Elan Microelectronics Corporation v. Apple, Inc.

Filing 293

Declaration of Jennifer Liu in Support of Plaintiff Elan Microelectronics Corporation's Reply to Apple, Inc.'s Opposition to Elan's Motion for Partial Summary Judgment of Infringement filed byElan Microelectronics Corporation. (Attachments: # 1 Exhibit A, # 2 Exhibit B, # 3 Exhibit C, # 4 Exhibit D, # 5 Exhibit E, # 6 Exhibit F, # 7 Exhibit G, # 8 Exhibit H, # 9 Exhibit N, # 10 Exhibit O)(Liu, Jennifer) (Filed on 6/16/2011)

EXHIBIT G

APPENDIX E

Appendix E: Tactile Sensing Mechanisms Invalidates the Asserted Claims of the ’352 Patent
Tactile Sensing Mechanisms (the “Fearing 1990 Reference”) Anticipates and/or Renders Obvious Claims 1, 7, 10,
18, 211

Claim Language

1. A method for detecting the operative
coupling of multiple fingers to a touch
sensor involving the steps of

Disclosure

The Fearing 1990 Reference discloses a method for detecting the operative
coupling of multiple objects (fingers are objects) to a touch sensor. Specifically,
the Fearing 1990 Reference discloses a capacitive touch sensor capable of sensing
multiple contacts.
See Fearing 1990 Reference at APEL0007543: “This article addresses
tactile finger design issues, the transduction mechanism, and determination
of contact and magnitude using linear filtering techniques.”
See Fearing 1990 Reference at APEL0007543: “A tactile sensor array was
packaged in a molded rubber finger tip for the Stanford/JPL hand as shown
in Figure 1. There are 7 X 12 tactile elements (tactels) on the cylindrical
portion and 1 X 12 elements underneath the hemispherical tip for a total of
8 X 12 capacitive sensing elements. Only an 8 X 8 subset (7 X 8 on
cylinder and 1 X 8 under the tip) is externally connected to the interface
electronics. The sensor density has been increased from previously
reported work (Fearing 1987a) to reduce aliasing and is now adequate for
local contact determination.”
See Fearing 1990 Reference at APEL0007548: “Figure 9 shows
interpolated strain profiles for two probes applied independently and
jointly. The sum of the individual strain profiles corresponds well to the

1

The citations to specific pages are made for exemplary purposes only. The entire reference, and not just the cited pages, disclose
the elements of the asserted claims as set forth herein.

Claim Language

Disclosure

joint strain profile. Thus superposition seems to be a valid assumption, at
least with forces less than 100 grams or so.”
See Fearing 1990 Reference at Fig. 9 at APEL0007549:

See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of
conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column. Unused rows and columns are switched to
ground potential to improve shielding and reduce cross-talk. The 8X8 array
was originally scanned at 15 Hz, but the rate was reduced to 7 Hz for
improved noise performance. This low scanning speed could be increased,
but since the analysis has been done for static forces only, it has not been a
limitation.”

2

Claim Language

Disclosure

See Fearing 1990 Reference at Fig. 2 at APEL0007544:

scanning the touch sensor to (a) identify a
first maxima in a signal corresponding to a
first finger, (b) identify a minima following
the first maxima, (c) identify a second
maxima in a signal corresponding to a
second finger following said minima, and
Apple’s and Elan’s construction:
“scanning the touch sensor”:
measuring the values generated by a touch
sensor to detect operative coupling and
determining the corresponding positions at
which the measurements are made

The Fearing 1990 Reference discloses scanning the touch sensor to identify a first
maxima in a signal corresponding to a first finger, identify a minima following the
first maxima, and identify a second maxima in a signal corresponding to a second
finger following the minima under Elan’s constructions but not under Apple’s and
the Staff’s constructions. Namely, the Fearing 1990 Reference does not disclose
the temporal order required by Apple’s and the Staff’s constructions. However,
one of ordinary skill in the art would have recognized that the Fearing 1990
Reference could be combine with either the Mehta Thesis or the ‘661 Application
to adapt the Fearing 1990 Reference to include the temporal requirement.
The Fearing 1990 Reference discloses that a straight line profile of the touch
sensor is generated.
See Fearing 1990 Reference at Fig. 9 at APEL0007549:

3

Claim Language

Disclosure

Staff’s construction:
“scanning the touch sensor”:
obtaining the values generated by a touch
sensor to detect operative coupling
Apple’s construction:
“identify”:
recognize a value to be
Staff’s construction:
“identify”:
to ascertain the origin, nature or definitive
characteristics of
Apple’s and Staff’s constructions:
“identify a first maxima in a signal
corresponding to a first finger”:
identify a first peak value in a finger profile
taken on a straight line obtained from
scanning the touch sensor

See Fearing 1990 Reference at Fig. 16 at APEL0007552:

“identify a minima following the first
maxima”:
identify the lowest value in the finger
profile taken on said straight line that occurs
after the first peak value and before another
peak value is identified

