Adaptive data recovery system with input signal equalization
View Patent ↗Data receivers often include equalizers for operating on received signals. The equalizers often have a plurality of taps, with signals from each tap weighted based on tap settings or values. The tap settings may be set based on bit error rates of data output from the equalizer. In some embodiments data output from the equalizer is split into two signals, and the two signals are processed to indicate a data eye of the data output from the equalizer. Preferred tap settings may be determined by setting tap settings to different values and using tap settings expected to maximize the data eye. This may be performed separately for different bit settings in an attempt to reduce effects of inter-signal interference.
1. A method for setting tap values for an adaptive equalizer, comprising:
setting the tap values for an adaptive equalizer to a first set of values;
for a plurality of different data cells, sampling and comparing an output of the adaptive equalizer at a first phase and first voltage, the first phase and the first voltage being within a data cell of the output of the adaptive equalizer;
for the plurality of different data cells, sampling and comparing the output of the adaptive equalizer at a second phase and second voltage, the second phase and the second voltage being within the data cell of the output of the adaptive equalizer;
determining if the sampled and compared output of the adaptive equalizer at the first phase and first voltage matches the sampled and compared output of the adaptive equalizer at the second phase and second voltage and the sampled and compared output at the first phase and voltage for at least some of the plurality of different data cells matches a predefined bit pattern; and
adjusting at least one tap value for the adaptive equalizer based on a determination that the sampled and compared output of the adaptive equalizer at the first phase and first voltage does not match the sampled and compared output of the adaptive equalizer at the second phase and second voltage.
2. The method of claim 1 further comprising performing, a plurality of times, each time of the plurality of times for different data cells:
sampling and comparing the output of the adaptive equalizer at the first phase and first voltage,
sampling and comparing the output of the adaptive equalizer at the second phase and second voltage, and
determining if the sampled and compared output of the adaptive equalizer at the first phase and first voltage matches the sampled and compared output of the adaptive equalizer at the second phase and second voltage.
3. The method of claim 2 wherein the adjusting at least one tap value for the adaptive equalizer based on the determination that the sampled and compared output of the adaptive equalizer at the first phase and first voltage does not match the sampled and compared output of the adaptive equalizer at the second phase and second voltage comprises adjusting at least one tap value for the adaptive equalizer based on a number of times the sampled and compared output of the adaptive equalizer at the first phase and first voltage does not match the sampled and compared output of the adaptive equalizer at the second phase and second voltage.
4. The method of claim 3 , wherein the sampled and compared data cell from the output of the adaptive equalizer is a data cell within the predetermined bit pattern.
5. The method of claim 4 wherein the adjusting at least one tap value is based on the number of times the sampled and compared output of the adaptive equalizer at the first phase and the first voltage does not match the sampled and compared output of the adaptive equalizer at the second phase and the second voltage for a particular data cell within the different data cells matching the predefined bit pattern.
6. The method of claim 4 further comprising changing the predefined bit pattern over time.
7. The method of claim 2 further comprising, for each of a plurality of different tap values, performing, a plurality of times, each time of the plurality of times for different data cells:
sampling and comparing the output of the adaptive equalizer at the first phase and first voltage,
sampling and comparing the output of the adaptive equalizer at the second phase and second voltage, and
determining if the sampled and compared output of the adaptive equalizer at the first phase and first voltage matches the sampled and compared output of the adaptive equalizer at the second phase and second voltage.
8. The method of claim 7 wherein the adjusting at least one tap value for the adaptive equalizer based on the determination that the sampled and compared output of the adaptive equalizer at the first phase and first voltage does not match the sampled and compared output of the adaptive equalizer at the second phase and second voltage comprises setting the tap values to the tap value of the plurality of different tap values for which the sampled and compared output of the adaptive equalizer at the first phase and first voltage does not match the sampled and compared output of the adaptive equalizer at the second phase and second voltage the least number of times.
9. The method of claim 1 wherein only one of either the first phase and the second phase or the first voltage and second voltage are different.
10. The method of claim 1 further comprising changing at least one of the first phase and the first voltage over time.
11. Circuitry for setting tap values for an equalizer, comprising:
first data recovery circuitry configured to recover data from a data stream, the data stream being recovered from an output of the equalizer;
second data recovery circuitry configured to recover data from the data stream, the data stream being recovered from the output of the equalizer;
comparison circuitry configured to compare data recovered from the first data recovery circuitry with data recovered from the second data recovery circuitry when data recovered from at least one of the first data recovery circuitry or the second data recovery circuitry matches a predefined bit pattern;
counting circuitry configured to count a number of times the comparison circuitry indicates data recovered from the first data recovery circuitry does not match data recovered from the second data recovery circuitry;
control circuitry configured to set at least one tap value of the equalizer based on the number of counted times.
12. The circuitry of claim 11 wherein the comparison circuitry is further configured to compare at least one of the data recovered from the first data recovery circuitry and the data recovered from the second data recovery circuitry with a predefined bit pattern.
13. The circuitry of claim 12 wherein the comparison circuitry is further configured to indicate that data recovered from the first data recovery circuitry does not match data recovered from the second data recovery circuitry only for data included in the predefined bit pattern.
14. The circuitry of claim 13 wherein the control circuitry is configured to set the tap values to a plurality of different sets of tap values, and to select a one set of tap values of the plurality of different sets of tap values based on the set of tap values of the plurality of different sets of tap values with the least number of counted times.
15. The circuitry of claim 14 further comprising equalizer circuitry implementing the equalizer, the equalizer circuitry including a summer for summing outputs of taps of the equalizer, the summer including an inductively peaked cascade stage.
16. A method of determining tap values for an equalizer in a receiver comprising:
setting tap values to a plurality of different sets of values and, for each of the plurality of different sets of values:
recovering data output from the equalizer in a first data channel and recovering data output from the equalizer in a second data channel; and
determining a number of times data recovered from the first data channel does not match data recovered from the second data channel and whether the output of one of the first data and the second data channel matches a predefined bit pattern; and
setting, the tap values to a one of the plurality of different sets of values for which the number of times data recovered from the first data channel does not match data recovered from the second data channel indicates the lowest bit error rate.
17. The method of claim 16 further comprising, for each of the plurality of different sets of values, setting a bit pattern to a plurality of different bit patterns, and wherein determining the number of times data recovered from the first data channel does not match data recovered from the second data channel comprising determining the number of times data recovered from the first data channel does not match data recovered from the second data channel for the set bit pattern.