Elan Microelectronics Corporation v. Apple, Inc.

Filing 237

Appendix A & B [AND EXHIBITS 1-8] TO THE DECLARATION OF RAVIN BALAKRISHNAN IN SUPPORT OF APPLE INC'S OPPOSITION TO ELAN MICROELECTRONICS CORP.'S MOTION FOR PARTIAL SUMMARY JUDGMENT OF INFRINGEMENT OF U.S. PATENT NO. 5,825,352 filed byApple, Inc.. (Attachments: # 1 Appendix B, # 2 Exhibit 1, # 3 Exhibit 2, # 4 Exhibit 3, # 5 Exhibit 4, # 6 Exhibit 5, # 7 Exhibit 6, # 8 Exhibit 7, # 9 Exhibit 8)(Greenblatt, Nathan) (Filed on 6/2/2011)

Exhibit 4

United States Patent

[11]

[19]

Evans
[54]

[45]

Inventor:

John W. Evans, La Jolla, Calif.

[73]

Assignee:

Mar. 22, 1988

Primary Examiner-Gerald L. Brigance
Assistant Examiner-Mahmoud Fatahi-yar .
Attorney, Agent, or Firm-Edward W. Callan

CAPACITANCE·VARIATION·SENSITIVE
TOUCH SENSING ARRAY SYSTEM

[75]

4,733,222

Patent Number:
Date of Patent:

Integrated Touch Arrays, Inc., La
Jolla, Calif.

[57]

ABSTRACT

OTHER PUBLICATIONS

A capacitance-variation-sensitive touch sensing array
system includes an array of electrodes where each electrode is a connected series of conductive tabs and forms
either a row or a column ofthe electrode array; an array
of drive lines where each drive line is capacitively cou:
pled to a plurality of the electrodes; a drive signal generator for applying alternating signal packets to the
drive lines; an array of sense lines where each sense line
is capacitively coupled to a plurality of the electrodes so
that signals are derived from the electrodes when drive
signals are applied to the drive lines. The number of
electrodes is equal to the product of the number of drive
lines and the number of sense lines. Based on values
derived from signals on the sense lines, a microprocessor provides information associated with touch by an
operator. The array of electrodes may be configured so
as to result in a touch-actuated keyboard where the
number of keys is equal to the product of the number of
row electrodes and the number of column electrodes; or
the array of electrodes may be configured so that finger
touch position in a single large touch responsive area
can be sensed and processed for use in control applications.

Irwin-IBM Technical Disclosure Bulletin, Jan. 1983,
vol. 25, No.8, p. 4097.

45 Claims, 22 Drawing Figures

[21]

Appl. No.: 853,428
Apr. 18, 1986

[22] Filed:

Related U.S. Application Data
[63]

Continuation-in-part of 8er. No. 566,045, Dec. 27,
1983, abandoned.

[51] Int. Cl.4
[52] U.S. Cl
[58]

[56]

G06F 3/02
340/365 C; 340/365 R;
200IDIG.1
Field of Search
340/365 C, 365 R, 365 S,
3401712,365 VL; 178/18; 200IDIG. 1;
400/485, 477
References Cited
U.S. PATENT DOCUMENTS

3,921,166 11/1975 Volpe
4,305,135 12/1981 Dahl et aI
4,359,720 11/1982 Chai et aI

340/365 C
340/3658
340/365 C

49

RESET

51

r--

·48

-~

PRO~~~~OR

l-.
:

17
DRIVE SIGNAL
GENERATOR

COLUMN DRIVE
LINES
DL5-B

ROW DRIVE LINES DLI-4

!

-J

~
49)

RESET

r-~

AID

45

16

,

I

CONTROL
LOGIC

41 J

~

51,

~

,48

~-~

I
DATA
:
:L...
PROCES SOR .JI

42
DRIVE SIGNAL
17
PATTERN
DRIVE SIGNAL
GEN ERATOR
DRIVE SIGNAL
TIMING

ft

a
~

J~

Iooool

~

00

rESLC 1-64
f'

ELECTRODE-TO-SENSE LINE

r

CAPACITORS
tSENSE
LINES
SL 1-8

10
f\

I

~

ROW ELECTRODES E

FIG. I

-

DLEC 1-64

....---.-1 COLUMN ELECTRODES E 33-64
ROW ELECTRODES E 1-32

Iooool

8,

TOUCH
SENSING
ARRAY
COLUMN
'f.---I-"'-ELECTRODES
E33-64

Iooool

N

COLUMN DR IVE
LI N ES
DL 5-8

------n

11

--1 DRIVE L1NE-TO- ELECTRODE

v

CAPAC ITORS

f\

11
ROW DRIVE

L1NES;~L 1-4

~
-..

~

w
-..

~

u.s. Patent

DLEC 1

-

DL I
DL 2

)

4~ DL,EC

-

DL 3

lj~

i--

---

DLEC 3

4~

DL4

DLEC

41
DL 5

cIDLEC

-q:

DL6

33
DLEC34

DLEC35

DL8

ES LC

SL2

SL5
SL6
SL7
SL8

lf~

'Y-

~~

lrt

AS I.C 3)

_. -

r-rOD<

I

a~

I

I

I
I
1

1

I

-- ,- ---- --I

I

A4 I r<!D-

I
I

I
I

-- -- --- ----+

I

-}- I
I

_L._
I

I

TOUCH
SENSING
ARRAY 10
TC

ES LC 64

-+-+-+-1--

-

-

ES

--

ES

I

I

- ---= -::- -:.:- -,
I

---

I

I

I

-----

I

1r<J1>

I

r<J(>

I
I

: I
I

I
I

~r----:---?'7

E31

E64

FIG. 2

ES LC4

-

lit lt~
ES LC6 ~ ES LC 63
I
I

~~

.

-In:

-

r

E LC2

If.- lti- ~t"

---

I

E30

ES LC I,

ESLJC61

I

E29

4~

ES LC 37 ES LC 39
---

SL4

E3

4~ DlfC
64

~

ES LC 34

SL3

E2

-~~~LEC62 63
DLEC

LC 35
I ES LC 36
~~ ~ :4'1- ~~ AS L~ 38 ES l;C40

SLI

-

DLEC 61

~~

t..~

EI

--

~~ DLEC 3-9
~~ DLEC 36 ~l DLEC 40

4'

DL7

DLEC 37
4~ DLEC38

4,733,222

Sheet 2 of 12

Mar. 22, 1988

I

I

I

I

I

I

u.s. Patent

Mar. 22, 1988

----o

rI

I

DLEC

r" E

1

I

Sheet 3 of 12

I

4,733,222

-

I. 1
---

:

ESLC

1-1_LLri-+:(c=:~""I-~_.
.,.---~\d.TTA~B sJ
:
\~VCG
I

SCG

......

_J._

-,--

_A._
"

FIG. 3

DIMENS ION OF TAB
COMPLEX PRINTED
ON PROTECT IVE FI LM

26,,}oo-------- -------..j

27.-'-)==========;;;1'
-

OVERLAY (OPTIONAL)

5)'---" PROTECTIVE FILM, BACK22
II--------~-- ....... PRINTED WITH INDICIA

22)
=:.L_

20)
--I:=======::J.',.==:::;=:::r--

21)

i..

- -.l~

3_)

ROW ELECTRODE TAB
SUBSTRATE
COLUMN ELECTRODE TAB

MECHANICAL SUPPORT

SHIELDING

2_4....1)

FIG. 4

~t!~33

m~

FIG. 5

SL I ESLC 35::
ESLC36
DL5
OL I OL2 OL3 OL6

SL I
SL 2

~

:l ~I ;1 1:1 ~, ;, ,:,:,

I'

I

"

''II

:I I: :
III I' ',II
II II

II

!:

'I

I: '.
II::
II
IL

1---.

II

~

EI

<--II

E2J

:

Gt-r

.

---'j
I

....-.-.

--"

~

J

Ii

I!

:1

~

L...J"" "
=._lL~~\~-i·

E29)

I II

.!

I

ll-t"
:

If ::; I: :-_JI
:: i :b.;1f{OLEC 31 E 32

LI.,,;tDL EC 32

)

II

~

00

22
~

'117!v'
:Z~
.,

...

•••

TOUCH
SENSING
ARRAY 10

:::::

,t1

:,

~

ur-

i2;!Jw

. •...

:

~--

,n---,

•

1

I

ESLC 32

AT>

\ ---.

<:'--i1

ES~C-29.:p1r=GJffiI~:: 29 E30 ~
::DLEC
ESLC 30
J II') JLI!:! DLEC 30 E31
:"I!" I,'
IbJj
ESLC 31

-->:

,

IL-:f'-!L

.

I

)
".o.o

• ••• I,~~

II I l!:__",
II'L_

II "1

~

~
....

I-

:' I'~l iriN :: !Ljf
M-~-~
I

~

(IIIIfoo..

:1

rr,-l'"""

,I 1---

~
.......

l_.,1

'I

I;;;

'I II

ESLC~~:

/ ,L/~ "Ql&iESLC~O
ES C37 E:t

=9i'IL=_~-r*: =ii'F=iic'~~;==;====:=====r==;rAII=;-:;~,....-:ti=,:iI=,~=;-LJf=r--;;Ji=rL.!i==

II II I
I, I I

~
rn

OL4

\:,

SL3~i':
',',:"
LECI
~1111--SL4~II!IIIIII, ,1""- -- 1'1 ESLC I
I
II It
'.-IhJ!!ii"\L~C 2
S L 5~1 IIU'I : 1'1 'I' ~u
III III,
S L 7 --.....! 01
""'HI
SL 8 ------tl II

l

38

I \ _\ gt~ :::
~ J:v
I' I'

I 'I I II'
,
L
I I I
1)1 lir=.-=;;-

SL6~

7

C34

~

gJ

:~~

(D

t

~

d

a

.....
~
~

N

.a;..

-..

~
~

-..

u.s. Patent

Mar. 22, 1988

DSP I
DL

DSP 3

DSP2

4,733,222

Sheet 5 of 12

DSP 4

DSP I

DSP2

DSP3

DSP 4

I

1

i
I
I

:---lI-l--T---7'--;--;--V ee
I :
I I
Vss

I

IDS~:

DL 2

I

-'

:

I

I

I

I

I

I

I

I

DL3
I

1IllllIII:
I

Vss

I

I

I,

Vee

I

-1-,..-_+'-+-~-;:--Vss

__

I , '
I

I

I

I

•

I

.:......-----<:c--'I-

DL 4

:I

1

I

I

DL5

"

I

I

I

DL6

I

::

I

I

I,

:

V

I.