4

Claim Language

Disclosure

“identify a second maxima in a signal
corresponding to the second finger
following said minima”:
after identifying the lowest value in the
finger profile taken on said straight line,
identify a second peak value in the finger
profile taken on said straight line
Elan’s constructions:
“identify”:
establish the identity of; to ascertain the
origin, nature or definitive characteristics of
“identify a first maxima in a signal
corresponding to a first finger”:
identify a first peak value in a finger profile
taken on a straight line obtained from
scanning the touch sensor
“identify a minima following the first
maxima”:
identify the lowest value in the finger
profile taken on a straight line that occurs
after the first peak value
“identify a second maxima in a signal
corresponding to the second finger
following said minima”:

The Fearing 1990 Reference further discloses that the location of the contact is
localized to the maxima.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
Points that satisfy the following conditions are maxima, while points that only
satisfy the first derivative condition, but not the second derivative condition, are
minima.

5

Claim Language

identify a second peak value in the finger
profile taken on a straight line following the
minimum.

Disclosure

See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
The parties’ construction for “scanning the touch sensor” requires “measuring the
values generated by a touch sensor to detect operative coupling and determining
the corresponding positions at which the measurements are made.” The Fearing
1990 Reference discloses a touch sensor that is scanned to measure values to
detect the coupling of fingers to the touchpad and their respective positions.
Accordingly, the Fearing 1990 Reference discloses this limitation consistent with
parties’ construction.
See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of
conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column. Unused rows and columns are switched to

6

Claim Language

Disclosure

ground potential to improve shielding and reduce cross-talk. The 8X8 array
was originally scanned at 15 Hz, but the rate was reduced to 7 Hz for
improved noise performance. This low scanning speed could be increased,
but since the analysis has been done for static forces only, it has not been a
limitation.”
The Staff’s construction for “scanning the touch sensor” requires “obtaining the
values generated by a touch sensor to detect operative coupling.” The Fearing
1990 Reference discloses a touch sensor that is scanned to measure values to
detect the coupling of fingers to the touchpad. Accordingly, the Fearing 1990
Reference discloses this limitation consistent with the Staff’s construction.
See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of
conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column. Unused rows and columns are switched to
ground potential to improve shielding and reduce cross-talk. The 8X8 array
was originally scanned at 15 Hz, but the rate was reduced to 7 Hz for
improved noise performance. This low scanning speed could be increased,
but since the analysis has been done for static forces only, it has not been a
limitation.”
Apple's construction for “identify” requires “recognize a value to be.” The Fearing
1990 Reference discloses a logic process, described above, that determines which
points are the extrema (using the first order differential) and further identifies
which points are the maxima (using the second order differential). Points that
satisfy the first order differential condition but not the second differential order
condition are minima. Points that satisfy both conditions are maxima.
Accordingly, the Fearing 1990 Reference discloses this limitation consistent with

7

Claim Language

Disclosure

Apple’s construction.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
The Staff's and Elan’s constructions for “identify” require at least “to ascertain the
origin, nature or definitive characteristics of.” The Fearing 1990 Reference
discloses a logic process, described above, that analyzes the finger profile to

8

Claim Language

Disclosure

ascertain whether the signal profile contains maxima and minima. Accordingly,
the Fearing 1990 Reference discloses this limitation consistent with the Staff’s
construction.
Apple's and the Staff's construction for “identify a first maxima in a signal
corresponding to a first finger” requires “identify a first peak value in a finger
profile taken on a straight line obtained from scanning the touch sensor.” As
described above, the Fearing 1990 Reference discloses a touch sensor that
identifies a first maxima using a first and second order differential.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two

9

Claim Language

Disclosure

probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
Apple's and the Staff's construction for “identify a minima following the first
maxima” requires “identify the lowest value in the finger profile taken on said
straight line that occurs after the first peak value and before another peak value is
identified.” While the Fearing 1990 Reference identifies the minima, it does not
explicitly state that the minima is identified after first identifying the first maxima.
Accordingly, the Fearing 1990 Reference does not disclose this limitation under
Apple and the Staff’s construction. It would have been obvious, however, to
combine the Fearing 1990 Reference with any of the references below to adapt the
Fearing 1990 Reference to operate in the required manner.
Apple's and the Staff's construction for “identify a second maxima in a signal
corresponding to the second finger following said minima” requires “after
identifying the lowest value in the finger profile taken on said straight line,
identify a second peak value in the finger profile taken on said straight line.” As
with the previous limitation, the Fearing 1990 Reference discloses identifying a
second maxima, but does not explicitly state that the second maxima is identified
after identifying the minima. Accordingly, the Fearing 1990 Reference does not
disclose this limitation under Apple and the Staff’s construction. It would have
been obvious, however, to combine the Fearing 1990 Reference with any of the
references below to adapt the Fearing 1990 Reference to operate in the required
manner.
Elan construes “identify a first maxima in a signal corresponding to a first finger”
to mean “identify a first peak value in a finger profile taken on a straight line
obtained from scanning the touch sensor”; “identify a minima following the first
maxima” to mean “identify the lowest value in the finger profile taken on a

