NOISE DETECTION IN MULTI-TOUCH SENSORS

• US 20080158169A1
• Filed: 01/03/2007
• Published: 07/03/2008
• Est. Priority Date: 01/03/2007
• Status: Active Grant

• Current Assignee: Apple Inc.
• Original Assignee: Apple Computer Incorporated (Apple Inc.)
• Inventors: Carl Westerman, Wayne, O'Connor, Sean Erik
• Granted Patent: US 7,643,011 B2
• Time in Patent Office: Days
• US Class Current: 345/173
• CPC Class Codes: G06F 3/0418 for error correction or compensation, e.g. based on parallax, calibration or alignment

Abstract

Multi-touch touch-sensing devices and methods are described herein. The touch sensing devices can include multiple sense points, each of which can be stimulated with a plurality of periodic waveforms having different frequencies to measure a touch value at the sense point. Noise at one or more of the frequencies can interfere with this measurement. Therefore, various noise detection (and rejection) techniques are described. The noise detection techniques include two-clean-frequency noise detection, one-clean-frequency noise rejection, and combined two-clean-frequency/one-clean-frequency noise detection. Each of the noise detection techniques can include statistical analyses of the sample values obtained. The touch sensing methods and devices can be incorporated into interfaces for a variety of electronic devices such as a desktop, tablet, notebook, and handheld computers, personal digital assistants, media players, and mobile telephones.

Claims

1. A method of detecting noise in one of at least three touch images, each touch image comprising a plurality of touch sample values obtained by stimulating a touch sensor at a unique frequency, the method comprising:

determining a mismatch of sample values obtained at each pair of frequencies;

comparing the mismatches to identify one or more noisy frequencies; and

if one or more noisy frequencies are identified, computing a touch value as a function of sample values obtained at frequencies other than the identified noisy frequencies.

2. The method of claim 1 wherein a mismatch is determined as a sum of variances of sample values obtained at each pair of frequencies.

3. The method of claim 1 wherein the plurality of touch sample values are baselined to remove variations in touch sensor gain and offset prior to determining the sum of variances of sample values obtained at each frequency.

4. The method of claim 1 wherein comparing the mismatches to identify a noisy frequency comprises comparing the determined mismatches to a predetermined threshold.

5. The method of claim 4 wherein the predetermined threshold is an adaptive threshold.

6. The method of claim 2 wherein:

determining the sum of variances of sample values obtained at each pair of frequencies further comprises;

subtracting sample values obtained at a first frequency from sample values obtained at a second frequency to obtain the sum of variances of a first frequency pair;

subtracting sample values obtained at a second frequency from sample values obtained at a third frequency to obtain a sum of variances of a second frequency pair; and

subtracting sample values obtained at a third frequency from sample values obtained at a first frequency to obtain a sum of variances of a third frequency pair; and

comparing the determined variances to identify a noisy frequency further comprises;

computing a first variance corresponding to the sum of variances of the first frequency pair;

computing a second variance corresponding to the sum of variances of the second frequency pair;

computing a third variance corresponding to the sum of variances of the third frequency pair; and

performing an analysis of variance on the computed variances.

7. The method of claim 6 wherein performing an analysis of variance on the computed variances further comprises:

identifying a minimum variance of the computed variances;

computing variance ratios of the computed variances to the minimum variance; and

comparing the variance ratios to a threshold determined from a statistical test.

8. The method of claim 7 wherein the statistical test is an F-test

9. The method of claim 7 wherein the statistical test is Levine'"'"'s test.

56. The method of claim 55 wherein the two-clean-frequency noise detection method comprises a method according to any of claims 1, 6, or 7.

65. The touch sensing device of claim 59 wherein the two-clean-frequency noise detection method comprises a method according to any of claims 1, 6, or 7.