?S ~L-_:-L:--I--T---j-:~---"IIij:
I

I

I

~

I
:

1
I

'IDS

I

I

I

I

1,---.JL-i-_l-...I--+-7--'_~-""""'---r---t-.....,.-Vcc

'
I

I

I

I"';

I

:
I

I

V~~

I

I

I

I

IIlI111I
IUIlll!I
:
1

Vee

I
I

Vss

I

I

I

I
I

I
I

I
I

I

I'
I

:

I':

:

I'

I

DL7

V
ee
Vss

I

I

I

I

I

,

I

I

I

,

I

I

:

I;DS:Ui

:

L

I

DL8

I

I

I

'

. , --+--+----c,---;_-;r-+--~
L

:::'::~---l---l---+-..,..---t--rl -

~

IpOWER ON, RESET

Vss

I

I

-

2)

& PROCESS SSV 1-32

J
6~
PROCESS
UNTOUCHED
STATE

3

I

Vee

£1

liN ITIALIZE

f SCAN

Vee
V
ss

FIG. 6

o
J)

+

,

YES

CHANGES
SMALL?
NO

I

7)

~
NO

VALID
I
TOUCH?

I

I

YES

I

INVALID
RESPONSE

~

FIG. 7

5)

PROCESS VALID
TOUCH

J PROCESS

I
I

u.s. Patent

Mar. 22, 1988
FROM STEP

4,733,222

Sheet 6 of 12
0

INPUT SSVI-32, STORE AS
BVI-32 AND ALSO AS RVI-32

FIG. 8

TO STEP 2

FROM STEP 1

!

r 2 .1

INPUT SSV 1- 321

~

(2.2

2.4~

RV DRIFT
COMPENSATION
DUE?

YES

_I RVk:=RVk+SIGN(SSVk-RVk)

I

FOR 1 :5 k :5 32

I

I

NO

2.5)

2.3
YES

IISSVk-RVk! > RVB
ANY I :5 k :5 32?

I RVk:= SSVk

I BRC;=O

FOR

1:5 k :5 321

J

NO
2.6

I

SSVVk: =SSVk- BVk
FOR 1:5 k:5 32
~

3.1

l

IllSSVVk
< VVB
EACH I :5 k :5 32?
NO!
TO STEP

YES CHANGE SMALL
TO STEP

CHANGE NOT SMALL
4

FIG. 9

6

~

rJ).

•
UNTOUCHED STATE
SMALL CHANGES
6.1
FROM STEP 3
BDCC:= BDCC + 2

~

~

BVk:= BVk + SIGN (SSVVk)
FOR I s k S 32

FIG. 10
NO

~
~

6.4

NO

I

t

lj

. TO STEP

2

'"

l-l

~

7.1

NO
NOT VALID TOUCH
FROM STEP 4

I
I IF

BRC:= BRC + I
BDCC > 0 THEN BDCC: =BDCC - I

I BRC = BRCL
NO

flo+.

i

NO

I

~

FIG. II

=--./

1

7.2)

BVk: = RVk
I S k S 32

I FOR

I

~ T O STEP 7

I

f.'"t

YES

? I

NO

NOT VAll D TOUCH

I

-:l

8,
l-l

N

7.3

FURTHER PROCESSING OF
INVALID RESPONSE

gJ
~

TO STEP

2
~

5.2
FURTHER PROCESSING
OF VALID TOUCH

I

~ TO

FIG. 12
STEP

2

~

ti

u.s. Patent

Mar. 22, 1988

4,733,222

Sheet 8 of 12

EI

E2
E3
E4

E5

FIG. 13

641
63)

, OVERLAY (OPTIONAL)
PROTECTIVE FILM, BACKPRINTED

I WITH INDICIA
54 1

"

I
I

I

I

55)

I
I
I
i

56)

ROW ELECTRODE (E 1- 32)
THIN PLASTIC FILM SUBSTRATE
COLUMN ELECTRODE (E33-64)
THIN PLASTIC FILM SUBSTRATE

60 3
58 1
57)

59)
61")

ROW DRIVE LINE OR SENSE LINE
THIN PLASTIC FILM SUBSTRATE
COLUMN DRIVE LINE OR SENSE LINE

I

MECHANICAL SUPPORT

1
62)

SHIELDING

FIG. 14

~

rn
EI~

E2""'/::

:::::::

..

: :: :::::

::

~

ft

::

a

'....---

FIG. /5 A

-------

FIG. 158

~

~'~~~::::::";'~

~

.- .
E31~:;;:::;::::::::'" - - - -------E32-,'
::: . :., ::::::::::::::::

\C
00
00

.
-

-~&~-

-_._

r

~

........~_.... .
o
&.A.-..oo...a

...

__ a

• •

-

~---.---

-

~

h

I

"

II

•

•••

•••••

I

Ii'

_.

II

~

OL2 OL3
OLI-" )) (OL4
$

~t1~~mm!!
§tg~;11ili:!

-ii· .

FIG. 15C
I

rufruf~TItU1
S!~~~ ll!11l$
•

SL6

SL7 SL8

$

~

tD

OL5
OL6
OL7
OL8

FIG. /5 D

a

\C

8,
~

N

~
....

~

N
~

u.s. Patent

Mar. 22, 1988

4,733,222

Sheet 10 of 12

VALID TOUCH
FROM STEP
5.1

!

5.2.1

k: = THE NUMBER OF THE MAXIMUM
OF ERVV 1-32
j: = k - I
m:= k+1
5.2.2
ERVVO:=O
ERVV33:=0
PREMAX: =ERVV j
MAX: =ER VVk
POSTMAX: =ERVVm
ERVV 33: = POS I T I VE PART
OF SSVVI7
5.2.3)
Y-COORDINATE:=ROUND TO NEAREST INTEGER
[4 (k+ {( POSTMAX- PREMAX) I (MAX+!POSTMAX-PREMAXI) })'3

]

5.2.4
k:=THE NUMBER OF THE MAXIMUM
OF ER VV33-64
j: = k- I
m: = k+ I
5.2.5
ERVV32: =0
ERVV 65: =0
PREMAX: = ERVVj
MAX:= ERVVk
POSTMAX: =ERVVm

5.2.6/
X-COORD INATE: = ROUND TO NEAREST INTEGER
[4 (k+{( POSTMAX-PREMAX)I (MAX+I POSTMAX-PREMAX I))

)-3]

5.2.7
ADDITIONAL PROCESSING
OF VALID TOUCH

TO ST EP

FIG.

16

2

u.s. Patent

Mar. 22, 1988

4,733,222

Sheet 11 of 12

FIG. 17
DL I

DSP I
I

DSP2

DSP3

DSP4

DSP5

DSP6

::tf----+~--i;:
:;;;;mT---+~;----7---J----+--ii-Vee

Vss
:::.D=.L..=2c...-;.:-;-----d-mm----'-!----'-t----'--t----''--+-Vee
I

:

Vss

I

I

I

~e

I

~L-_+-:-:---i---i---J..--;---'--,---i--,- Vss
.

DL 4
DL

:

I

i:
I

I

I

1

IDS

1 ------.....

'

Vee
Vss

.1"1

III,

5: :

II

I
I

DL6

Vee
Vss

I

I

j---t-I-;- Vee

:

IDS

Vss

I

Vee

I

1
:::.D:::.L..:..7-!..~ _ _...!---.---'_-t--7-_ _-=111"",1'!-_-!--..1-_--I----'-Vss

I

I

'~e
:::.D:=.L..:::8-:--i-----;--T---!-:-----;I,.--r---F=:--~;__';_Vss

I
I

FROM STEP 0

2.2
CALCULATE SSVVk
ADk: = ADk + SS VVk
FOR IS kS 48

FIG. 18
IF I SSVVk I < S8
FOR
ISkS 48 THEN SC: = SC+I
t -_ _
y_ES

-+i

ADk:=O FOR
SC:=O

NO
y ES
t -_ __ _

TO STEP 4

.... TO STEP 6

u.s. Patent

Mar. 22, 1988

Sheet 12 of 12

F/G./9

4,733,222

NO

4.3
.-----ooool SECOND ADk OVER THR 2 REQUIRED?

YES
4.5
ROWMAX 2 >THR 2" I-'-'-NO~_ _-..I

4.6
ONLY ROWMAX I

a ROWMAX

2> TOl I?

