Touch panel apparatus and methods

1. An apparatus for improving touch panel accuracy, comprising:

a calibration sampling controller to command an X-Y coordinate positioning apparatus to position a touch-emulating stylus at a known X-Y coordinate position relative to a touch panel and to depress the stylus;

a touch sampling module communicatively coupled to the calibration sampling controller to receive, sample and store a plurality of calibration sample values, each calibration sample value corresponding to a signal amplitude associated with a touch panel node, each node corresponding to an X-Y coordinate position associated with a touch panel row/column cross point, the signal amplitude corresponding to a physical proximity of a node to the X-Y coordinate position of the depressed stylus; and

a touch panel characteristic modeler communicatively coupled to the calibration sampling controller to receive the plurality of node calibration sample values and to generate a characteristic model of the plurality of node calibration sample values as a function of a path of the touch-emulating stylus across the touch panel,

a peak detection module communicatively coupled to the run-time buffer to compare the plurality of sample values and to retain a selected number of the largest-signal sample values; and

a sample sorting module communicatively coupled to the peak detection module to order the largest-signal X-Y node sample values according to magnitude.

2. The apparatus of claim 1 , each calibration sample value representing a magnitude of a corresponding row or column calibration signal at a calibration sampling time corresponding to a position along a corresponding row or column.

3. The apparatus of claim 1 , further comprising:

a calibration sample buffer communicatively coupled to the calibration sampling controller to store at least one of a set of column calibration sample values associated with stylus movement along a row or a set of row calibration sample values associated with stylus movement along a column.

4. An apparatus for improving touch panel accuracy, comprising:

a touch sampling module to receive and store a plurality of sample values, each sample value corresponding to a signal amplitude associated with a touch panel node, each node corresponding to an X-Y coordinate position associated with a touch panel row/column cross point, the signal amplitude corresponding to a physical proximity of a node to an X-Y coordinate position of a run-time touch event;

a distance projection module communicatively coupled to the touch sampling module to substitute a sample value associated with a largest-signal node on a path most strongly associated with the run-time touch panel event into a touch panel calibration model and to solve for a distance between the largest-signal node and a point-of-projection from the position of the run-time touch event onto the most strongly-associated path,

a run-time buffer coupled to the touch sampling module to store the plurality of sample values;

a peak detection module communicatively coupled to the run-time buffer to compare the plurality of sample values and to retain a selected number of the largest-signal sample values; and

a sample sorting module communicatively coupled to the peak detection module to order the largest-signal X-Y node sample values according to magnitude.

5. The apparatus of claim 4 , further comprising:

a touch panel characteristic modeler communicatively coupled to the touch sampling module to receive a superset of node calibration sample values and to generate a characteristic model of the superset of node calibration sample values as a function of a path of a touch-emulating stylus moving across the touch panel.

6. The apparatus of claim 5 , each calibration sample value representing a magnitude of a corresponding row or column calibration signal at a calibration sampling time corresponding to a position along a corresponding row or column.

7. A method of improving touch panel accuracy, comprising:

sampling at least one of a set of row signals or a set of column signals to acquire a plurality of sets of X-Y node calibration sample values at times corresponding to a series of calibration touch events created at known locations along a path on a touch panel, each X-Y node calibration sample value representing a magnitude of a corresponding row or column calibration signal at a calibration sampling time corresponding to a position along a corresponding row or column;

curve-fitting the set of X-Y node calibration sample values to obtain a touch panel calibration model of the path;

sampling at least one of a set of row signals or a set of column signals associated with a run-time touch panel event to acquire a set of X-Y node run-time sample values;

determining at least one of a row or column as a path most strongly associated with a run-time touch panel event;

determining a ratio of a normalized largest-possible peak value to the largest-signal X-Y node on the most strongly-associated path, the largest-signal X-Y node sample value normalized to the largest-possible peak value;

determining a natural logarithm of the ratio;

calculating a square root of the natural logarithm of the ratio to determine the distance between the largest-signal X-Y node on the most strongly-associated path and the point-of-projection of the run-time touch event location onto the most strongly-associated path; and

determining a distance between a largest-signal X-Y node on the most strongly-associated path and a point-of-projection of a run-time touch event location onto the most strongly-associated path by substituting a magnitude of a largest-magnitude X-Y node run-time sample value associated with the most strongly-associated path into the touch panel calibration model and solving for the distance.

