System and Method for the Adjustment of Compensation Applied to a Signal Using Filter Patterns

• US 20070300119A1
• Filed: 05/24/2007
• Published: 12/27/2007
• Est. Priority Date: 05/30/2006
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Hidaka, Yasuo
• Granted Patent: US 7,804,894 B2
• Time in Patent Office: Days
• US Class Current: 714/746
• CPC Class Codes: H04L 25/03885 adaptive
• CPC Class Codes: H04L 25/063 Setting decision thresholds using feedback techniques only

Abstract

In one embodiment of the present invention, a method for adjusting a signal includes applying at least one of a loss compensation for frequency-dependent distortion and an offset compensation for DC-offset distortion to a data signal before or after the distortion occurs to generate an output signal. The method also includes, using a clock signal, sampling the output signal to generate a plurality of data values and an error value, each data value comprising either a high value or a low value based on the sampling of the output signal, the error value indicating residue of the distortion based on the sampling of the output signal. The method further includes selecting one or more filter patterns from a set of a plurality of filter patterns, each filter pattern comprising a plurality of data values. The method also includes monitoring the output signal for one or more patterns of data values corresponding to the one or more selected filter patterns. The method further includes detecting a pattern of data values in the output signal corresponding to one of the one or more selected filter patterns. The method also includes adjusting the at least one of the loss compensation and the offset compensation applied to the data signal based on the error value.

Claims

1. A method for adjusting a signal, comprising:

applying at least one of a loss compensation for frequency-dependent distortion and an offset compensation for DC-offset distortion to a data signal before or after the distortion occurs to generate an output signal;

using a clock signal, sampling the output signal to generate a plurality of data values and an error value, each data value comprising either a high value or a low value based on the sampling of the output signal, the error value indicating residue of the distortion based on the sampling of the output signal;

selecting one or more filter patterns from a set of a plurality of filter patterns, each filter pattern comprising a plurality of data values;

monitoring the output signal for one or more patterns of data values corresponding to the one or more selected filter patterns;

detecting a pattern of data values in the output signal corresponding to one of the one or more selected filter patterns; and

adjusting the at least one of the loss compensation and the offset compensation applied to the data signal based on the error value.

2. The method of claim 1, wherein the error value comprises either a high value or a low value.

3. The method of claim 1, wherein:

the error value comprises at least one boundary value sampled at a transition in value between two successive data values; and

each filter pattern includes a transition in value between the corresponding two successive data values.

4. The method of claim 1, wherein the adjustment of compensation applied to the data signal is further based on the high or low values of one or more data values in the detected pattern of data values.

5. The method of claim 1, wherein:

the error value comprises a pulse-width value;

the pulse-width value is derived from two successive boundary values and an in-between data value in three successive data values with two transitions; and

the loss compensation applied to the data signal is adjusted based on the pulse-width value.

6. The method of claim 1, wherein the set of the plurality of filter patterns is pre-determined and fixed.

7. The method of claim 1, wherein the set of the plurality of filter patterns is dynamically adaptable based on the data signal.

8. The method of claim 7, further comprising:

monitoring the output signal for patterns of data values in the data signal; and

updating the set of filter patterns based on the patterns of data values observed in the output signal.

9. The method of claim 8 wherein:

the data signal comprises even data and odd data;

monitoring the output signal for patterns of data values comprises;

monitoring the data beginning with even data for patterns of data values and the data beginning with odd data for patterns of data values; and

comparing the observed patterns in the data beginning with even data and in the data beginning with odd data; and

updating the set of filter patterns based on the patterns of data values observed comprises updating the set to include only those filter patterns observed both in the data beginning with even data and in the data beginning with odd data.

10. The method of claim 8, wherein:

the data signal comprises even data and odd data;

monitoring the output signal for patterns of data values comprises;

monitoring the data beginning with even data for patterns of data values and the data beginning with odd data for patterns of data values; and

comparing the observed patterns in the data beginning with even data and in the data beginning with odd data; and

updating the set of filter patterns based on the patterns of data values observed comprises updating the set to include the same number of filter patterns observed only in the data beginning with even data as the number of filter patterns observed only in the data beginning with odd data.