YES
VALID

ONSET

NO

4.7

NO
PRESENT? I--_ _-.I

NO

4.10

SECOND ADk OVER THR 2 REQUIRED?
YES

NO

4.12

COLMAX2> THR4? a--:-;N:..:;;O_ _--..l
YES
ONLY COlMAX 1

a

4.13

COlMAX 2>TOl2? N

'---------~YES

VAll 0 ONSET

4.14
N
PRESENT? I--....;;O_---"_...... TO STEP 7

YES
5,1
FURTH ER PROCESSING OF VALID TOUCH...--_TO STEP 2

1

4,733,222

CAPACITANCE·VARIATION·SENSITIVE TOUCH
SENSING ARRAY SYSTEM

RELAnON TO PRIOR APPLICATION
5
This is a continuation-in-part of co-pending application Ser. No. 566,045 fIled on Dec. 27, 1983, now abandoned.
BACKGROUND OF THE INVENTION
10
The present invention pertains to improvements in
capacitance-variation-sensitive touch sensing array systems.
A touch sensing array system may include an array of
individual touch responsive areas which may be oper- 15
ated as keys on a keyboard or may include a single
relatively large touch responsive area which may be
operated for control purposes as a touch pad to provide
an indication associated with a given position in the
array. In a capacitance-variation-sensitive touch sensing 20
array system, the touching of a touch responsive area
results in a variation of the capacitance between one or
more electrodes and circuit ground. This variation results in the variation of a sense signal produced on one
or more sense lines coupled to the respective electrodes. 25
In a typical capacitance-variation-sensitive touch sensing array system, each electrode is driven by an alternating current or pulse signal. Sense signals are produced on the respective sense lines coupled to the respective driven electrodes. The sense signals are pro- 30
cessed to determine whether a touch responsive area
has been touched by an operator.
A typical touch sensing array has three types of conductors: drive lines, sense lines, and electrodes.
A drive line is a conductor coupled to a drive signal 35
generator and capacitively coupled to one or more
electrodes. Typically an enlarged portion of the drive
line constitutes a plate of a drive line-to-electrode capacitor.
A sense line is a conductor coupled to a sense signal 40
detector and capacitively coupled to one or more electrodes. Typically an enlarged portion of the sense line
constitutes a plate of an electrode-to-sense line capacitor.
An electrode is a conductor capacitively coupled to 45
at least one drive line and capacitively coupled to at
least one sense line. Typically an electrode has one or
more enlarged portions or plates, hereafter called tabs,
which furnish a capacitor plate to complete a capactive
coupling between the electrode and the body of an 50
operator, and thereby to system ground, when the tab
or the area overlying the tab is touched by a fmger of
the operator. The electrode part of a drive line-to-electrode capacitor or electrode-to-sense line capacitor may
be one or more tabs of the electrode, or may be a con- 55
ductive plate of the electrode not serving as a tab.
U.S. Pat. No. 4,145,748 to Eichelberger and Butler is
relevant to the present invention. This patent describes
a touch sensing array involving drive lines, sense lines,
and electrodes. In this system, there is a unique elec- 60
trode at the intersection of each drive line and sense
line, and each electrode is associated with only one key.
The number of keys is equal to the product of the number of drive lines and the number of sense lines. Under
the direction of control logic means, voltage pulses are 65
applied sequentially to the individual drive lines and the
resulting signals on the sense lines are individually processed by an AID converter to produce a digital output

2

count uniquely associated with a pair made up of a
single drive line and a single sense line and therefore
with the unique electrode which lies at the intersection
of said single drive line and said single sense line. The
digital output associated with each electrode is compared to a "no-touch" digital count unique to that electrode which is stored in a memory and updated periodically by a count of at most one. Each comparison of a
digital count associated with a specific electrode to the
"no-touch" digital count associated with that electrode,
results in an independent determination of the presence
or absence of touch of that electrode. In the event of an
extended dwell time of operator touch, the "no-touch"
digital count becomes incorrect as an indicator of the
absence of touch for the key(s) touched. This is because
the periodic updates occur regularly, even during the
dwell period. This condition may result in "false negatives" wherein the touch array is not responsive to
operator touch of affected keys. The condition is corrected after a sufficient number of periodic updates in
the absence of operator touch.
Other relevant teachings in the prior art are described
below.
Several touch sensing array systems are described in
U.S. Pat. No. 4,233,522 to Grummer and Hendriks.
Each of these systems involves one drive line, a plurality of sense lines and a plurality of electrodes. The sense
lines are divided into row sense lines and column sense
lines and there are a number of keys equal to the product of the number of row sense lines and the number of
column sense lines. Each sense line is connected to a
unique detection circuit. A common feature of these
systems is an array of touch actuated keys. Each key is
composed in part of two tabs; each tab is a directly
touchable conductive plate. The tabs are arranged on
the top side of a substrate in a coordinate grid of rows
and columns where each row of tabs is associated with
a unique row sense line and each column of tabs is associated with a unique column sense line. Each key has
one row tab and one column tab. Actuation of a selected
key is brought about by concurrent direct touching of
the two tabs of the key. In a first and second embodiment, the tabs of each row and the tabs of each column
are electrically connected to form electrodes. This requires crossing over of selected conductor runs connecting the tabs. In the first embodiment, on the bottom
surface of the substrate, there is a single drive line
which delivers a common drive signal to each of the
electrodes; the coupling of the drive line to the electrode is capacitive and is effected by a plurality of expanded conductive plates linked to the drive line and
underlying a portion of each of the tabs of each electrode; also on the bottom surface of the substrate are a
plurality of sense lines; there is a unique sense line for
each row electrode and for each column electrode; each
sense line is capacitively coupled to a particular row or
column electrode by expanded conductive plates linked
to the sense line and underlying the tabs of the electrode. Crossing over of selected conductor runs is required on the bottom surface of the substrate. In the
second embodiment, there are no conductors on the
bottom side of the substrate and the capacitive couplings of the drive line and of the sense lines to row and
column electrodes are carried out using discrete components in an electrical circuit located separate from the
touch panel. A third embodiment resembles the second
embodiment except that there are two layers of conduc-

3

4,733,222

tive traces on the top side of the supporting substrate
and crossing over of conductive runs is not required;
these two layers of conductive traces are separated by a
thin dielectric mm; the tabs of the columns lie on the top
side of this dielectric film and are connected by conduc- 5
tive traces as in embodiments one and two; the tabs of
each row are also on the top side of the dielectric film
but are not linked to form a single conductor; each tab
in a row overlies an expanded portion of a single conductor on the bottom side of the dielectric mm and 10
thereby is strongly capacitively coupled to this single
conductor which is coupled to the common drive line
and to a unique sense line as in embodiment two. In all
three of the embodiments, the detection circuits connected to the sense lines are scanned in a sequential 15
manner to detect touch. Because of the direct touch of
the touchable plates, a capacitive coupling to ground on
the order of loo pfis brought about by operator touch.
Changes of this large magnitude can be detected by
very simple means. Apart from the scanning process, 20
each detection circuit operates independently to report
a state of touch or no touch for a particular row or
column. The logic of key touch reporting is based on
the outputs of the independent detection circuits.
In U.S. Pat. No. 4,290,052 to Eichelberger et al., a 25
physical layout of a touch sensing array is given. Touch
sensing circuitry is fabricated on a double sided printed
circuit board adhesively bonded to the bottom surface
"Mf an insulative substrate such as glass. Squarish touch
:.electrodes are formed on the top side of the printed 30
circuit board and transmitter (drive line) and receiver
(sense line) electrode tabs in registration therewith are
·formed on the bottom side of the printed circuit board.
Electrical components and interconnection leads are
fabricated as an integral part of the conductor patterns 35
on the surfaces of the printed circuit board. The number
of keys is equal to the product of the number of drive
.Jines and the number of sense lines.
In another type of touch sensing array system, the
;;;drive lines and sense lines are combined to form a plu- 40
';;,rality of drive/sense lines. A touch sensing array of this
type is described in U.S. Pat. No. 3,757,322 to Harold
Barkan et al. In this device, there is a coordinate grid
made up of row and column electrodes wherein each
electrode is coupled to a unique drive/sense line; there 45
is a key at the intersection of each row and column
electrode; and the number of keys is equal to the product of the number of row electrodes and the number of
column electrodes. The operator concurrently and directly touches a tab of a row electrode and a tab of a 50
column electrode to activ:ate a key. A further touch
sensing array having combined drive/sense lines is described in U.S. Pat. No. 4,288,786 to Ferenc Ledniczki.
In this device, there are n pairs of drive/sense lines and
2 to the nth power electrodes; each electrode is capaci- 55
tively coupled to one line in each pair of drive/sense
lines; and each electrode is associated with a unique key.
Other types of keyboards share with touch sensing
array systems the features of drive lines, sense lines, and
sense signal processing. In particular, numerous capaci- 60
tance actuated keyboards based on changes in capacitance brought about by mechanical displacement exist
in the prior art and are in widespread use. Such keyboards are disclosed, for example, in U.S. Pat. Nos.
3,921,166 to Volpe, 4,305,135 to Dahl et al., and 65
4,359,720 to Chai et al.
Prior art capacitance-variation-sensitive touch sensing array systems are further described in U.S. Pat. Nos.

4

4,290,052 to Eichelberger et ali 4,379,287 to Tyler et ali
4,305,135 to Dahl et ali 4,157,539 to Hunts et al;
3,691,555 to Looschen; 4,321,479 to Ledniczki et ali
4,103,252 to Bobick; and 3,921,167 to Fox, and in Publication No. T904,oo8 by Crouse. Capacitance-variation
tabs also are described in U.S. Pat. No. 3,492,440 to
Cerbone et al and in IBM Technical Disclosure Bulletin, Vol. 17, No.1, June 1974, page 166-7, "TouchSensing Circuit", by J. A. Williams.
SUMMARY OF THE INVENTION
The present invention provides an improved capacitance-variation-sensitive touch sensing array system
involving no moving parts for which the array may be
inexpensively fabricated in a monolithic construction
that is easy to maintain. The nature of the basic combination of elements in the system provides the following
advantages:
(i) Only a small number of connections are needed
between passive circuitry capactively coupled to an
array of electrodes and active circuitry used to detect
touch.
(ii) The array of electrodes may be configured so as to
result in a touch-actuated keyboard with each of a plurality of touch responsive areas serving as an individual
key; or the array of electrodes may be configured so
that finger touch position in a single touch responsive
area of a touch pad can be sensed and processed for use
in control applications.
(iii) Relatively few and simple parts are used to construct a large array of many more keys then ordinarily
are associated with conventional keyboards.
(iv) The touch responsive areas may be legendized by
overlays, which may be sheets of computer printout.
(v) Even with considerable stray capacitive coupling
in the sytem, sense signals produced in a complete scan
of the system may be processed as one or more groups
to accurately recognize whether variations in the sense
signals are the result of a valid touch.
(vi) Sense signals produced in the system may be
processed in such a manner as to allow portions of the
touch sensing array not in use to be covered with nonconducting materials such as papers or books.
The capacitance-variation-sensitive touch sensing
array system of the present invention includes an array
of electrodes with each electrode including one or a
connected series of conductive tabs; an array of drive
lines capacitively coupled to the electrodes; a drive
signal generator for applying drive signals to the drive
lines to drive the electrodes coupled thereto; an array of
sense lines capacitively coupled to the electrodes to
derive sense signals from the electrodes when drive
signals are applied to the drive lines; and means coupled
to the sense lines for sensing the sense signals on the
respective sense lines, with the amplitudes of the respective sense signals being dependent upon whether the
capacitance between a driven electrode coupled to the
sense line and circuit ground is varied by an operator
touching an area overlying a tab of a driven electrode;
and a signal processing system coupled to the sensing
means for processing the sense signals to provide information associated with touch by an operator. The electrodes are disposed in relation to the drive lines and the
sense lines to provide a system in which the number of
electrodes exceeds the sum of the number of drive lines
and the number of sense lines and in which the number
of keys is greater than the number of electrodes. Preferably the sense signals are converted by analog-to-digital

