Strategically selecting data captures for signal predistortion
An approach for determining corrective predistortion functions includes maintaining a history of recent estimations of a plurality of corrective predistortion functions. Each of those functions corresponds to a respective signal characteristic. By determining a feature of each of those functions from the history, one of the corrective predistortion functions is selected for a next estimation based upon which of the functions has a feature within the history that satisfies a selected criteria. In a disclosed example, the function having the lowest number of estimations within a recent history is selected as the function to have the highest priority for a next estimation.
1. A method of determining a corrective predistortion function, comprising the steps of:
determining a history of estimations of a plurality of corrective predistortion functions that each correspond to a respective signal characteristic;
determining a feature of each of the plurality of corrective predistortion functions from the history; and
selecting one of the corrective predistortion functions for a next estimation based on the selected corrective predistortion function having the determined feature that satisfies a selected criterion.
2. The method of claim 1 , wherein the signal characteristic comprises a signal power level.
3. The method of claim 1 , wherein the selected criterion comprises a number of times that each corrective predistortion function has been estimated within the determined history.
4. The method of claim 3 , comprising
selecting the corrective predistortion function having the lowest number of times.
5. The method of claim 1 , comprising
sampling a signal a plurality of times to obtain a plurality of capture sets each comprising a plurality of signal samples, each capture set corresponding to a corresponding one of the signal characteristics; and
selecting the capture set corresponding to the signal characteristic of the selected corrective predistortion function for the next estimation.
6. The method of claim 5 , comprising
estimating the selected corrective predistortion using the selected capture set.
7. The method of claim 5 , comprising
obtaining the plurality of capture sets within an estimation interval;
obtaining a new plurality of capture sets within each of a plurality of successive estimation intervals; and
at least temporarily storing the capture sets from each estimation interval for at least the duration of the corresponding estimation interval.
8. The method of claim 7 , comprising
selecting the corrective predistortion function at the end of each of the estimation intervals.
9. A device for determining a corrective predistortion function, comprising
a history module configured to store a history of estimations of a plurality of corrective predistortion functions that each correspond to a respective signal characteristic; and
an estimator configured to
determine a feature of each of the plurality of corrective predistortion functions from the history and
select one of the corrective predistortion functions for a next estimation based on the selected corrective predistortion function having the determined feature that satisfies a selected criterion.
10. The device of claim 9 , wherein the signal characteristic comprises a signal power level.
11. The device of claim 9 , wherein a selected criterion comprises a number of times that each corrective predistortion function has been estimated within the stored history.
12. The device of claim 11 , wherein the estimator is configured to select the corrective predistortion function having the lowest number of times.
13. The device of claim 9 , wherein the estimator is configured to
sample a signal a plurality of times to obtain a plurality of capture sets each comprising a plurality of signal samples, each capture set corresponding to a corresponding one of the signal characteristics; and
select the capture set corresponding to the signal characteristic of the selected corrective predistortion function for the next estimation.
14. The device of claim 13 , wherein the estimator is configured to estimate the selected corrective predistortion using the selected capture set.
15. The device of claim 13 , wherein the estimator is configured to
obtain the plurality of capture sets within a sampling window;
obtain a new plurality of capture sets within each of a plurality of successive sampling windows; and
at least temporarily store the capture sets from each sampling window for at least the duration of the corresponding sampling window.
16. The device of claim 15 , wherein the estimator is configured to select the corrective predistortion function at the end of each of the sampling windows.