11. The method of claim 1, wherein:

filter patterns are selected sequentially from the set of the plurality of filter patterns;

only one filter pattern is selected at a time;

the output signal is monitored for patterns of data values corresponding only to the selected filter pattern; and

after a pattern of data values in the output signal corresponding to the selected filter pattern is detected, the next filter pattern in the set is selected.

12. The method of claim 1, wherein:

filter patterns are selected randomly from the set of the plurality of filter patterns;

only one filter pattern is selected at a time;

the output signal is monitored for patterns of data values corresponding only to the selected filter pattern; and

after a pattern of data values in the output signal corresponding to the selected filter pattern is detected, a next filter pattern is selected randomly from the set.

13. The method of claim 1, wherein:

all filter patterns in the list are selected initially; and

the output signal is monitored for patterns of data values corresponding to the selected filter patterns;

the method further comprising;

after a pattern of data values in the output signal corresponding to one of the selected filter patterns is detected, deselecting the filter pattern corresponding to the detected pattern such that the output signal is no longer monitored for the detected pattern corresponding to the deselected filter pattern; and

reselecting all filter patterns in the set after patterns in the output signal corresponding to all filter patterns in the set are detected.

14. The method of claim 1, wherein, if the one or more patterns of data values corresponding to the one or more selected filter patterns is not detected in the output signal within a period of time, the one or more selected filter patterns is deselected and another one or more filter patterns is selected from the set.

15. An adaptive equalizer, comprising:

an equalizer configured to apply at least one of a loss compensation for frequency-dependent distortion and an offset compensation for DC-offset distortion to a data signal before or after the distortion occurs to generate an output signal;

a sampler configured to;

receive the output signal and a clock signal; and

using the clock signal, sample the output signal to generate a plurality of data values and an error value, each data value comprising either a high value or a low value based on the sampling of the output signal, the error value indicating residue of the distortion based on the sampling of the output signal; and

a controller configured to;

select one or more filter patterns from a set of a plurality of filter patterns, each filter pattern comprising a plurality of data values;

monitor the output signal for one or more patterns of data values corresponding to the one or more selected filter patterns;

detect a pattern of data values in the output signal corresponding to one of the one or more selected filter patterns; and

adjust the at least one of the loss compensation and the offset compensation applied to the data signal based on the error value.

16. The adaptive equalizer of claim 15, wherein the error value comprises either a high value or a low value.

17. The adaptive equalizer of claim 15, wherein:

the error value comprises at least one boundary value sampled at a transition in value between two successive data values; and

each filter pattern includes a transition in value between the corresponding two successive data values.

18. The adaptive equalizer of claim 15, wherein the controller is further configured to adjust the compensation applied to the data signal based on the high or low value of one or more data values in the detected pattern of data values.

19. The adaptive equalizer of claim 15, wherein:

the error value comprises a pulse-width value;

the pulse-width value is derived from two successive boundary values and an in-between data value in three successive data values with two transitions; and

the loss compensation applied to the data signal is adjusted based on the pulse-width value.

20. The adaptive equalizer of claim 15, wherein the set of the plurality of filter patterns is pre-determined and fixed.

21. The adaptive equalizer of claim 15, wherein the set of the plurality of filter patterns is dynamically adaptable based on the data signal.

22. The adaptive equalizer of claim 21, wherein the controller is further configured to:

monitor the output signal for patterns of data values in the data signal; and

update the set of filter patterns based on the patterns of data values observed in the output signal.

23. The adaptive equalizer of claim 22, wherein:

the data signal comprises even data and odd data;

monitoring the output signal for patterns of data values comprises;

monitoring the data beginning with even data for patterns of data values and the data beginning with odd data for patterns of data values; and

comparing the observed patterns in the data beginning with even data and in the data beginning with odd data; and

updating the set of filter patterns based on the patterns of data values observed comprises updating the set to include only those filter patterns observed both in the data beginning with even data and in the data beginning with odd data.