5

4,733,222

conversion to sense signal values for processing by the
signal processing system.
Preferably, each electrode is capacitively coupled to
only one drive line and to only one sense line; each of
the drive lines is capacitively coupled to a plurality of 5
electrodes; and each of the sense lines is capacitively
coupled to a plurality of electrodes. No given drive line
and sense line pair is connected to more than one electrode in common, and the number of electrodes is equal
to the product of the number of drive lines and the 10
number of sense lines. Thus the number of electrodes
can exceed the sum of the number of drive lines and the
number of sense lines.
The drive signal generator under the control of a
microprocessor associated control logic is adapted for 15
repetitively applying a sequence of drive signals having
different predetermined characteristics to the different
drive lines to cause differently characterized sense signals to be derived on a sense line coupled to a plurality
of different electrodes.
20
The drive signal on each drive line is a periodic occurence of an alternating signal packet that occurs either
at a time when all other drive lines have no signal packets or preferably at a time when one other drive line has
a synchronously occuring alternating signal packet of 25
opposing polarity.
Preferably, at anyone time, signals on the drive lines
consist of alternating signal packets simultaneously applied to a single pair of drive lines, with the paired
signal packets being synchronous with each other, of 30
approximately the same value, and of opposing polarities to thereby cause the amplitude of the sense signals
derived in response to said synchronous pair of drive
signal packets to be low in relation to the variation in
the amplitude of the sense signals resulting from touch 35
py an operator. This novel opposing polarity feature
enables the number of sense signals and related sense
signal processing to be reduced by fifty percent.
Preferably, balancing of the oppositely polarized
paired signal packets is achieved by selective trimming 40
of the drive line-to-electrode capacitors and/or the
electrode-to-sense line capacitors or alternatively by the
signal processing system controlling the amplitude of
one or both of the paired drive signals. The object of
this balancing is to make the amplitude of each of the 45
sense signals small in the absence of operator touch.
The touch sensing array is amenable to an inexpensive monolithic construction, wherein a first portion of
the electrode array is supported by one side of a thin
dielectric substrate and a second portion of the elec- 50
trode array is supported by the opposite side of the thin
dielectric substrate. In alternative embodiments, the
drive lines and sense lines are supported by opposite
sides of a second thin dielectric substrate that is separated by a third thin dielectric substrate from the first- 55
mentioned dielectric substrate supporting the electrodes; or the drive lines and sense lines are supported
on the same thin dielectric substrate as the electrodes. In
the latter embodiment, a drive line or sense line on one
side of the substrate is. coupled to a plurality of elec- 60
trodes on the opposite side of the substrate by a capacitor including a first plate supported by the one side of
the substrate and extending from the coupled drive line
or sense line; and a second oppositely disposed plate
supported by the opposite side of the substrate and 65
extending from each coupled electrode.
The touch sensing array system of the present invention can be configured for use as a keyboard. In a pre-

6

ferred embodiment the electrodes are arrayed in rows
and columns and each key of the keyboard overlies a
tab complex comprised of an enlarged portion, a tab, of
one row electrode and a tab of one column electrode;
and the number of keys is equal to the product of the
number of row electrodes and the number of column
electrodes. Thus the number of keys can greatly exceed
the number of electrodes. The sense signals derived
from the row and column electrodes are processed to
provide information associated with whichever key is
touched by an operator.
The touch sensing array system of the present invention also can be configured so that fmger touch position
can be sensed. In such embodiments the electrode array
is dimensioned and disposed so that an operator firmly
touching his fmger to the interior area of a single touch
responsive panel overlying and coextensive with the
electrode array necessarily varies the capacitance to
ground of a plurality of electrodes; and the signal processing system processes the sense signals to provide
information associated with the position of touch. The
processing system processes the relative magnitudes of
variations in the sense signals and performs an interpolation routine using three relative magnitudes to provide
an estimate of touch position.
In another aspect of the present invention that is
applicable in general to capacitance-variation-sensitive
touch sensing array systems and not merely to the system of the present invention having arrays of drive lines
and sense lines capacitively coupled to arrays of electrodes, sense values in digital form are provided and the
processing system processes complete sets of digital
sense values iteratively as a group to ascertain the values of variations in the digital sense values and processes the variation values to determine whether said
variations resulted from a touch responsive area being
touched by an operator. In the processing, operator
touch at a specific location is judged to be present when
the variation values match a pattern of variation values
characteristically associated with touch at that specific
location. Such pattern matching constitutes a more
powerful and versatile signal processing routine than is
found in the prior art and allows the recognition of a
valid touch response in capacitance-variation-sensitive
touch sensing array systems in which, because of stray
capacitive coupling, a significant change may occur in
sense signal values not directly associated with electrodes underlying the position of operator touch.
Preferably the sense signal values are processed independently in one or more groups. Preferably for each
group this processing proceeds in two stages. In a first,
feature extraction stage a search is made for features of
the group of variations of sense signal values indicative
of touch. A second, confirmation stage is carried out
when an affmnative determination is made in the first
stage. In the confmnation stage, tests are carried out
which confmn that a pattern of variation characteristic
of touch by an operator is or is not present. In some
embodiments, further stages of processing may be
needed to more completely characterize the response of
the touch sensing array system.
In a further aspect of the present invention that likewise. is applicable to capacitance-variation-sensitive
touch sensing array systems in general, sense signal
values in digital form are provided in accordance with
the amplitudes of the respective sense signals and the
processing system processes the sense signals iteratively
to determine whether a touch responsive area has been

4,733,222
7

8

touched by an operator by comparing the sense signal
DESCRIPTION OF THE PREFERRED
values with baseline values derived from sense signal
EMBODIMENTS
values provided when no touch responsive area is
Referring the FIG. 1, a preferred embodiment of the
touched by an operator; and revises the baseline values
in response to sensed variations in the sense signal val- 5 capacitance-variation-sensitive touch sensing array system of the present invention includes a 1024-key touch
ues which are stable over a predetermined number of
sensing array 10, constructed from thirty-two row eleccomplete scans of the touch sensing array occurring in
trodes EI-32 and thirty-two column electrodes E33-64,
the absence of valid touch by an operator. This feature
a drive line array that includes four row drive lines
allows portions of the touch sensing array not in use to
be covered with nonconducting materials, such as 10 DLI-4 and four column drive lines DL5-8, an array of
eight sense lines SLI-8, an array of sixty-four drive
books or papers, without rendering the array inoperaline-to-electrode capacitors DLECI-64, an array of
tive.
sixty-four electrode-to-sense line capacitors ESLCI-64,
Additional features of the present invention are dea multiplexer 11, an amplifier circuit 12, a synchronous
scribed in relation to the description of the preferred 15
demodulator 13, an analog-to-digital (AID) converter
embodiments.
14, a microprocessor IS, a control logic circuit 16 and a
drive signal generator 17.
BRIEF DESCRIPTION OF THE DRAWINGS
While only thirty-two row electrodes and thirty-two
FIG. 1 is a block diagram of the system of the present
column electrodes are included in this embodiment, it is
20 to be understood that with a number "n" of drive lines
invention.
FIG. 2 is a partial schematic circuit diagram showing
DL and a number "m" of sense lines SL there may be
the capacitive interconnection between the drive lines,
n Xm electrodes; and if a number "k" of the electrodes
electrodes, and sense lines and showing the tabs of the
are row electrodes, then there would be (nxm)=k
electrodes in one preferred embodiment of the system
column electrodes and k[(nxm)-k] keys. In particular,
of FIG. 1.
25 with a number n of drive lines and the same number of
FIG. 3 is a schematic circuit diagram illustrating the
sense lines, there may be n squared electrodes; and with
theory of touch detection in the touch sensing array of
half, or (n squared)/2, of the electrodes being row electhe system of FIG. 1.
trodes and with half being column electrodes, there
,~
FIG. 4 is a schematic exploded cross-sectional diamay be (n to the fourth power)/4 keys. Thus with only
-':.gram illustrating the construction of one preferred em- 30 2n connections between active circuitry and the drive
.',··bodiment of the touch sensing array of the system of
and sense lines in combination, a system of (n to the
>:FIG.l.
fourth power)/4 keys may be obtained.
FIG. 5 is a partial plan view illustrating the construcReferring to FIG. 2, illustrated schematically are tab
tion and interconnection of the drive lines, electrodes,
complexes TC, each comprised of a tab of one of the
and sense lines and the tabs of the electrodes in one 35 row electrodes EI-32 and a tab of one of the column
electrodes E33-64. Each of the electrodes EI-64 conpreferred embodiment of the system of FIG. 1.
tributes a tab to thirty-two tab complexes TC. The drive
FIG. 6 illustrates the timing and polarity of the differlines DI-8 are capacitively coupled to the electrodes
.o-ent drive signal patterns applied to the drive lines by the
EI-64 by the drive line-to-electrode capacitors DLEC:drive signal generator in one preferred embodiment of
::the system of FIG. 1.
40 1-64. The electrodes EI-64 are capacitively coupled to
:c. FIGS. 7 through 12 are flow charts of the program the sense lines SLI-8 by the electrode-to-sense line
capacitors ESLCI-64. Each electrode EI-64 is capaciused by the microprocessor to process sense signal valtively coupled to only one drive line DL and to only
ues in one preferred embodiment of the system of FIG.
one sense line SL. Each of the eight drive lines DLI-8
1.
FIG. 13 is a partial plan view illustrating the con- 45 is capacitively coupled to eight of the sixty-four electrodes EI-64; and each of the eight sense lines SLI-8 is
struction and relationship of the electrode array and the
coupled to eight of the sixty-four electrodes EI-64. In
single touch responsive area in an alternative preferred
effect each electrode EI-64 completes a T-bridge conembodiment of the system of FIG. 1.
nection between a pair made up of a unique one of drive
FIG. 14 is a schematic exploded cross-sectional view 50
lines DLI-8 and a unique one of sense lines SLI-8.
illustrating the construction of the touch sensing array
The theory of touch detection is described with referin the alternative preferred embodiment of FIG. 13.
ence to FIG. 3, wherein a tab of an electrode is shown.
FIGS. 15A through 15D illustrate the relationship
A drive line DL is capacitively coupled to the electrode
between layered arrays of electrodes, drive lines, and
E by the drive line-to-electrode capacitance DLEC;
sense lines in the alternative preferred embodiment of 55 and the electrode E is capacitively coupled to a sense
FIGS. 13 and 14.
line SL by an electrode-to-sense line capacitance
FIG. 16 is a flow chart of microprocessor programESLC. A drive signal is provided on the drive line DL
ming applicable to the further preferred embodiment of
by a drive signal source V to drive the electrode E. A
FIG. 13 that is used in addition to the programming
variable capacitance to ground (VCG) is produced in
60 accordance with the proximity to the tab of an operaillustrated by FIGS. 7-12.
FIG. 17 illustrates the timing and polarity of the
tor's finger F, as shown in the equivalent circuit 18.
different drive signal patterns applied to the drive lines
There is also a stray capacitance from the electrode E to
by the drive signal generator in a further preferred
ground as represented by the capacitor SCG.
embodiment of the system of FIG. 1.
When a drive signal (OS) consisting of an applied AC
FIGS. 18 and 19 are flow charts of microprocessor 65 voltage is placed on the drive line DL to drive the
programming applicable to the further preferred emelectrode E, a sense signal (SS) is provided on the sense
bodiment of FIG. 17 that is used in lieu of the programline SL. The amplitude of the sense signal on the sense
ming illustrated in FIGS. 8-12.
line SL is dependent upon whether the capacitance