10

Claim Language

Disclosure

straight line that occurs after the first peak value”; and “identify a second maxima
in a signal corresponding to the second finger following said minima” to mean
“identify a second peak value in the finger profile taken on a straight line
following the minimum.” Under these constructions, the Fearing 1990 Reference
discloses the identification steps. Specifically, the Fearing 1990 Reference
discloses that the touch contacts are localized to the peak values of the force on the
touch pad.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition

11

Claim Language

Disclosure

seems to be a valid assumption, at least with forces less than 100 grams or
so.”
The above equations also solve for the minima. The first derivative solves for all
the extrema (maxima and minima), and the second derivative determines which
extrema are maxima and which are minima. Extrema satisfying the condition of
the equation are maxima and the rest are minima. Accordingly, under Elan’s
constructions, the Fearing 1990 reference discloses these limitations.
The Mehta Thesis discloses scanning the touch sensor to identify a first maxima in
a signal corresponding to a first finger, identify a minima following the first
maxima, and identify a second maxima in a signal corresponding to a second
finger following the minima under Apple’s and the Staff’s constructions. More
specifically, the Mehta Thesis discloses a touch sensor that operates in the
following manner. First, a digitized picture of the touch surface is created. The
digitized image includes the outline of the key overlay and, if multiple fingers are
present, blobs representing fingers. The digitized image is binary, meaning the
image only includes values that are minima, “0s”, and maxima, “1s.”
See Mehta Thesis at APEL0006843: “The first task at hand is that of
picture digitization. A binary digitizer has been implemented; however the
decision threshold is set by software. Three binary thresholding techniques
were considered. They are detailed in 5.1. After setting the threshold, the
image processor is directed to digitize a frame. On completion, the picture
buffer contains a digitized image of the scene. This would contain outlines
of the keys, and possibly, blobs corresponding to fingers pressed on the
overlay.”
See Mehta Thesis at APEL0006822: “However, in the present application,
image processing occurs only on ‘binary images’ - that is, outlines on an
overlay are allowed to be only black or white. Even if available, greyscale

12

Claim Language

Disclosure

images would have to be converted into binary images by setting all pixels
darker than predetermined threshold to black and the rest to white.”
Second, the background image (i.e., the image of overlay) is subtracted from the
digitized image, leaving only blobs corresponding to fingers.
See Mehta Thesis at APEL0006843: “The image is subtracted from one
containing only outlines of keys that was taken immediately after 'powerup'. The result is a picture containing only blobs corresponding to fingers.”
Third, each finger blob is reduced (a process called “thinning") to a single
representative point. The thinning process checks the frame buffer for horizontal
lines. These horizontal lines are maxima within the digitized data set.
Accordingly, the thinning process scans the frame buffer for a maximum. When a
horizontal line is encountered, the horizontal line is reduced (i.e., thinned) to a
single point. This is accomplished by replacing the horizontal line with a single
pixel at the centroid of the horizontal line. Implicit in this is that the thinning
process must recognize when the horizontal line concludes, i.e., transitions from
the maxima to a minima. Accordingly, the thinning process identifies a minima
following the maxima. In the case where another finger is present on the input
device on the same X or Y axis, the thinning process will encounter another set of
horizontal lines corresponding to the second finger. In other words, the scanning
processing encounters a second maxima following the minima.
See Mehta Thesis at APEL0006846: “The thinning process is similar to
noise reduction, and has been combined with it. The frame buffer is
scanned to check for horizontal lines. When encountered, they are replaced
by single pixels at the centre of the line. Hence a blob, which is essentially
a set of horizontal lines, is reduced to an approximately vertical line. This
line is further reduced to a single centroid point by another pass over the
raster, this time checking for vertical lines. Hence, on exit from this

13

Claim Language

Disclosure

process, all blobs on the raster are thinned down to single pixels close to, or
exactly on, their centroids. This ‘centroid’ information is stored in a CG
file for subsequent use by the tracking and code output subroutines.”
See Mehta Thesis at Figs. 6.15-6.16 at APEL0006862 depicting two
contact points:

The Apple’s construction for “scanning the touch sensor” requires “measuring the
values generated by a touch sensor to detect operative coupling and determining
the corresponding positions at which the measurements are made.” The Mehta
Thesis discloses that a video camera takes a snapshot of the surface and the output

