SEVERE ERROR DETECTORS, METHODS AND COMPUTER PROGRAM PRODUCTS THAT USE CONSTELLATION SPECIFIC ERROR EVENT THRESHOLDS TO DETECT SEVERE ERROR EVENTS DURING DEMODULATION OF A SIGNAL COMPRISING SYMBOLS FROM A PLURALITY OF SYMBOL CONSTELLATIONS

• US 6,553,518 B1
• Filed: 10/29/1999
• Issued: 04/22/2003
• Est. Priority Date: 03/08/1999
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Ware, Malcolm Scott, Davis, Gordon Taylor
• Primary Examiner(s): Decady, Albert
• Assistant Examiner(s): Torres, Joseph D.
• Application Number:
  US09/430,693
• Time in Patent Office: 1,271 Days
• Field of Search: 747/04, 375/232, 375/233, 375/287
• US Class Current: 714/704
• CPC Class Codes: H04L 2025/03363 Multilevel H04L2025/03369 takes precedence
• CPC Class Codes: H04L 2025/03694 Stop and go
• CPC Class Codes: H04L 25/03057 with a recursive structure H04L25/03031 takes precedence
• CPC Class Codes: H04L 25/4927 using levels matched to the quantisation levels of the channel

Abstract

Severe error detectors, methods, and computer program products use constellation specific error event thresholds to detect severe error events in a signal comprising symbols from a plurality of symbol constellations. In particular, a plurality of constellation thresholds are defined that correspond to a plurality of symbol or signal constellations. An output signal sample, which is associated with one of the plurality of symbol constellations, is received from a decision feedback equalizer. The absolute value of the output signal sample is compared with the constellation threshold that corresponds to the particular symbol constellation associated with the output signal sample. If the absolute value of the output signal sample exceeds the constellation threshold, then a severe error is recognized. The ITU V.90 Recommendation allows up to six different symbol or signal constellations to be used in the same communication session. The multiple constellations may be designed such that an error threshold suitable for one constellation may not be suitable for another constellation. As a result, a severe error detection circuit or module using a single constellation threshold for detecting severe errors may either under report or over report severe error events. Advantageously, multiple constellation thresholds may allow severe errors to be detected with greater precision than could be detected using a single constellation threshold.

Claims

1. A severe error detector, comprising:

means for defining a plurality of constellation thresholds that correspond to a plurality of symbol constellations;

means for receiving an output signal sample that is associated with one of the plurality of symbol constellations, the means for receiving comprising a decision feedback equalizer which comprises a digital filter characterized by a set of coefficients;

means for comparing an absolute value of the output signal sample with the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

means, responsive to the means for comparing, for recognizing a severe error only if the absolute value of the output signal sample exceeds the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

means for determining if a number of severe error occurrences exceeds a first error event threshold; and

means, responsive to the means for determining, for preventing the coefficients from being updated to new values if the number of severe error occurrences exceeds the first error event threshold.

2. A severe error detector as recited in claim 1, wherein the means for determining comprises:

means, responsive to the means for recognizing, for incrementing an error counter by a first constant if a severe error has been recognized; and

means for comparing the error counter with the first error event threshold.

3. A severe error detector as recited in claim 2, further comprising:

means, responsive to the means for recognizing, for decrementing the error counter by a second constant if a severe error has not been recognized;

means for comparing the error counter with a second error event threshold; and

means, responsive to the means for comparing the error counter with the second error event threshold, for setting the error counter equal to the second error event threshold if the second error event threshold exceeds the error counter.

4. A severe error detector as recited in claim 1, wherein the means for determining comprises:

means, responsive to the means for recognizing, for decrementing an error counter by a first constant if a severe error has been recognized; and

means for comparing the error counter with the first error event threshold.

5. A severe error detector as recited in claim 4, further comprising:

means, responsive to the means for recognizing, for incrementing the error counter by a second constant if a severe error has not been recognized;

means for comparing the error counter with a second error event threshold; and

means, responsive to the means for comparing the error counter with the second error event threshold, for setting the error counter equal to the second error event threshold if the error counter exceeds the second error event threshold.

6. A severe error detector as recited in claim 1, further comprising:

means, responsive to the means for determining, for invoking an error recovery protocol if the number of severe error occurrences exceeds the first error event threshold.

7. A method of detecting severe error events, comprising the steps of:

defining a plurality of constellation thresholds that correspond to a plurality of symbol constellations;

receiving an output signal sample that is associated with one of the plurality of symbol constellations from a decision feedback equalizer, the decision feedback equalizer comprising a digital filter characterized by a set of coefficients;

comparing an absolute value of the output signal sample with the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

recognizing a severe error only if the absolute value of the output signal sample exceeds the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

determining if a number of severe error occurrences exceeds a first error event threshold; and

preventing the coefficients from being updated to new values if the number of severe error occurrences exceeds the first error event threshold.

8. A method as recited in claim 7, wherein the determining step comprises the steps of:

incrementing an error counter by a first constant if a severe error has been recognized; and

comparing the error counter with the first error event threshold.

9. A method as recited in claim 8, further comprising the steps of:

decrementing the error counter by a second constant if a severe error has not been recognized;

comparing the error counter with a second error event threshold; and

setting the error counter equal to the second error event threshold if the second error event threshold exceeds the error counter.