4,733,222

9

W

between the driven electrode E and circuit ground is
varied by an operator touching the tab. When the operator touches the tab (or even moves his fmger F in close
proximity to the tab) the variable capacitance to ground
VCG increases to thereby cause the amplitude of the 5
sense signal on the sense line SL to decrease. By providing appropriate means for processing the seJ,1s~ signal
provided on the line SL, it is possible to ascertam when
the tab has been touched by an operator.
A preferred construction of the touch sensing array 10
in this embodiment is shown in FIG. 4. Each tab complex TC includes two conductive tabs 20 and 21 that a~e
respectively parts of different ~lectrode~. One ~b 20.ls
supported on one side of a thin (0.001 mch) dielectnc
Mylar fUm substrate 22 and the other tab 21 is supported 15
on the other side of the thin dielectric substrate 22.
Referring to FIG. 5, it is seen that the row electrodes
E1-32 (represented by solid lines) are supported on the
top side of the substrate 22, and that the column electrodes E33-64 (represented by broken lines) are sup- 20
ported on the bottom side of the substrate 22. Each
top-side tab 20 is an enlarged portion of a row electrode;
and each corresponding bottom-side tab 21 of the same
tab complex is an enlarged portion of a column electrode and is adjacent to the position underlying the 25
top-side tab 20 so that the capacitance between both
electrodes having the respective tabs 20 and 21 and
ground is varied whe~ the tou~h. sensing array is
touched by an operator m the proX1m1ty of the tab com·
plex comprising tabs 20 and 2 1 . .
30
Referring again to FIG. 4, t~e dielectnc .substrate 22
is supported by a nonconductive mechamcal; support
layer 23, which is covered on its bottom-side by a
grounded conductive, electricalltsheilding layer ~.
The touch sensing array further mcludes a pr~tectlve 35
0.005 inch L:xan plastic fUm layer 25 covenng the
unsuPP?rted Side of the ~ubstrate. 22 and .the. ta~ ~O. The
protective fUm layer 25 is backpnnted ~Ith indiCia ~sociated with each of the tab com~lexes m ord~r to identify the tab co~plexes ~ underlymg keys which are ~he 40
touch responsl~e ar:as m a keyb~ard. The dashed line
26 sh?ws the ~trnenslOn of an outl~e of the tab coml?lex
that IS backpnnted .on the protective ~lm 25. Optionally, the touch SenSl1:1g ~ay may' al~o. mclude a paper
ov.erlay .lay~~ 27 hav~g. dlffe~ent indiCia than the back- 45
pnnted indiCia when It IS deSired to use the touch sens~g. ~rray in an applic~tion in which the backprinted
mdicla are. not appropnate. .
FIG. 4 Illustrates the r:latlve placement of the ~onstituents of the touch sensing array only; and the thick- 50
nesses of the l~yers 22, 23, 24, 25, ~nd 27 are not pro~ortionate.. In this pref~rred ~mbod~ent th~ mecham~al
suppo~mg layer 23 IS 0.2~ Inch thi~k. It will be ~eadily
appreciated that de~rea;smg the thickness.
this supporting layer 23 will Increase the proX1m1ty of the 55
grounde~ shield~g layer 24 to each o.f electrodes E1-64
thereby mcreaslng the stray c~pacltances t~ ground
SCG of each of E1-;64: The entire. touch sens~g array
10 is p~ o~ a monolithic construc~lOn that also Includes
the dnve lme-to-electrode capacitors DLEC1-64 and 60
the electrode-to-sense line capacitors ESLC1-64.
The width of each of the respective electrodes E1-64
is very narrow (0.02 inch) in areas where a row electrode E1-32 on one side of the substrate 22 crosses the
position of a column electrode E33-64 on the opposite 65
side of the substrate 22 in order to minimize capacitive
coupling between electrodes on opposite sides of the
substrate 22.

or

Referring again to FIG. 5, both the drive lines DL1-8
and the sense lines SL1-8 are supported by the same
thin dielectric substrate 22 as supports the electrodes
E1-64 and the tab complexes TC. The row drive lines
DL1-4 (represented by broken lines) are supported on
the bottom side of the substrate 22; and the column
drive lines DL5-8 (represented by solid lines) are supported on the top side of the substrate 22.
Each of drive lines DL1-8 on one side of the substrate 22 is coupled to electrodes on the opposite side of
the substrate 22 by the drive line-to-electrode capacitors
DLECl-64. Each capacitor includes a fIrst plate supported by one side of the substrate 22 and extending
from the drive line and a second oppositely disposed
plate supported by the other side of the substrate 22 and
extending from the coupled electrode. For example,
consider drive line-to-electrode capacitor DLEC40
which includes a fIrst plate 29 supported on the top side
of the substrate 22 and extending from column drive
line DL8 on the top side of the substrate 22 and a second
oppositely disposed plate 30 supported on the bottom
side of the substrate 22 and extending from column
electrode E40 on the bottom side of the substrate 22.
The sense lines SL1-8 are supported on both sides of
the substrate 22, as may be discerned from the representative respective solid and broken lines in FIG. 5. The
row electrodes E1-32 on the top side of the substrate 22
are capacitively coupled to the portions of eight sense
lines SL1-8 on the bottom side of the substrate 22 respectively by thirty-two of the electrode-to-sense line
capacitors, ESLC1-32. Each of the eight sense lines is
capacitively coupled to four of the row electrodes. The
column electrodes E33-64 of the bottom side of the
substrate 22 are capacitively coupled to the portions of
the eight sense lines SL1-8 on the top side of the substrate 22 respectively by the remaining thirty-two electrode-to-sense line capacitors, ESLC33-64. Each. such
capacitor includes a fIrst plate supported by one Side of
the substrate 22 and extending from the electrode and a
second plate supported by the other side of the substrate
22 and extending from the coupled sense line, such as
shown for capacitors ESLC1-4. This condition is ~so
satisfIed by capacitors ESLC37-40 wherein a capacitor
plate on one side of the substrate constitutes the first
plate for several different capacitors that are defIned by
separate second plates on the opposite side of the substrate. In capacitors ESLC37-40, a first plate 34 S?Pported on the top side of the substrate 22 and extendmg
from the sense line SL2 supported on the top side of the
substrate 22 is in common to each of these capacitors
ESLC37-40, and separate second plates 37, 38, 39 and
40 are supported on the bottom side of the substrate 22
and respectively extend from the electrodes E37, E38,
E39, and E40 supported on the bottom side of the sub22
str~~ble '1 shows which one of the drive lines DL1-8
and which one of the sense lines SL1-8 are respectively
coupled to each of the electrodes E1-64. No given
drive line and sense line pair is connected to more than
1
d'
one e ectro e In common.
TABLE I

ROW
DRIVE
LINES

DLi
DL2
DL3
DL4

Electrode Number
SLi SL2 SLJ SL4 SL5 SL6
I 5 9 13 17 21
2
6 10 14 18 22
3
7 II
15
19 23
4
8 12
16 20 24