14

Claim Language

Disclosure

is fed into an analog-to-digital converter which creates a digital image of the scene
as viewed by the camera. This digital image is a set of values generated by the
touch sensor to detect the operative coupling and position of fingers or other
objects. Accordingly, the Mehta Thesis discloses scanning the touch sensor
consistent with Apple’s construction.
See Mehta Thesis at APEL0006822: “The TV Camera video output is fed
into an analog-to-digital converter in the attached computer which creates a
digital image of the scene as viewed by the camera. Assuming standard
video rates, the active view time per line is about 53 microseconds. It is
also assumed that the outline resolution of the interface is about 2 mm., and
the overlay is 50 cm. long x 20 cm. wide. Under these conditions, the
number of points sampled per line (called pixels) must be 500 mm./2 mm.
= 250. The time per pixel equals is /us/250 = 212 ns. This implies that the
conversion rate of the A/D converter must be 1/212 ns. or about 5
megasamples per second. Commercial ‘Flash’ converters capable of
digitizing analog signals at speeds of 5 megasamples per second are
prohibitively expensive. The reason is that they are designed to resolve the
analog signal into 4-8 bits of digital data per pixel, which requires
extensive hardware. However, in the present application, image processing
occurs only on ‘binary images’ - that is, outlines on an overlay are allowed
to be only black or white. Even if available, greyscale images would have
to be converted into binary images by setting all pixels darker than
predetermined threshold to black and the rest to white.”
See Mehta Thesis at APEL0006843: “The image is subtracted from one
containing only outlines of keys that was taken immediately after 'powerup'. The result is a picture containing only blobs corresponding to fingers.”
The Staff’s construction for “scanning the touch sensor” requires “obtaining the
values generated by a touch sensor to detect operative coupling.” As explained

15

Claim Language

Disclosure

above, the Mehta Thesis discloses that a video camera takes a snapshot of the
surface and the output is fed into an analog-to-digital converter which creates a
digital image of the scene as viewed by the camera. This digital image is a set of
values generated by the touch sensor to detect the operative coupling of fingers or
other objects. Accordingly, the Mehta Thesis discloses scanning the touch sensor
consistent with the Staff’s construction.
Apple's construction for “identify” requires “recognize a value to be.” The Mehta
Thesis discloses that the thinning process scans for horizontal lines in the digitized
image. In other words, the thinning process evaluates each pixel and recognizes
whether a pixel represents a maxima or not. Accordingly, the Mehta Thesis
discloses “identify” consistent with Apple's construction.
The Staff's construction for “identify” requires “to ascertain the origin, nature or
definitive characteristics of.” The Mehta Thesis discloses that the thinning
process scans for horizontal lines in the digitized image. In other words, the
thinning process evaluates each pixel to ascertain whether the pixel is a maxima or
a minima. Accordingly, the Mehta Thesis discloses “identify” consistent with the
Staff's construction.
Apple's and the Staff's construction for “identify a first maxima in a signal
corresponding to a first finger” requires “identify a first peak value in a finger
profile taken on a straight line obtained from scanning the touch sensor.” Apple's
and the Staff's construction for “identify a minima following the first maxima”
requires “identify the lowest value in the finger profile taken on said straight line
that occurs after the first peak value and before another peak value is identified.”
Apple's and the Staff's construction for “identify a second maxima in a signal
corresponding to the second finger following said minima” requires “after
identifying the lowest value in the finger profile taken on said straight line,
identify a second peak value in the finger profile taken on said straight line.” The
Mehta Thesis discloses a thinning process that traverses each row of pixels

16

Claim Language

Disclosure

searching for horizontal lines (the process also occurs vertically). When a
horizontal line is encountered (i.e., a series of maxima is identified), it is reduce to
a single pixel. Implicit in this is the recognition of where the horizontal line ends.
In other words, the thinning process first identifies the horizontal line (i.e.,
maxima) and then subsequently identifies where the horizontal line ends, which is
signified by the presence of a minimum. After identifying a first horizontal line
(i.e., a first maxima) and a minima following that first horizontal line (i.e., where
the horizontal line terminates), the thinning process continues searching for any
other horizontal lines (i.e. additional maxima). Accordingly, when two fingers are
present on the same axis (i.e., finger profile), a first maxima will be identified in
the profile, followed by a minima, and then a second maxima. Therefore, the
Mehta Thesis discloses max/min/max terms consistent with Apple's and the Staff's
construction.
See Mehta Thesis at Figs. 6.15-6.16 at APEL0006862 depicting two
contact points:

17

Claim Language

Disclosure

See Mehta Thesis at APEL0006846: “The thinning process is similar to
noise reduction, and has been combined with it. The frame buffer is
scanned to check for horizontal lines. When encountered, they are replaced
by single pixels at the centre of the line. Hence a blob, which is essentially
a set of horizontal lines, is reduced to an approximately vertical line. This
line is further reduced to a single centroid point by another pass over the
raster, this time checking for vertical lines. Hence, on exit from this
process, all blobs on the raster are thinned down to single pixels close to, or
exactly on, their centroids. This ‘centroid’ information is stored in a CG
file for subsequent use by the tracking and code output subroutines.”