24. The adaptive equalizer of claim 22, wherein:

the data signal comprises even data and odd data;

monitoring the output signal for patterns of data values comprises;

monitoring the data beginning with even data for patterns of data values and the data beginning with odd data for patterns of data values; and

comparing the observed patterns in the data beginning with even data and in the data beginning with odd data; and

updating the set of filter patterns based on the patterns of data values observed comprises updating the set to include the same number of filter patterns observed only in the data beginning with even data as the number of filter patterns observed only in the data beginning with odd data.

25. The adaptive equalizer of claim 15, wherein:

filter patterns are selected sequentially from the set of the plurality of filter patterns;

only one filter pattern is selected at a time;

the output signal is monitored for patterns of data values corresponding only to the selected filter pattern; and

after a pattern of data values in the output signal corresponding to the selected filter pattern is detected, the next filter pattern in the set is selected.

26. The adaptive equalizer of claim 15, wherein:

filter patterns are selected randomly from the set of the plurality of filter patterns;

only one filter pattern is selected at a time;

the output signal is monitored for patterns of data values corresponding only to the selected filter pattern; and

after a pattern of data values in the output signal corresponding to the selected filter pattern is detected, a next filter pattern is selected randomly from the set.

27. The adaptive equalizer of claim 15, wherein:

all filter patterns in the list are selected initially;

the output signal is monitored for patterns of data values corresponding to the selected filter patterns; and

the controller is further configured to;

after a pattern of data values in the output signal corresponding to one of the selected filter patterns is detected, deselect the filter pattern corresponding to the detected pattern such that the output signal is no longer monitored for the detected pattern corresponding to the deselected filter pattern; and

reselect all filter patterns in the set after patterns in the output signal corresponding to all filter patterns in the set are detected.

28. The adaptive equalizer of claim 15, wherein, if the one or more patterns of data values corresponding to the one or more selected filter patterns is not detected in the output signal within a period of time, the controller is further configured to deselect the one or more selected filter patterns and select another one or more filter patterns from the set.

29. A method for adjusting a signal, comprising:

applying at least one of a loss compensation for frequency-dependent distortion and an offset compensation for DC-offset distortion to a data signal before or after the distortion occurs to generate an output signal;

using a clock signal, sampling the output signal to generate a plurality of data values and an error value, each data value comprising either a high value or a low value based on the sampling of the output signal, the error value indicating residue of the distortion based on the sampling of the output signal;

selecting one or more filter patterns from a set of a plurality of filter patterns, each filter pattern comprising a plurality of data values;

monitoring the output signal for one or more patterns of data values corresponding to the one or more selected filter patterns;

detecting a pattern of data values in the output signal corresponding to one of the one or more selected filter patterns; and

adjusting the at least one of the loss compensation and the offset compensation applied to the data signal based on the detected pattern.

30. An adaptive equalizer, comprising:

an equalizer configured to apply at least one of a loss compensation for frequency-dependent distortion and an offset compensation for DC-offset distortion to the data signal before or after the distortion occurs to generate an output signal;

a sampler configured to;

receive the output signal and a clock signal; and

using the clock signal, sample the output signal to generate a plurality of data values and an error value, each data value comprising either a high value or a low value based on the sampling of the output signal, the error value indicating residue of the distortion based on the sampling of the output signal; and

a controller configured to;

select one or more filter patterns from a set of a plurality of filter patterns, each filter pattern comprising a plurality of data values;

monitor the output signal for one or more patterns of data values corresponding to the one or more selected filter patterns;

detect a pattern of data values in the output signal corresponding to one of the one or more selected filter patterns; and

adjust the at least one of the loss compensation and the offset compensation applied to the input data signal based on the detected pattern.