SL7 SL8
25
26
27
28

29
30
31
32

4,733,222

11

TABLE 2-continued

TABLE I-continued
Electrode Number
SLI SL2 SL3 SL4 SL5 SL6
COLUMN
DRIVE
LINES

DL5
DL6
DL7
DL8

33
34
35
36

37
38
39
40

41
42
43.
44

45
46
47
48

12

49

50

51
52

53
54
55
56

DSPI
SL7 SLS
57
58
59

61 5
62
63

60

64

DSP2

DSP3

DSP4

Location of DS and IDS
on DLl-8 during DSPI-4

During each drive signal pattern, the paired drive
signal DS and inverted drive signal IDS occur simultaneously and are synchronous with each other, of apReferring again to FIG. 1, drive signals are applied to 10 proximately the same value, and of opposing polarities.
the drive lines DLI-8 by the drive signal generator 17.
The drive signal pattern information on lines 42 indiThe drive signal generator 17 is adapted to respond to
cates the sequence in which the drive signal packets are
drive signal timing information provided on line 41
placed upon given drive lines. Each of the drive lines
from the control logic circuit 16 and to drive signal 15 DSI-8 is dedicated to receiving either drive signal DS
pattern information provided on lines 42 from the mior inverted drive signal IDS exclusively, as shown in
croprocessor 15 by repetitively applying a sequence of
FIG. 6.
drive signals having different predetermined characterReferring again to FIG. 1, the multiplexer 11 timeistics to the different drive lines DLI-8 to cause differdivision multiplexes the eight different sense signals on
ently characterized sense signals to be derived on sense 20 sense lines SLI-8 to provide time-division multiplexed
lines SLI-8 coupled to a plurality of different elecsense signals on line 43 to the amplifier 12. The amplifier
trodes EI-64. The drive signal on each drive line is a
12 is a band pass amplifier centered at the predeterperiodic occurence of an alternating signal packet that
mined frequency of the alternating signal within the
occurs at a time when one other drive line has a syn- 25 signal packets of the drive signals. The synchronous
chronously occuring alternating signal packet of opposdemodulator frequency demodulates the signal on line
ing polarity. Referring to FIG. 6, each signal packet
44 from the amplifier 12. The control logic circuit 16
contains sixteen cycles of an alternating signal that varcontrols the operation of the synchronous demodulator
ies at a predetermined frequency. At anyone time, the
13 by a signal on lines 45 to cause the demodulation of
signals on the drive lines consist of alternating signal 30 a signal packet on line 44 to be synchronous with the
generation of the corresponding drive signal packet by
packets simultaneously applied to a signal pair of drive
lines, with the paired signal packets being synchronous
the drive signal generator 17. The demodulated signals
then are converted from analog signals to digital signals
with each other, of approximately the same value, and
of opposing polarities to thereby cause the amplitude of 35 by the AID converter 14 which provides a respective
sense signal value (SSV) on lines 46 for the sense signal
the sense signals derived in response to such synchroderived from one of the sense lines SLI-8 that is senous pairs of drive signal packets to be low in relation to
lected by the multiplexer 11 during one of drive patthe variation in the amplitude of such sense signals reterns DSPI-4 (FIG. 6). The multiplexing rate is such
sulting from the keys being touched by an operator.
Further means of providing that each of the sense sig- 40 that a sense signal from only one of the sense lines
SLI-8 is provided on line 43 and subsequently demodunals is of small amplitude in the absence of operator
lated during each occurrence of a drive signal pattern
touch are described in conjunction with Table 3 and
DSP. The sense signal values indicate both the amplithereafter in conjunction with Table 6.
Each of the drive signal packets of one polarity is 45 tude and the polarity of the sense signals. Because opposing polarity drive signal patterns are provided, the
referred to as drive signal DS; and each of the drive
number of sense signal values that must be processed to
signal packets of the opposite polarity is referred to as
scan completely the touch sensing array is one-half the
inverted drive signal IDS. In the context of this Specifinumber of electrodes. Thus, for this embodiment, only
cation when no distinction is made between the drive
signal DS and the inverted drive signal IDS, the term 50 thirty-two sense signal values need be processed during
each complete scan of the touch sensing array. In this
drive signal refers to whichever type of drive signal
embodiment, the AID converter 14 is an 8-bit AID
pattern (DS or IDS) is applied.
converter yielding a number between 0 and 255. The
Referring again to FIG. 6, the drive signals are prosynchronous demodulator 13 is biased. so that a midvided on the drive lines DLI-8 by the drive signal
range sense signal value of 128 is obtained when no
generator 17 in a repeating sequence of four different 55
drive signals are applied to the drive lines, and consedrive signal patterns DSPl, DSP2, DSP3, and DSP4.
quently no drive signals are transmitted to the sense
During DSPl, a drive signal DS is provided on drive
lines, during a period of AID conversion.
line DLI and an inverted drive signal IDS is provided
on drive line DL3. Table 2 shows the pair of drive lines 60 Table 3 shows a processing sequence identification
number for each of the sense signal values SSVI-32 and
from DSI-8 receiving respectively the drive signal DS
the sign, under an appropriate choice of polarity for DS
and the inverted drive signal IDS during each of the
and IDS, of the variation in the sense signal value defour different drive signal patterns DSPI-4.
rived from each of the sense lines SLI-8 during each of
_ _ _ _ _ _ _.....:T;,,;A..;:,B;;;,L;;;;,;:;E:.,;2;;..65 the respective drive signal patterns DSPI-4 whenever a
DSPI
DSP2
DSP3
DSP4
key is touched by an operator with respect to the electrodes EI-64 that supply the tabs of the tab complex of
LINE RECEIVING DS
DLi
DL2
DL5
DL6
LINE RECEIVING IDS
DL3
OL4
DL7
DL8
the key.

4,733,222

13

14

TABLE 3
COLUMN

ROW
DSPI
SSV SIGN
SLi

I

+

SL2

2

+
+

SL3
4

+

SL5

5

+

SL6

6

+

SL7

7

+

SL8

8

+

SL4

E

DSP2
SSV SIGN

I
9
3
5
10
7
9
It
11
13
12
15
17
13
19
21
14
23
25
15
27 .
29
16
31

+
+
+
+
+
+
+
+

E

DSP3
SSV SIGN

2
4
6
8

17

+

18

+

10

19

+

20

+

21

+

22

+

23

+

24

+

12
14
16
18
20
22
24
26
28
30
32

E

SSV

DSP4
SIGN

33
35
37
39
41
43
45
47
49
51
53
55
57
59
61
63

25

+

26

+

27

+

28
29

+
+

30

+

31

+

32

+

For example, if the tab complex with a tab from row 20
electrode El and a tab from column electrode E48 is
touched by an operator, a positive variation in the sense
signal value is derived from sense line SLI during DSPI
and a negative variation in the sense signal value is
derived from sense line SL4 during DSP4 provided that 25
the polarity of drive signal DS is such that a decrease in
the amount of the drive signal DS appearing on the
appropriate sense line during a drive signal pattern results in a positive variation in the sense signal value.
During manufacture of the touch sensing array sys- 30
tem, Table 3 can be referred to in trimming drive lineto-electrode capacitors and/or electrode-to-sense line
capacitors. The object of this trimming step is to balance the transmitted drive signal DS and inverted drive
signal IDS appearing on each sense line during each one 35
of DSPI-4. In this embodiment, a sense signal value of
128 is indicative of perfect balancing of DS and IDS on
a sense line during a drive signal pattern. As an example,
if SSV15 is found to be 151 in the absence of operator
touch, trimming either the drive line-to-electrode ca- 40
pacitor DLEC28 or the electrode-to-sense line capacitor ESLC28 is seen, on consulting Table 3, to make the
value of SSV15 smaller. Balancing of the entire touch
sensing array system of this embodiment proceeds by
iterating a trimming procedure in which, for each k 45
between 1 and 32, the difference between SSVk and
128, if exceeding in magnitude a predefmed tolerance
value, is reduced by half by trimming the appropriate
DLEC and/or ESLC on consulting Table 3. Thus, the
trimming is terminated when each ofSSVI-32 is within 50
the predefmed tolerance of the target value 128 in the
absence of operator touch. An iterative method of trimming is preferred since, in the presence of stray capacitive coupling, the trimming ofDLEC and/or ESLC of
a particular electrode may produce changes in sense 55
signal values not associated with that electrode. During
manufacture, this trimming operation is carried out
when the assembly of the touch sensing array system is
sufficiently advanced so that the relations of the arrays
of conductors of the system to the protective layer 25, 60
the mechanical support 23, and the shielding layer 24 of
FIG. 4 are fixed.
An alternative method of attaining the desired balance of DS and IDS on each of the sense lines SLI-8
during each of DSPI-4 is discussed below beginning 65
with the paragraph preceding Table 6.
The microprocessor 15 processes the sense signal
value on lines 46 to provide information on lines 47

E
34
36
38
40
42
44
46
48
50
52
54
56
58
60
62
64

associated with any particular key that is touched by an
operator. In this embodiment the touch sensing array 10
is used as a keyboard for a data processor 48. Accordingly, the information provided by the microprocessor
on lines 47 constitutes a keyboard output signal to the
data processor 48.
The operation of the multiplexer 11 is controlled by
signals provided on lines 49 by the microprocessor 15.
The microprocessor also produces an initialization signal on line 50 to the control logic circuit 16, which
coordinates the operation of the synchronous demodulator 13, the AID converter 14, and the drive signal
generator 17. The operation of the microprocessor 15 is
reset by the data processor 48 by a signal on line 51.
The microprocessor 15 is programmed to process the
sense signal values (SSV) on lines 46 iteratively to ascertain the signs and values of the variations (SSVV) in
the sense signal values (SSV) and to process the sense
signal variation values (SSVV) as a group to determine
whether the variations resulted from a key being
touched by an operator. The flow charts for this program are shown in FIGS. 7 through 12.
The sense signal values are processed iteratively by
comparing the sense signal values with baseline values
derived from sense signal values provided when the
touch sensing array is untouched by an operator to
determine which one of three states is present. The first
of the three states is an UNTOUCHED state. The second of the three states is a VALID TOUCH state in
which a valid touch by an operator is present. The third
of the three states is an INVALID RESPONSE state
which occurs when neither the first state or the second
state is present.
Referring to FIG. 7, step 0 represents a power on,
reset operation; and in step 1, the variable values are
initialized. In step 2, a scan of the toueh sensing array 10
is carried out with the production of sense signal values
SSVI-32 as program input. Initial processing is carried
out. In step 3, a test is performed to determine if the
sense signal variation· values SSVVI-32 are all small. If
so, the program branches to step 6 where the UNTOUCHED state is processed. If any of SSVVI-32 is
found not to be small in step 3, the program branches to
step 4 where a test is performed to identify the presence
or absence of the VALID TOUCH state. The VALID
TOUCH state is processed in step 5. In the absence of
the VALID TOUCH state the program branches from
step 4 to step 7 where an INVALID RESPONSE is