18

Claim Language

Disclosure

The ‘661 Application discloses scanning the touch sensor to identify a first
maxima in a signal corresponding to a first finger, identify a minima following the
first maxima, and identify a second maxima in a signal corresponding to a second
finger following the minima under Apple’s and the Staff’s constructions.
Specifically, the ‘661 Application discloses that when a “switch” is contacted, a
switch detector transmits an interrupt to a microprocessor. At this point, each line
is scanned along the X direction and the Y direction.
See ‘661 Application at Fig. 1 at APEL0059803:

See ‘661 Application at APEL0059793: “By scanning all the lines along
the x-axis and by reading a y-axis line every time, in such a manner, data

19

Claim Language

Disclosure

for all intersections in negative logic relative to ON-switches are obtained.”
The data is then projected onto the X and Y axes creating two finger profiles taken
on straight lines (X and Y).
See ‘661 Application at Fig. 5 at APEL0059803:

See ‘661 Application at APEL0059793: “If it is determined by the empty
data judgment part (11) that there are some data, [the flow] jumps to an xaxis projection data maker (11), coordinate data are projected on the x-axis
as shown in Figure 5, and x-axis projection data (23) are obtained.”
Once the finger profiles are derived, logic determines the number of fingers by
identifying the number of maxima in the profile. Specifically, the logic looks for a
first maxima (“1”), followed by a minima (“0”), followed by a second maxima
(“1”). In the case where there are three or more fingers, the logic is capable of

20

Claim Language

Disclosure

identifying these additional fingers. Applying the logic to the x axis of Figure 5, a
“1” is seen for the maxima 23 in the signal. Maxima 23 corresponds to finger 21
on touch pad 1. The logic then looks for a minima following maxima 23, which is
where the signal drops to “0.” The logic then looks for another “1,” which in
Figure 5 is the maxima at 25 that corresponds to finger 22 on touch pad 1.
See ‘661 Application at APEL0059793: “First, when considered in positive
logic, a determination as to a division is carried out, first, by finding a point
where a change occurs from a ‘1’ to ‘0’ which is at the end of the first data
group and, then, by finding a point where a change occurs from ‘0’ to ‘1’
which is at the beginning of the second data group. The point of division is
set at the above-stated point of change from ‘0’ to ‘1.’”
Apple’s construction for “scanning the touch sensor” requires “measuring the
values generated by a touch sensor to detect operative coupling and determining
the corresponding positions at which the measurements are made.” The ‘661
Application discloses that a touch pad is scanned to measure values to detect the
coupling of fingers to the touchpad and their respective positions. Accordingly,
the ‘661 Application discloses this limitation consistent with Apple’s construction.
See ‘661 Application at APEL0059793: “By scanning all the lines along
the x-axis and by reading a y-axis line every time, in such a manner, data
for all intersections in negative logic relative to ON-switches are obtained.”
The Staff’s construction for “scanning the touch sensor” requires “obtaining the
values generated by a touch sensor to detect operative coupling.” The ‘661
Application discloses that the touch pad is scanned to measure values to detect the
coupling of fingers to the touchpad. Accordingly, the ‘661 Application discloses
this limitation consistent with the Staff’s construction.
See ‘661 Application at APEL0059793: “By scanning all the lines along
the x-axis and by reading a y-axis line every time, in such a manner, data

21

Claim Language

Disclosure

for all intersections in negative logic relative to ON-switches are obtained.”
Apple's construction for “identify” requires “recognize a value to be.” The ‘661
Application discloses a logic process, described above, that recognizes that “1s”
correspond to maxima in the data set and “0s” correspond to minima.
Accordingly, the ‘661 Application discloses this limitation consistent with Apple’s
construction.
See ‘661 Application at APEL0059793: “First, when considered in positive
logic, a determination as to a division is carried out, first, by finding a point
where a change occurs from a ‘1’ to ‘0’ which is at the end of the first data
group and, then, by finding a point where a change occurs from ‘0’ to ‘1’
which is at the beginning of the second data group. The point of division is
set at the above-stated point of change from ‘0’ to ‘1.’”
The Staff's construction for “identify” requires “to ascertain the origin, nature or
definitive characteristics of.” The ‘661 Application discloses a logic process,
described above, that analyzes the finger profile to ascertain whether the signal
profile contains maxima (“1s”) and minima (“0s”). Accordingly, the ‘661
Application discloses this limitation consistent with the Staff’s construction.
See ‘661 Application at APEL0059793: “First, when considered in positive
logic, a determination as to a division is carried out, first, by finding a point
where a change occurs from a ‘1’ to ‘0’ which is at the end of the first data
group and, then, by finding a point where a change occurs from ‘0’ to ‘1’
which is at the beginning of the second data group. The point of division is
set at the above-stated point of change from ‘0’ to ‘1.’”
Apple's and the Staff's construction for “identify a first maxima in a signal
corresponding to a first finger” requires “identify a first peak value in a finger
profile taken on a straight line obtained from scanning the touch sensor.” Apple's
and the Staff's construction for “identify a minima following the first maxima”