10. A touch sensing device comprising:

a touch sensitive surface having a plurality of touch sensors;

drive circuitry configured to apply drive signals to the touch sensors wherein the drive signals comprise at least three periodic waveforms each having a unique frequency;

sense circuitry configured to derive a touch image for each frequency, each touch image comprising a plurality of touch sample values obtained at each frequency; and

additional circuitry configured to detect noise in one of the touch images by;

determining a mismatch of sample values obtained at each pair of frequencies;

comparing the determined mismatches to identify a noisy frequency; and

if a noisy frequency is identified, computing a touch value as a function of sample values obtained at frequencies other than the identified noisy frequency.

11. The touch sensing device of claim 10 wherein a mismatch is determined as a sum of variances of sample values obtained at each pair of frequencies.

12. The touch sensing device of claim 10 wherein one of the sense circuitry and the additional circuitry is configured to baseline the plurality of touch sample values obtained at each frequency by removing variations in sensor gain and offset.

13. The touch sensing device of claim 10 wherein the additional circuitry is part of the sense circuitry.

14. The touch sensing device of claim 10 wherein the additional circuitry comprises at least one microcontroller.

15. The touch sensing device of claim 10 wherein the touch sample values correspond to a capacitance measurement at each touch sensor.

16. The touch sensing device of claim 10 wherein the touch sensitive surface is part of a touch screen.

17. The touch sensing device of claim 10 wherein the additional circuitry compares the mismatches to identify a noisy frequency by comparing the mismatches to a predetermined threshold.

18. The touch sensing device of claim 17 wherein the predetermined threshold is an adaptive threshold.

19. The touch sensing device of claim 11 wherein:

the additional circuitry determines a sum of variances of sample values obtained at each pair of frequencies by;

subtracting sample values obtained at a first frequency from sample values obtained at a second frequency to obtain a sum of variances of a first frequency pair;

subtracting sample values obtained at a second frequency from sample values obtained at a third frequency to obtain a sum of variances of a second frequency pair; and

subtracting sample values obtained at a third frequency from sample values obtained at a first frequency to obtain a sum of variances of a third frequency pair; and

the additional circuitry compares the determined variances to identify a noisy frequency by;

computing a first variance corresponding to the sum of variances of the first frequency pair;

computing a second variance corresponding to the sum of variances of the second frequency pair;

computing a third variance corresponding to the sum of variances of the third frequency pair; and

performing an analysis of variance on the computed variances.

20. The touch sensing device of claim 19 wherein performing an analysis of variance on the computed variances further comprises:

identifying a minimum variance of the computed variances;

computing variance ratios of the computed variances to the minimum variance; and

comparing the variance ratios to a threshold determined from a statistical test.

21. The method of claim 20 wherein the statistical test is an F-test

22. An electronic device having as an input device a touch sensing device according to any of claims 10, 16, 17, 19, or 20.

23. The electronic device of claim 22 wherein the electronic device is selected from the group consisting of a desktop computer, a tablet computer, and a notebook computer.

24. The electronic device of claim 22 wherein the electronic device comprises at least one of a handheld computer, a personal digital assistant, a media player, and a mobile telephone.

25. A mobile telephone having as an input device a touch sensing device according to any of claims 10, 16, 17, 19, or 20.

26. A method of detecting noise in one or more of at least two touch images, each touch image comprising a plurality of touch sample values obtained by stimulating a touch sensor at a unique frequency, the method comprising:

computing a smoothed image corresponding to each frequency from each of the at least two touch images;

computing a high pass image corresponding to each frequency from the touch image at each frequency and the smoothed image at each frequency;

computing a spatial variance for each frequency from the high pass image for each frequency, the spatial variance being representative of noise energy in each of the high pass images;

computing a temporal variance representative of noise energy for each frequency as a function of time, wherein the temporal variance is a function of the computed spatial variance for each frequency and one or more prior spatial variances for each frequency; and

comparing the computed temporal variances for each frequency to identify a noisy frequency.

27. The method of claim 26 further comprising:

if no noisy frequencies are identified, computing a touch value as a function of sample values obtained at all frequencies;

if one or more noisy frequencies are identified, computing a touch value as a function of sample values obtained at all frequencies other than the one or more noisy frequencies;

if all frequencies are identified as noisy, computing a touch value as a function of sample values obtained at all frequencies.