15

4,733,222

processed. Following each of steps 5, 6, or 7 there is a
return to step 2.
Expansions of steps 1 through 7 of FIG. 7 are shown
in FIGS. 8 through 12, with a step numbering system
5
which agrees in the fIrst digit with that of FIG. 7.
Referring to FIG. 8, the initialization is shown in
steps 1.1 and 1.2 to consist of inputting the sense signal
values SSVl-32 and storing these values as baseline
values BVl-32 and also as reserve values RVl-32.
Next, a baseline drift compensation count (BDCC) and 10
a baseline revision count (BRC) are each initialized to
zero. In step 1.2 and in later steps, the "assign" symbol
:=is used in the usual way to indicate that the value to
the right of said symbol is assigned to the variable to the
left of said symbol. It is intended in operation that base- 15
line values represent the sense signal values that would
be obtained in the absence of operator touch and thus
that the touch sensing array 10 should be untouched
during initialization.
Referring to FIG. 9, the microprocessor 15 is pro- 20
grammed to process SSVl-32 to determine whether
any of the keys has been touched by an operator by
comparing SSVl-32 with the respective BVl-32 derived from the corresponding SSVl-32 provided when
the., keys are untouched by an operator to determine 25
whether any of the sense signal variation values
SSVVl-32 derived from such comparison exceeds a
predetermined variation value bound (VVB); and to
,.:',revise BVl-32 in response to sensed variations in
,SSVl-32 over a predetermined number of iterations 30
,".c, (each iteration corresponds to a sequence in which each
of drive signal patterns DSPl-4 is repeated for each of
:, the sense lines SLl-8) occurring in the absence of valid
touch by an operator. The baseline values are revised in
response to sensed variations in the sense signal values 35
which are stable over a predetermined number of iterations and which indicate the INVALID RESPONSE
state of the touch sensing array 10.
In the course of revising the baseline values BVl-32,
:.::the microprocessor 15 (a) stores a set of reverse values 40
5. RVl-32 derived from a set including each ofthe respec..;;tive sense signal values SSVl-32; (b) compares each
sense signal value to each corresponding reserve value;
(c) revises the stored set of RVl-32 to correspond to a
current set of SSVl-32 whenever the difference be· 45
tween anyone of the sense signal values and the corresponding one of the reserve values exceeds a predetermined reserve value bound (RVB); (d) processes sense
signal variation values SSVVl-32 derived from a current set of SSVl-32 whenever anyone of SSVVl-32 50
derived from such set exceeds VVB to determine
whether VVB was exceeded as a result of a key being
touched by an operator; (e) increments the baseline
revision count BRC in response to each current set of
SSVl-32 for which it is determined that the exceeding 55
of VVB was not the result of a key being touched by an
operator; (t) resets BRC to zero each time the stored set
of RVl-32 is revised; and (g) revises BVl-32 to correspond to RVl-32 when BRC equals a predetermined
baseline revision count limit (BRCL). Step (d) is shown 60
in the flow chart of FIG. 10 and steps (e) and (g) are
shown in the flow chart of FIG. 12. Accordingly, it is
seen that the baseline values are revised to correspond
to the reserve values when the INVALID RESPONSE
state is present for a predetermined number of iterations 65
during which time no revision of the reserve values
occurs, no revision of baseline values to correspond to
reserve values occurs, and the VALID TOUCH state is

16

not present. It is also seen that the baseline values are
updated by either incrementing or decrementing each
of the respective baseline values in accordance with
whether the baseline value is less or greater than the
respective last-provided sense signal value when the
UNTOUCHED state is present for a predetermined
number of consecutive iterations during which time no
updating of the baseline values occurs.
Sense signal values SSVl-32 are input in step 2.1.
Next, in step 2.2 a test is done on the baseline revision
count BRC. At certain preselected and fIxed values of
BRC there is a branch to step 2.4 where each of the
reserve values RVl-32 is altered as shown so as to
become closer by one if possible to the respective value
of SSVl-32. The SIGN function is employed in this
step and in later steps in the usual way where the SIGN
of a positive number is plus one, the SIGN of zero is
zero, and the SIGN of a negative number is minus one.
This operation of step 2.4 is called reserve value drift
compensation. Part of the signifIcance of the operation
of step 2.4 is shown by the test in step 2.3 where if any
of RVl-32 is found to depart by more than the reserve·
value bound RVB from the corresponding value among
SSVl-32 there is a branch to step 2.5 where all of
RVl-32 are given the respective values of the current
SSVl-32 and where BRC is reset to zero. In step 2.6 the
sense signal variation values SSVVl-32 are calculated
as the respective differences of SSVl-32 and BVl-32.
In step 3.1, if the magnitude of each of SSVVl-32 is
found to be smaller than the variation value bond VVB,
the touch sensing array 10 is judged to be in the UNTOUCHED state and there is a branch to step 6. Otherwise there is a branch to step 4.
Referring next to FIG. 10, in the course of processing
SSVVl-32 the microprocessor 15 (a) in a feature ex·
traction stage determines whether the maximum of the
group of variation values derived from the row electrodes El-32 (ROWMAX) exceeds a predetermined
threshold value (THR); (b) in a confIrmation stage determines whether none of the other variation values
derived from El-32 exceeds a tolerance value (TOL)
that is a predetermined fraction of ROWMAX; (c) in a
feature extraction stage determines whether the maximum of the group of variation values derived from the
column electrodes E33-64 (COLMAX) exceeds THR;
and (d) in a confIrmation stage determines whether
none of the other variation values derived from E33-64
exceeds a tolerance vlaue TOL that is a predetermined
fraction of COLMAX. All of the four immediately
preceding recited determinations (a), (b), (c) and (d)
being affIrmative indicates that the maximum variation
value ROWMAX derived from the row electrodes and
the maximum variation value COLMAX derived from
the column electrodes each resulted from only a single
one of the keys being touched instantaneously by an
operator.
Tests for a valid touch response are carried out in
steps 4.1 through 4.6. The variable ROWMAX is always assigned the value of the largest magnitude of the
current set of sense signal variation values SSVVl-16
derived from the row electrodes El-32. In step 4.1
ROWMAX is compared to the predetermined thresh·
old value THR. If ROWMAX exceeds THR there is a
branch to step 4.2 where the tolerance value TOL is set
to some fIxed fraction of the value of ROWMAX. Next,
in step 4.3 each of the magnitudes of SSVVl-16 other
than the one chosen as ROWMAX is compared to
TOL. If none of these values exceeds TOL then a

17

4,733,222

VALID ROW TOUCH response is judged to be present and the program branches to step 4.4 where the
largest magnitude COLMAX of the current set of sense
signal value variations SSVV17-32 derived from the
column electrodes E33-64 is determined. COLMAX is 5
compared to THR. If COLMAX exceeds THR, then
TOL is recalculated as a fIxed fraction of COLMAX in
step 4.5; and in step 4.6, each of the magnitudes of
SSVV17-32 other than the one chosen as COLMAX is
compared to TOL. If none of these values exceeds 10
TOL, the tests for a VALID TOUCH have been passed
and there is an exit to steps 5.1 and 5.2 where a VALID
TOUCH is processed. Failing any of the tests of steps
4.1, 4.3, 4.4, or 4.6 causes a branch to step 7 where an
INVALID RESPONSE is processed. In step 5.1 which 15
begins the processing of a valid touch, the counts
BDCC and BRC are each reset to zero. The signifIcance of these resets will be appreciated when the program flow charts shown in FIGS. 11 and 12 are described. In step 5.2 further processing of a VALID 20
TOUCH response is carried out including the appropriate encoding of the touched key by row number and by
column number which numbers may be calculated
using the information in Table 3.
It is to be understood that the separate processing of 25
sense signal variation values SSVVl-16 derived from
row electrodes and of sense signal variation values
SSVV17-32 derived from column electrodes as described in conjunction with FIG. 10 is appropriate to
the present embodiment with drive signal patterns 30
DSPl-4ofTable 2 and with the geometric layout of the
conducting arrays SLl-8, DLl-8, and El-64 of FIG. 5
where none of DLl-8 is a drive line for both a row
electrode and a column electrode. In other embodiments, the patterns of variation values characteristically 35
associated with touch may differ and in particular may
not allow separate processing of the row and column
sense signal values.
Referring to FIG. 11, the microprocessor 15·is further programmed to compensate for drift in the sense 40
signal values SSVl-32 that are provided when the
touch sensing array is in the untouched state. To effect
such compensation, the microprocessor 15 (a) increments the baseline drift compensation count BDCC in
response to each current set of sense signal values 45
SSVl-32 in which none of the sense signal variation
values SSVVl-32 exceeds the variation value bound
VVB; (b) revises the baseline values BVl-32 when the
baseline drift compensation count BDCC equals or
exceeds a predetermined baseline drift compensation 50
count limit (BDCCL) by adding to BVk the SIGN of
the SSVk for each k between 1 and 32; and (c) resets the
baseline drift compensation count BDCC to zero upon
the baseline values being revised.
The processing of the UNTOUCHED state is shown 55
in steps 6.1 through 6.4. In step 6.1 BDCC is incremented by two counts. Next, in step 6.2, BDCC is compared to the predetermined BDCCL. If BDCC has
reached or exceeded BDCCL, there is a branch to step
6.3 where BDCC is reset to zero. Next, in step 6.4 each 60
of BVl-32 is altered as shown so as to become closer by
one if possible to the respective value of the current
SSVl-32.
Referring to FIG. 12, the process of compensating for
baseline value drift further includes decrementing the 65
baseline drift compensation count BDCC to not less
than zero in response to each current set of sense signal
values SSVl-32 for which at least one of the variation

18

values SSVVl-32 exceeds the predetermined variation
value bound VVB except when it is determined that the
exceeding of VVB was the result of a key being touched
by an operator. When the INVALID RESPONSE
state occurs only intermittently, the updating of the
baseline· values is delayed by degree. In step 7.1 the
baseline revision count BRC is incremented. Next,
BDCC is decremented if BDCC is greater than zero.
Next, in step 7.2 BRC is compared to the predetermined
baseline revision count limit BRCL. If there is equality
there is a branch to step 7.4 where the baseline values
BVl-32 are assigned the respective reserve values
RVl-32, as discussed above in relation to the description of the program flow chart shown in FIG. 9. Further processing of the INVALID RESPONSE state is
carried out in step 7.3 as needed.
Referring again to the program flow chart of FIG. 7,
it is seen that baseline value drift compensation occurs
in step 6, where the UNTOUCHED state is processed,
when BDCC reaches BDCCL. BDCC is reset to zero in
step 5, incremented by two in step 6, and is decremented
by one if positive in step 7. Since there is passage
through only one of steps 5, 6, or 7 during each iteration, the condition which brings about baseline value
drift compensation is apparent. Also it is seen that complete baseline revision with the resulting replacement of
BVl-32 by RVl-32 occurs in step 7, where the INVALID RESPONSE state is processed, when BRC
reaches BRCL. BRC is incremented in step 7, is reset to
zero in step 5, and is possibly reset to zero in step 2.
Thus the conditions needed for complete baseline value
revision are apparent: first, sufficient stability of
SSVl-32 so that the resetting ofBRC is avoided in step
2, and second, passage through step 7 BRCL times
before passing through step 5.
Further processing of VALID TOUCH or INVALID RESPONSE beyond that indicated in the flow
charts of FIGS. 7-12 may proceed according to wellestablished techniques in keyboard scanner design.
All numbers are stored in the microprocessor 15 as
8-bit binary numbers and the sense signal variation values SSVVl-32 are encoded as signed binary numbers
lying between decimal minus 128 and decimal 127. It is
assumed that, in normal use, changes in each of
SSVl-32 will not exceed decimal 127 in magnitude and
thus that the sign of such changes will be given a correct interpretation in the microprocessor program.
As an example of program parameter choices, suppose that sense signal variation values of over 16 units
are reliably produced by operator touch under typical
operating conditions. THR may be set equal to 8. RVB
and VVB may be set equal to 4. BDCCL may be chosen
so that baseline value drift compensation occurs at the
rate of one per second when all SSVl-32 are consistently less than or equal to VVB in magnitude. In the
process of FIG. 10 in step 4.2 TOL may be assigned the
value of ROWMAX/4 and in step 4.5 TOL may be
assigned the value of COLMAX/4. BRCL may be
chosen so that a complete baseline value revision occurs
after two seconds of stabilized SSVs in the INVALID
RESPONSE state. The preselected values of BRC
which lead to a branch to reserve value drift compensation from step 2.2 in the process of FIG. 9 may be
chosen as 1, 2, 4, 8, ••• up to BRCL.
In an alternative preferred embodiment, the touch
sensing array 10 is adapted so that instead of the e1ectrades serving to form keys of a keyboard, an array of
electrodes is dimensioned and disposed so that an opera-