8. The method of improving touch panel accuracy of claim 7 , further comprising:

determining the row or column most strongly associated with the run-time touch panel event as a row or column associated with a largest-magnitude X-Y node run-time sample value compared to other X-Y node run-time sample values associated with surrounding rows or columns;

for peripheral X-Y nodes, determining a direction of the point-of-projection from the largest-signal X-Y node on the most strongly-associated path as being toward a panel edge; and

for interior X-Y nodes, determining a direction of the point-of-projection from the largest-signal X-Y node on the most strongly-associated path as being toward a second-largest X-Y node run-time sample value.

9. The method of improving touch panel accuracy of claim 7 , further comprising:

pressing the touch panel and sweeping along a row or column between opposite edges of the touch panel to create the set of row or column calibration signals.

10. The method of improving touch panel accuracy of claim 9 , further comprising:

sweeping along the row or column at a substantially constant rate to create an even distribution of calibration samples per increment of time.

11. A method of improving touch panel accuracy, comprising:

sampling at least one of a set of row signals or a set of column signals associated with a run-time touch panel event to acquire a set of X-Y node run-time sample values;

determining at least one of a row or column as a path most strongly associated with a run-time touch panel event;

determining a ratio of a normalized largest-possible peak value to the largest-signal X-Y node on the most strongly-associated path, the largest-signal X-Y node sample value normalized to the largest-possible peak value;

determining a natural logarithm of the ratio;

calculating a square root of the natural logarithm of the ratio to determine the distance between the largest-signal X-Y node on the most strongly-associated path and the point-of-projection of the run-time touch event location onto the most strongly-associated path, and

determining a distance between a largest-signal X-Y node on the most strongly-associated path and a point-of-projection of a run-time touch event location onto the most strongly-associated path by substituting a magnitude of a largest-magnitude X-Y node run-time sample value associated with the most strongly-associated path into the touch panel calibration model and solving for the distance.

12. The method of improving touch panel accuracy of claim 11 , further comprising:

choosing a row or column corresponding to a largest-magnitude X-Y node sample value of all X-Y node sample values acquired from the touch panel during a sampling sequence.

13. The method of improving touch panel accuracy of claim 11 , further comprising:

for peripheral X-Y nodes, determining a direction of the point-of-projection from the largest-signal X-Y node on the most strongly-associated path as being toward a panel edge; and

for interior X-Y nodes, determining a direction of the point-of-projection from the largest-signal X-Y node on the most strongly-associated path as being toward a second-largest X-Y node run-time sample value.

14. The method of improving touch panel accuracy of claim 11 , the touch panel calibration model consisting of a model selected from one of an exponential model, a parabolic model, or a linear model.

15. The method of improving touch panel accuracy of claim 11 , further including:

normalizing the largest-possible peak value to 100.

16. The method of improving touch panel accuracy of claim 11 , further comprising:

determining a numerical difference between a normalized largest-possible peak value and the largest-signal X-Y node on the most strongly-associated path, the largest-signal X-Y node sample value normalized to the largest-possible peak value;

determining a ratio of the numerical difference to a touch panel characteristic parameter value, the touch panel characteristic parameter value selected to approximate a shape of a touch panel row or column signal strength characteristic curve; and

calculating a square root of the ratio to determine the distance between the largest-signal X-Y node on the most strongly-associated path and the point-of-projection of the run-time touch event location onto the most strongly-associated path.

17. The method of improving touch panel accuracy of claim 16 , further comprising:

normalizing the largest- possible peak value to 100.

18. The method of improving touch panel accuracy of claim 11 , further comprising:

determining an absolute value of a numerical difference between a normalized largest-possible peak value and the largest-signal X-Y node on the most strongly-associated path, the largest-signal X-Y node sample value normalized to the largest-possible peak value; and

determining the distance between the largest-signal X-Y node on the most strongly-associated path and the point-of-projection of the run-time touch event location onto the most strongly-associated path as plus or minus a ratio of the absolute value of the numerical difference to a touch panel characteristic parameter value, the characteristic parameter value selected to approximate a shape of a touch panel row or column signal strength characteristic curve.