10. A method as recited in claim 7, wherein the determining step comprises the steps of:

decrementing an error counter by a first constant if a severe error has been recognized; and

comparing the error counter with the first error event threshold.

11. A method as recited in claim 10, further comprising the steps of:

incrementing the error counter by a second constant if a severe error has not been recognized;

comparing the error counter with a second error event threshold; and

setting the error counter equal to the second error event threshold if the error counter exceeds second error event threshold.

12. A method as recited in claim 7, further comprising the step of:

invoking an error recovery protocol if the number of severe error occurrences exceeds the first error event threshold.

13. A computer program product for detecting severe error events, comprising:

a computer readable storage medium having computer readable program code embodied therein, the computer readable program code comprising;

computer readable program code for defining a plurality of constellation thresholds that correspond to a plurality of symbol constellations;

computer readable program code for receiving the output signal sample that is associated with one of the plurality of symbol constellations from a decision feedback equalizer, the decision feedback equalizer comprising a digital filter characterized by a set of coefficients;

computer readable program code for comparing an absolute value of the output signal sample with the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

computer readable program code, responsive to the computer readable program code for comparing, for recognizing a severe error only if the absolute value of the output signal sample exceeds the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

computer readable program code for determining if a number of severe error occurrences exceeds a first error event threshold; and

computer readable program code, responsive to the computer readable program code for determining, for preventing the coefficients from being updated to new values if the number of severe error occurrences exceeds the first error event threshold.

14. A computer program product as recited in claim 13, wherein the computer readable program code for determining comprises:

computer readable program code, responsive to the computer readable program code for recognizing, for incrementing an error counter by a first constant if a severe error has been recognized; and

computer readable program code for comparing the error counter with the first error event threshold.

15. A computer program product as recited in claim 14, further comprising:

computer readable program code, responsive to the computer readable program code for recognizing, for decrementing the error counter by a second constant if a severe error has not been recognized;

computer readable program code for comparing the error counter with a second error event threshold; and

computer readable program code, responsive to the computer readable program code for comparing the error counter with the second error event threshold, for setting the error counter equal to the second error event threshold if the second error event threshold exceeds the error counter.

16. A computer program product as recited in claim 13, wherein the computer readable program code for determining comprises:

computer readable program code, responsive to the computer readable program code for recognizing, for decrementing an error counter by a first constant if a severe error has been recognized; and

computer readable program code for comparing the error counter with the first error event threshold.

17. A computer program product as recited in claim 16, further comprising:

computer readable program code, responsive to the computer readable program code for recognizing, for incrementing the error counter by a second constant if a severe error has not been recognized;

computer readable program code for comparing the error counter with a second error event threshold; and

computer readable program code, responsive to the computer readable program code for comparing the error counter with the second error event threshold, for setting the error counter equal to the second error event threshold if the error counter exceeds the second error event threshold.

18. A computer program product as recited in claim 13, further comprising:

computer readable program code, responsive to the computer readable program code for determining, for invoking an error recovery protocol if the number of severe error occurrences exceeds the first error event threshold.

19. A severe error detector, comprising:

means for defining a plurality of constellation thresholds that correspond to a plurality of symbol constellations;

a decision feedback equalizer that is configured to receive an output signal sample that is associated with one of the plurality of symbol constellations, the decision feedback equalizer comprising a digital filter characterized by a set of coefficients;

means for comparing an absolute value of the output signal sample with the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

means, responsive to the means for comparing, for recognizing a severe error if the absolute value of the output signal sample exceeds the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

means for determining if a number of severe error occurrences exceeds an error event threshold; and

means, responsive to the means for determining, for preventing the coefficients from being updated to new values if the number of severe error occurrences exceeds the error event threshold.

20. A method of detecting severe error events, comprising the steps of:

defining a plurality of constellation thresholds that correspond to a plurality of symbol constellations;

receiving an output signal sample that is associated with one of the plurality of symbol constellations from a decision feedback equalizer, the decision feedback equalizer comprising a digital filter characterized by a set of coefficients;

comparing an absolute value of the output signal sample with the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

recognizing a severe error if the absolute value of the output signal sample exceeds the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

determining if a number of severe error occurrences exceeds an error event threshold; and

preventing the coefficients from being updated to new values if the number of severe error occurrences exceeds the error event threshold.

21. A computer program product for detecting severe error events, comprising:

a computer readable storage medium having computer readable program code embodied therein, the computer readable program code comprising;

computer readable program code for defining a plurality of constellation thresholds that correspond to a plurality of symbol constellations;

computer readable program code for receiving an output signal sample that is associated with one of the plurality of symbol constellations from a decision feedback equalizer, the decision feedback equalizer comprising a digital filter characterized by a set of coefficients;

computer readable program code for comparing an absolute value of the output signal sample with the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

computer readable program code, responsive to the computer readable program code for comparing, for recognizing a severe error if the absolute value of the output signal sample exceeds the constellation threshold that corresponds with the symbol constellation associated with the output signal sample;

computer readable program code for determining if a number of severe error occurrences exceeds an error event threshold; and

computer readable program code, responsive to the computer readable program code for determining, for preventing the coefficients from being updated to new values if the number of severe error occurrences exceeds the error event threshold.