28. The method of claim 26 wherein the smoothed image corresponding to each frequency is computed by performing an unsharp mask on a corresponding touch image.

29. The method of claim 26 wherein the high pass image corresponding to each frequency is computed by subtracting the smoothed image at each frequency from the touch image at each frequency.

30. The method of claim 26 wherein the temporal variance is directly related to a length of a curve of a plot of a plurality of spatial variances as a function of time.

31. The method of claim 26 wherein comparing the computed temporal variances for each frequency to identify a noisy frequency further comprises performing an analysis of variance on the temporal variances.

32. The method of claim 31 wherein performing an analysis of variance on the temporal variances further comprises:

identifying a minimum temporal variance of the temporal variances for each frequency;

computing variance ratios of the temporal variances for each frequency to the minimum temporal variance; and

comparing the variance ratios to a threshold determined from a statistical test.

33. The method of claim 32 wherein the statistical test is an F-test.

34. The method of claim 26 further comprising:

computing a quietness value as a function of the computed spatial variances;

comparing the quietness value to a predetermined threshold; and

if the quietness value does not exceed the predetermined threshold, determining that there is low noise on all frequencies.

35. The method of claim 34 wherein the quietness value is computed as a function of the maximum absolute deviation of the spatial variances.

57. The method of claim 55 wherein the one-clean-frequency noise detection method comprises a method according to any of claims 26, 27, 31, or 32.

66. The touch sensing device of claim 59 wherein the one-clean-frequency noise detection method comprises a method according to any of claims 26, 27, 31, or 32.

36. A touch sensing device comprising:

a touch sensitive surface having a plurality of touch sensors;

drive circuitry configured to apply drive signals to the touch sensors wherein the drive signals comprise at least three periodic waveforms each having a unique frequency;

sense circuitry configured to derive a touch image for each frequency, each touch image comprising a plurality of touch sample values obtained at each frequency; and

additional circuitry configured to detect noise in one of the touch images by;

computing a smoothed image corresponding to each frequency from each of the at least two touch images;

computing a high pass image corresponding to each frequency from the touch image at each frequency and the smoothed image at each frequency;

computing a spatial variance for each frequency from the high pass image for each frequency, the spatial variance being representative of noise energy in each of the high pass images;

computing a temporal variance representative of noise energy for each frequency as a function of time, wherein the temporal variance is a function of the computed spatial variance for each frequency and one or more prior spatial variances for each frequency; and

comparing the computed temporal variances for each frequency to identify a noisy frequency.

37. The touch sensing device of claim 36 wherein one of the sense circuitry and the additional circuitry is configured to baseline the plurality of touch sample values obtained at each frequency by removing variations in sensor gain and offset

38. The touch sensing device of claim 36 wherein the additional circuitry is part of the sense circuitry.

39. The touch sensing device of claim 36 wherein the additional circuitry comprises at least one microcontroller.

40. The touch sensing device of claim 36 wherein the touch sample values correspond to a capacitance measurement at each touch sensor.

41. The touch sensing device of claim 36 wherein the touch sensitive surface is part of a touch screen.

42. The touch sensing device of claim 36 wherein the additional circuitry is further configured to:

if no noisy frequencies are identified, computing a touch value as a function of sample values obtained at all frequencies;

if one or more noisy frequencies are identified, computing a touch value as a function of sample values obtained at all frequencies other than the one or more noisy frequencies;

if all frequencies are identified as noisy, computing a touch value as a function of sample values obtained at all frequencies.

43. The touch sensing device of claim 36 wherein the additional circuitry computes the smoothed image corresponding to each frequency by performing an unsharp mask on a corresponding touch image.

44. The touch sensing device of claim 36 wherein the additional circuitry computes the high pass image corresponding to each frequency by subtracting the smoothed image at each frequency from the touch image at each frequency.

45. The touch sensing device of claim 36 wherein the temporal variance is directly related to a length of a curve of a plot of a plurality of spatial variances as a function of time.