22

Claim Language

Disclosure

requires “identify the lowest value in the finger profile taken on said straight line
that occurs after the first peak value and before another peak value is identified.”
Apple's and the Staff's construction for “identify a second maxima in a signal
corresponding to the second finger following said minima” requires “after
identifying the lowest value in the finger profile taken on said straight line,
identify a second peak value in the finger profile taken on said straight line.” As
described above, the ‘661 Application discloses that the touch sensor is scanned to
create a finger profile taken on a straight line. Logic then identifies a first “1” (i.e.,
maxima) in the profile. The logic then looks for a “0” (i.e., minima) following the
first “1” (i.e., first maxima), but before identifying the second “1” (i.e., second
maxima) that corresponds to the second finger. Subsequently, the logic identifies
a second “1” (i.e., second maxima) in the signal. Accordingly, the ‘661
Application discloses this limitation consistent with Apple’s and the Staff’s
construction.
See ‘661 Application at Fig. 5 at APEL0059803:

23

Claim Language

Disclosure

See ‘661 Application at APEL0059793: “First, when considered in positive
logic, a determination as to a division is carried out, first, by finding a point
where a change occurs from a ‘1’ to ‘0’ which is at the end of the first data
group and, then, by finding a point where a change occurs from ‘0’ to ‘1’
which is at the beginning of the second data group. The point of division is
set at the above-stated point of change from ‘0’ to ‘1.’”
providing an indication of the simultaneous
presence of two fingers in response to
identification of said first and second
maxima.
Apple’s and Staff’s constructions:
“in response to”:

The Fearing 1990 Reference discloses that an indication of the presence of two
objects is provided in response to the identification of said first and second
maxima under the parties’ and the Staff’s constructions. Specifically, the Fearing
1990 Reference discloses that the extrema, and in particular the maxima, are
identified using the following equation.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and

24

Claim Language

after and in reaction to

Elan’s constructions:
“in response to”:
plain meaning

Disclosure

magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
As disclosed in Fearing 1990 Reference, the number of maxima corresponds
directly to the number of objects on the touch sensor. Accordingly, the solution to
the equations (i.e., which points are the maxima), is the indication of the presence
of two contacts on the touch sensor (in the case where there are just two contacts).
Apple’s and the Staff’s construction for “in response to” requires “after and in
reaction to.” The Fearing 1990 Reference discloses that if it is determined that

25

Claim Language

Disclosure

there are two maxima, there are considered to be two objects. Accordingly, the
Fearing 1990 Reference discloses this limitation consistent with Apple’s and the
Staff’s construction.
Elan contends that the construction for “in response to” is the plain meaning. I
understand the plain meaning of “in response to” to be the construction set forth by
Apple and the Staff.
6. The method of claim 1 wherein said
touch sensor includes a plurality of lines,
said maxima being a largest local variation
in a signal value on one of said lines due to
capacitive coupling of a finger.

The Fearing 1990 Reference discloses a capacitive touch pad with a plurality of
lines, the maxima being a largest local variation in a signal value on one of said
lines due to capacitive coupling resulting from an object.
See Fearing 1990 Reference at Fig. 2 at APEL0007544:

See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of

26

Claim Language

Disclosure

conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column.”
The Fearing 1990 Reference further discloses that the maxima is the largest local
variation in a signal value on one of said lines due to capacitive coupling of a
contact object.
See Fearing 1990 Reference at Fig. 9 at APEL0007549:

See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of
conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column.”
See Fearing 1990 Reference at APEL0007545:

27

Claim Language

7. The method of claim 6 wherein said
maxima are peaks.

Disclosure

The Fearing 1990 Reference discloses that the maxima identified in the profile are
peaks.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
See Fearing 1990 Reference at Fig. 9 at APEL0007549:

28

Claim Language

Disclosure

See Fearing 1990 Reference at Fig. 16 at APEL0007552:

29

Claim Language

Disclosure

10. The method of claim 1 further
comprising the step of: detecting a distance
between said first and second maxima.

The Fearing 1990 Reference does not disclose that the distance between the first
and second maximum is detected. However, it would have been obvious to one of
ordinary skill in the art for the reasons discussed in my report to combine the
Fearing 1990 Reference with the Mehta Thesis.
The Mehta Thesis discloses that the distance between the first and second
maximum is detected. Specifically, the Mehta Thesis discloses that several tests
were run that considered the distance between touches and the ability of the system
to discern multiple objects where the objects were in close proximity to each other.
See Mehta Thesis at APEL0006858: “Finger - Finger Separation
Threshold: The last parameter determined was the minimum distance
allowed between two fingers simultaneously placed on an overlay before
they were mistaken for a single finger.”

