically reallocating further comprises the steps of:

deallocating one bit from the worst bin;

deallocating one bit from the second worst bin; and

allocating two bits to the best bin.

18. The method of claim 16, further comprising the steps of:

determining if a projected signal-to-noise margin of a second best bin is greater than a sum of a current signal-to-noise margin of the worst bin plus a second threshold value; and

determining if the current number of bits allocated to the second best bin is greater than zero.

19. The method of claim 18, wherein the second threshold value is non-zero.

20. The method of claim 18, wherein the second threshold value is zero.

21. The method of claim 18, wherein said step of dynamically reallocating further comprises the steps of:

deallocating one bit from the worst bin;

deallocating two bits from the second worst bin;

allocating two bits to the best bin; and

allocating one bit to the second best bin.

22. The method of claim 12, wherein said step of dynamically reallocating comprises the steps of:

if error correction is being used and a number of used bins is odd, dynamically reallocating one bit from the worst bin to the best bin, and deallocating an error correction bit from the worst bin.

23. The method of claim 22, wherein the error correction is Trellis error correction.

24. The method of claim 22, wherein said step of dynamically reallocating comprises the step of:

if error correction is not used or the number of used bins is even, identifying a second best bin within the first portion of the plurality of bins.

25. The method of claim 16, further comprising the steps of:

determining if a projected signal-to-noise margin of a second best bin is greater than a sum of a current signal-to-noise margin of the worst bin plus a second threshold value.

26. The method of claim 25, wherein the second threshold value is non-zero.

27. The method of claim 25, wherein the second threshold value is zero.

28. The method of claim 25, further comprising the steps of:

determining if the current number of bits allocated to the second best bin is greater than zero.

29. The method of claim 25, wherein said step of dynamically reallocating comprises the steps of:

determining if a current number of bits allocated to the second worst bin is greater than two.

30. The method of claim 29, wherein said step of dynamically reallocating further comprises the steps of:

deallocating two bits from the worst bin;

deallocating one bit from the second worst bin;

allocating one bit to the best bin; and

allocating two bits to the second best bin.

31. The method of claim 29, wherein said step of dynamically reallocating further comprises the steps of:

deallocating two bits from the worst bin;

allocating one bit to the best bin; and

allocating one bit to the second best bin.

32. A method for dynamically reallocating data in a discrete multi-tone communication system having a plurality of bins for sending data, the method comprising the steps of:

determining a signal-to-noise value for each of the plurality of bins;

storing a plurality of reference signal-to-noise values;

defining a current reference signal-to-noise value for each of the plurality of bins by selecting one of said plurality of reference signal-to-noise values for each of the plurality of bins, wherein the current reference signal-to-noise value for each of the plurality of bins is determined by a number of bits currently allocated to a corresponding one of the plurality of bins;

defining a projected reference signal-to-noise value for each of the plurality of bins by selecting one of said plurality of reference signal-to-noise values for each of the plurality of bins, wherein the projected reference signal-to-noise value for each of the plurality of bins is determined by a number of bits projected to be allocated to the corresponding one of the plurality of bins;

calculating a signal-to-noise margin for each of the plurality of bins based on the signal-to-noise value and the current reference signal-to-noise value;

calculating a projected signal-to-noise margin for each of the plurality of bins based on the signal-to-noise value and the projected reference signal-to-noise value;

identifying a first best bin within the plurality of bins, the first best bin having a maximum projected signal-to-noise margin;

identifying a second best bin within the plurality of bins, the second best bin having a next maximum projected signal-to-noise margin;

identifying a first worst bin within the plurality of bins, the first worst bin having a minimum current signal-to-noise margin;

identifying a second worst bin within the plurality of bins, the second worst bin having a next minimum current signal-to-noise margin; and

dynamically reallocating a first number of bits from at least one of the first and second best bins to at least one of the first and second worst bins based on bit loading.