46. The touch sensing device of claim 36 wherein the additional circuitry compares the computed temporal variances for each frequency to identify a noisy frequency by performing an analysis of variance on the temporal variances.

47. The touch sensing device of claim 46 wherein the additional circuitry performs an analysis of variance on the temporal variances by:

identifying a minimum temporal variance of the temporal variances for each frequency;

computing variance ratios of the temporal variances for each frequency to the minimum temporal variance; and

comparing the variance ratios to a threshold determined from a statistical test.

48. The touch sensing device of claim 47 wherein the statistical test is an F-test.

49. The touch sensing device of claim 36 wherein the additional circuitry is further configured to:

compute a quietness value as a function of the computed spatial variances;

compare the quietness value to a predetermined threshold; and

if the quietness value does not exceed the predetermined threshold, determine that there is low noise on all frequencies.

50. The touch sensing device of claim 49 wherein the quietness value is computed as a function of the maximum absolute deviation of the spatial variances.

51. An electronic device having as an input device a touch sensing device according to any of claims 36, 41, 42, 46, 47, or 49.

52. The electronic device of claim 51 wherein the electronic device is selected from the group consisting of a desktop computer, a tablet computer, and a notebook computer.

53. The electronic device of claim 51 wherein the electronic device comprises at least one of a handheld computer, a personal digital assistant, a media player, and a mobile telephone.

54. A mobile telephone having as an input device a touch sensing device according to any of claims 36, 41, 42, 46, 47, or 49.

55. A method of detecting noise in one or more of at least three touch images, each touch image comprising a plurality of touch sample values obtained by stimulating a touch sensor at a unique frequency, the method comprising:

determining whether noise in any of the at least three touch images exceeds a background noise level;

if noise in any of the touch images exceeds a background noise level performing a one-clean-frequency noise detection method; and

if noise in all of the touch images does not exceed a background level or if the one-clean-frequency noise detection method finds noise in any one of the touch images, performing a two-clean-frequency noise detection method.

58. The method of claim 55 wherein determining whether noise in any of one or more touch images exceeds a background noise level further comprises:

computing a quietness value as a function of a plurality of computed spatial variances of the one or more touch images;

comparing the quietness value to a predetermined threshold; and

if the quietness value does not exceed the predetermined threshold, determining that there is low noise on all frequencies.

59. A touch sensing device comprising:

a touch sensitive surface having a plurality of touch sensors;

drive circuitry configured to apply drive signals to the touch sensors wherein the drive signals comprise at least three periodic waveforms each having a unique frequency;

sense circuitry configured to derive a touch image for each frequency, each touch image comprising a plurality of touch sample values obtained at each frequency; and

additional circuitry configured to detect noise in one of the touch images by;

determining whether noise in any of the at least three touch images exceeds a background noise level;

if noise in any of the at least three touch images does not exceed a background noise level performing a two-clean-frequency noise detection method; and

if noise in any of the at least three touch images exceeds a background noise level performing a one-clean-frequency noise detection method.

60. The touch sensing device of claim 59 wherein one of the sense circuitry and the additional circuitry is configured to baseline the plurality of touch sample values obtained at each frequency by removing variations in sensor gain and offset.

61. The touch sensing device of claim 59 wherein the additional circuitry is part of the sense circuitry.

62. The touch sensing device of claim 59 wherein the additional circuitry comprises at least one microcontroller.

63. The touch sensing device of claim 59 wherein the touch sample values correspond to a capacitance measurement at each touch sensor.

64. The touch sensing device of claim 59 wherein the touch sensitive surface is part of a touch screen.

67. An electronic device having as an input device a touch sensing device according to claim 59.

68. The electronic device of claim 67 wherein the electronic device is selected from the group consisting of a desktop computer, a tablet computer, and a notebook computer.

69. The electronic device of claim 67 wherein the electronic device comprises at least one of a handheld computer, a personal digital assistant, a media player, and a mobile telephone.

70. A mobile telephone having as an input device a touch sensing device according to claim 59.