30

Claim Language

Disclosure

See Mehta Thesis at APEL0006858: “It will be observed that 1mm
separation causes the image of the metal piece to be reduced to single
point. The vertical line compression routine will only accept single dark
pixels if they are separated from another dark pixel(s) by at most one white
pixel.”
18. A touch sensor for detecting the
operative coupling of multiple fingers
comprising:
means for scanning the touch sensor to (a)
identify a first maxima in a signal
corresponding to a first finger, (b) identify a
minima following the first maxima, (c)
identify a second maxima in a signal
corresponding to a second finger following
said minima, and
Apple’s and Elan’s constructions:
“means for scanning the touch sensor”:

The Fearing 1990 Reference discloses this limitation. See claim 1, above.2

The Fearing 1990 Reference discloses this limitation under Elan’s construction but
not under either Apple’s or the Staff’s construction. Namely, the Fearing 1990
Reference does not disclose the temporal order required by Apple’s and the Staff’s
construction nor does it disclose the structure or equivalents thereto required by
the Staff. However, one of ordinary skill in the art, for the reasons discussed in my
report, would have known to combine the Fearing 1990 Reference with either the
Mehta Thesis or the ‘661 Application to render the claim obvious under Apple’s
constructions.3 The combination of the Fearing 1990 Reference with the Mehta
Thesis or the ‘661 Application, does not, however, disclose this limitation under

2

Claim 18 and its asserted dependent claims mirror independent claim 1 and its dependents. Accordingly, any discussion of claim
1 and its dependents is understood to apply to claims 18 and its dependents, including the discussion of the parties’ and the Staff’s
respective claim constructions for terms that appear in both claim 1 and its dependents and claim 18 and its dependents.
3

For a discussion of the requirements of this limitation beyond the means plus function language, see claim 1.

31

Claim Language

This limitation is governed by 35 U.S.C. §
112(6).
The recited function is scanning the touch
sensor.
The corresponding structure is an analog
multiplexer, a circuit to measure changes in
capacitance of sensor conductors, an analog
to digital converter, a microcontroller, and
equivalents thereof.
Staff’s constructions:
“means for scanning the touch sensor”:
This limitation is governed by 35 U.S.C. §
112(6).
The recited function is scanning the touch
sensor to (a) identify a first maxima in a
signal corresponding to a first finger, (b)
identify a minima following the first
maxima, and (c) identify a second maxima
in a signal corresponding to a second finger
following said minima.

Disclosure

the Staff’s constructions.
The parties agree that “means for scanning the touch sensor” is governed by 35
U.S.C. § 112(6) and the function is scanning the touch sensor. 4 The Fearing 1990
Reference discloses a device which performs this function.
See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of
conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column. Unused rows and columns are switched to
ground potential to improve shielding and reduce cross-talk. The 8X8 array
was originally scanned at 15 Hz, but the rate was reduced to 7 Hz for
improved noise performance. This low scanning speed could be increased,
but since the analysis has been done for static forces only, it has not been a
limitation.
The parties agree that the corresponding structure is an analog multiplexer, a
circuit to measure changes in capacitance of the sensor conductors, an analog to
digital converter, a microcontroller, and equivalents thereof.
The Fearing 1990 Reference discloses an analog multiplexer and a circuit to
measure changes in the capacitance.
See Fearing 1990 Reference at Fig. 2 at APEL0007544:

4

I understand that a prior art reference discloses a means plus function limitation if it performs the same function and discloses the
corresponding structure or an equivalent thereof.

32

Claim Language

Disclosure

The corresponding structure is an analog
multiplexer, a circuit to measure changes in
capacitance of sensor conductors, an analog
to digital converter, a microcontroller, the
algorithms at 400-440 in Fig. 5, the
algorithms at 200-278 in Fig. 6-1, and
equivalents thereof.

See Fearing 1990 Reference at APEL0007544: “As shown in Figure 2,
capacitors are formed at the intersection of rows and columns of
conductive strips. As in the method used by Siegel (1986), the capacitance
at a junction is measured by the amplitude of the output voltage for a
selected row and column. Unused rows and columns are switched to
ground potential to improve shielding and reduce cross-talk. The 8X8 array
was originally scanned at 15 Hz, but the rate was reduced to 7 Hz for
improved noise performance. This low scanning speed could be increased,
but since the analysis has been done for static forces only, it has not been a
limitation.
Simplicity of construction was an issue; thus the electronics are mounted
remotely from the sensor at the base of the hand, at a distance of about 30
cm. While performance would be significantly improved by having the

33

Claim Language

Disclosure

electronics built into the finger tip, it was not felt that the sensor design was
stable enough for the considerable cost of installing electronics internally.
A significant fraction of the signal is thus lost in the shielding capacitance.
The large capacitance of this sensor allows significant cable capacitance.
Two shielded cables are required to reduce coupling, one for driving the
array and one for sensing the array output. Improvements in sensitivity and
reduced wiring constraints could be accomplished by a custom hybrid
circuit embedded in the core of the finger.”
One of ordinary skill in the art would further recognize that an A/D converter and
a microcontroller are being used to derive and calculate the formulas discussed at
APEL0007545 and particularly to allow plotting of the following figures.
Accordingly, the Fearing 1990 Reference disclose all the structure required by the
parties.
See Fearing 1990 Reference at Fig. 9 at APEL0007549:

34

Claim Language

Disclosure

See Fearing 1990 Reference at Fig. 16 at APEL0007552:

35

Claim Language

means for providing an indication of the
simultaneous presence of two fingers in

Disclosure

The Staff disagrees with the parties’ understanding of the function and
identification of corresponding structure. The Staff understands the function to be
scanning the touch sensor to (a) identify a first maxima in a signal corresponding
to a first finger, (b) identify a minima following the first maxima, and (c) identify a
second maxima in a signal corresponding to a second finger following said
minima, in that order. The Fearing 1990 Reference does not perform the claim
limitations in order. Furthermore, the additional structure required by the Staff
includes the algorithm in Figs. 5 and 6. The Fearing 1990 Reference does not
disclose this required structure literally or by equivalence thereto.
The Fearing 1990 Reference discloses this limitation under the parties’
construction but does not disclose this limitation under the Staff’s construction.

36

Claim Language

response to identification of said first and
second maxima.
Apple’s constructions:
“means for providing an indication”:
This limitation is governed by 35 U.S.C. §
112(6).
The recited function is providing an
indication of the simultaneous presence of
two fingers in response to identification of
said first and second maxima.
The corresponding structure is the algorithm
found in Fig. 8-1, which sets a finger value
equal to two after determining if a scan in
either the X direction or the Y direction has
detected two fingers.
Staff’s constructions:
“means for providing an indication”:
This limitation is governed by 35 U.S.C. §
112(6).
The recited function is providing an
indication of the simultaneous presence of
5

Disclosure

Namely, the Fearing 1990 Reference does not disclose the structure or equivalents
thereto required by the Staff.5
The parties agree that “means for providing an indication” is governed by 35
U.S.C. § 112(6) and the function is providing an indication of the simultaneous
presence of two fingers in response to identification of said first and second
maxima. Apple’s understanding of the corresponding structure requires the
algorithm found in Fig. 8-1 (specifically, block 860), which sets a finger value
equal to two after determining if a scan in either the X direction or the Y direction
has detected two maxima. The Fearing 1990 Reference discloses this structure for
performing the required function (see claim 1 above for discussion of Fearing
1990 Reference’s disclosure of the function). Specifically, the Fearing 1990
Reference discloses that the first and second order derivate of the profile is taken
to identify the number of objects on the touch sensor.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach (1985). A more general approach
when there is no tangential force at the surface and the applied pressure is
an even, single maximum function, is to find the peak strain by
interpolation.”
Points that satisfy the following conditions are maxima, while points that only
satisfy the first derivative condition are minima.
See Fearing 1990 Reference at APEL0007558:

For a discussion of the requirements of this limitation beyond the means plus function language, see claim 1.

37

Claim Language

Disclosure

two fingers in response to the identification
of said first and second maxima.
The corresponding structure is a
microcontroller programmed as shown in
Fig. 5 (items 450-540) or as shown in Fig.
8-1 (item 850) to Fig. 8-2 (915).
Elan’s constructions:
“means for providing an indication”:
This limitation is governed by 35 U.S.C. §
112(6).
The function is providing an indication of
the simultaneous presence of two fingers in
response to identification of said first and
second maxima.
The corresponding structure is firmware or
software that provides data indicating the
presence of two fingers in response to the
identification of two maxima and
equivalents thereof.

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
Elan agrees with Apple regarding the function required by the limitation, but
understands the corresponding structure to be firmware or software that provides
data indicating the presence of two fingers in response to the identification of two
maxima and equivalents thereof. The Fearing 1990 Reference discloses firmware
or software for performing the required function (see claim 1 above for discussion
of the Fearing 1990 Reference’s disclosure of the function). Specifically, the
Fearing 1990 Reference, using a software algorithm, recognizes the number of
contacts on the touch sensor.
See Fearing 1990 Reference at APEL0007557: “Localization of a line force
could in principle be done by solving for location, angle of force, and
magnitude of force from the strain equations at three sensors as was
proposed in Fearing and Hollerbach ( 1985). A more general approach
when there is no tangential force at the surface and the applied pressure is

38

Claim Language

Disclosure

an even, single maximum function, is to find the peak strain by
interpolation.”
Points that satisfy the following conditions are maxima, while points that only
satisfy the first derivative condition are minima.
See Fearing 1990 Reference at APEL0007558:

See Fearing 1990 Reference at APEL0007548: “If the sensor behaves as a
linear space invariant system, there are many powerful techniques from
linear system theory that can be used to analyze it. If the sensor response is
linear and obeys the principle of superposition, the linear space invariant
assumption will hold. Figure 9 shows interpolated strain profiles for two
probes applied independently and jointly. The sum of the individual strain
profiles corresponds well to the joint strain profile. Thus superposition
seems to be a valid assumption, at least with forces less than 100 grams or
so.”
21. The touch sensor of claim 18 wherein
said maxima are peaks.

See claim 7, above.

39