19

4,733,222

tor fmnly touching the interior area of a touch responsive panel overlying and coextensive with the electrode
array with his finger necessarily varies the capacitance
to ground of a plurality of electrodes simultaneously. In
this alternative embodiment the microprocessor 15 is 5
programmed to response so as to provide an output
indication of the coordinate position on the touch sensing arry 10 that is touched by the operator. The indication of the position of fmger touch may be used in a
variety of control situations, such as, for example, se- 10
lecting the cursor position on a cathode ray tube constituting the display portion of a computer terminal. In
other applications, transparent conductive material may
be used for the electrodes and the touch sensing array
15
may overlay the cathode tube.
The touch sensing array of this alternative embodiment is shown in FIG. 13, where it is seen that there are
row electrodes E1, ... supported on one side of a thin
dielectric substrate, and column electrodes E33, . . .
supported on the opposite side of said substrate. Each 20
electrode consists of a connected series of conductive
tabs 53. The center line-to-center line distance D between the row electrodes E1, . . . is 0.25 inch. The
column electrodes E33, . . . are separated by the same
distance. The conductive tabs are square in shape and 25
are connected by conductive paths which traverse the
tabs diagonally. Each square has a diagonal dimension
of 0.2 inch. The tabs on one side of the substrate do not
,;:underlie the tabs on the opposite side of the substrate.
':~Instead, as shown in FIG. 13, the tabs on one side of the 30
" substrate are disposed symmetrically between the positions of the tabs on the opposite side of the substrate.
The touch sensing array of this alternative embodi'ment has the same basic monolithic construction as
described above for the touch sensing array shown in 35
,FIG. 4. The touch responsive panel is the protective
"·film 25, which ordinarily would be backprinted with
," indicia to show a coordinate grid. The use of the optional overlay 27 would be appropriate with this emcohodiment when it is desired to specify portions of the 40
,;"array as having particular significance. As in the first
"described preferred embodiment the entire touch sensing array 10 in this alternative preferred embodiment is
a part of a monolithic construction that also includes the
drive line-to-electrode capacitors DLECl-64 and the 45
electrode-to-sense line capacitors ESLCl-64 as described above with reference to FIG. 5.
An alternative monolithic construction of the touch
sensing array of FIG. 13 in combination with the drive
line-to-electrode capacitors DLECl-64 and electrode- 50
to-sense line capacitors ESLCl-64 is shown in FIGS.
14 and 15. Referring to FIG. 14, a row electrode 54 of
the array of row electrodes El-32 is supported by the
top side of a first thin dielectric Mylar mm substrate 55;
and a column electrode 56 of the array of column elec- 55
trodes E33-64 is supported by the bottom side of the
first substrate 55. The drive lines and sense lines are
supported by opposite sides of a second thin dielectric
Mylar film substrate 57. In the representative cross-sectional view of FIG. 14, a line 5S of either the array of 60
row drive lines DLl-4 or the array of sense lines SL1-S
is supported by the top side of the second substrate 57;
and a line 59 of either the array of column drive lines
DL5-S or the array of sense lines SL1-S is supported by
the bottom side of the second substrate 57. A third thin 65
dielectric Mylar film substrate 60 separates the first
substrate 55 and the second substrate 57. Each of these
three substrates is approximately 0.001 inch ~hkk.

20

FIGS. 15A, 15B, 15C and 15D show the respective
layers of the arrays containing the lines 54, 56, 58 and 59
shown in FIG. 14. The registration marks in these figures indicate the alignment of the respective layers. The
top layer, shown in FIG. 15A, contains the array of row
electrodes El-32. The next layer, shown in FIG. 15B,
contains the array of column electrodes E33-64. The
following layer, shown in FIG. 15C, contains the array
of row drive lines DLl-4 and portions of the array of
sense lines SL1-S. The bottom layer, shown in FIG.
lSD, contains the array of column drive lines DL5-S
and portions of the array of sense lines SLl-8. When
these fpur layers are aligned as indicated by their registration marks, each of the row electrodes El-32 is capacitively coupled primarily to one of the row drive
lines DLl-4 and to one of the sense lines SL1-S; and
each of the column electrodes E33-64 is capacitively
coupled primarily to one of the column drive lines
DL5-S and to one of the sense lines SL1-S.1t is obvious
that the alternative construction methods illustrated in
FIGS. 14 and 15 are equally applicable to embodiments,
such as that shown in FIG. 5, in which the touch sensing array is configured as a keyboard.
Referring again to FIG. 14, the second substrate 57 is
supported by a nonconductive mechanical support
layer 61, which is covered on its bottom side by a
grounded conductive, electrically shielding layer 62.
The touch sensing array further includes a protective
0.005 inch Lexan plastic film layer 63 covering the top
unsupported side of the first substrate 55. Optionally the
touch sensing array may also include a paper overlay
layer 64 having various indicia and/or patterns thereon
to particularly indicate specific portions of the array.
Referring again to FIG. 10, the microprocessor program for this alternative embodiment necessarily is
modified in steps 4.3 and 4.6, to accommodate the fact
that more than one of the row electrodes El-32 and
more than one of the column electrodes E33-64 are
affected during a valid touch. The program can be modified to exempt from the tolerance test the sense signal
variation values derived from the two electrodes on
either side of the respective electrodes from which
ROWMAX and COLMAX are derived.
The principle modification of the microprocessor
program is to add a stage of processing beyond that of
feature extraction and confirmation for valid touch.
This modification is described with reference to FIG.
16 which illustrates a flow chart for an interpolation
process for calculating an x coordinate and a y coordinate for fmger touch position. In this process it is convenient to redesignate ach of the sense signal value variations SSVVl-32 as a positive part minus a negative
part; where the positive part of a nonnegative number is
the number itself, and the negative part of a nonpositive
number is the absolute value of the number. In this way
64 variables called electrode referred variation values
ERVVl-64 are obtained from the 32 sense signal variation values SSVVl-32.The two of ERVVl-64 associated with each ofSSVVl-32 are obtained by consulting
Table 3 with the obvious interpretation in the present
context. For example: the positive and negative parts of
SSVV21 are assigned to ERVV49 and ERVV51 respectively upon consulting Table 3; a value of minus
twelve for SSVV21 would yield zero for ERVV49 and
twelve for ERVV51; a value of fifteen for SSVV21
would yield fifteen for ERVV49 and zero for ERVV51.
Referring to FIG. 16, the interpolation routine for
estimating the y coordinate of operator finger touch

21

4,733,222

position from ERVVl-32 is set forth in steps 5.2.1
through 5.2.3, and the interpolation process for estimating the x coordinate of operator touch position from
ERVV33-64 is set forth in steps 5.2.4 through 5.2.6.
These steps are to be considered as an expansion in this
preferred embodiment of step 5.2 of the process covered by the flow chart of FIG. 10 for the first described
preferred embodiment. These interpolation processes
produce x and y coordinates which indicate the position
of operator fmger touch. Each coordinate is a number
between I and 125.
In step 5.2.1 the variable k is assigned the number of
the maximum of ERVVl-32; the variable j is assigned
the value of k-I; and the variable m is assigned the
value of k + 1. For convenience, in step 5.2.2 a variable
called ERVVO is assigned the value zero and ERVV33
is temporarily assigned the value zero. Next, variables
PREMAX, MAX, and POSTMAX are assigned the
respective values ERVVj, ERVVk, and ERVVm. Following this, the value of ERVV33 is restored to the
value of the positive part of SSVV17. Next, in step 5.2.3
the variable Y-COORDINATE which gives a ycoordinate for fmger touch position is calculated from the
elementary interpolation formula:
Y=4[k+{(POSTMAX-PREMAX)/(MAX+ IPOSTMAX -PREMAX I){]-3

22

referring to FIG. 5, wherein each tab complex includes
conductive tabs 20 and 21 respectively of a row electrode El-32 and a column electrode E33-64. This encoding principle clearly could be extended so that the
5 electrodes could be divided into three or more groups
and that fmger contact with a touch responsive area
overlying a tab complex would result in an increase in
the variable capacitance to ground of an electrode from
each of the three or more groups.
10
Yet another change that can be made in the foregoing
described preferred embodiments can be appreciated by
referring to FIG. 4, wherein a cross-sectional view of
the touch sensing array is shown. It is apparent that, in
areas not supporting the tabs 20 and 21, the thin fIlm
15 substrate 22 need not remain close to the surface of the

20

25
(Eq. I)

Y is rounded to the nearest integer. It is evident that the
term POSTMAX-PREMAX appearing in the interpolation formula is never greater than MAX and thus that 30
the term (pOSTMAX-PREMAX)/(MAX+ IPOSTMAX - PREMAX I) which is added to k always lies
between minus one half and plus one half. Next in steps
5.2.4 through 5.2.6 the interpolation routine of steps
5.2.1 through 5.2.3 is repeated with obvious modiflca- 35
tions to calculate the variable X-COORDINATE. In
steps 5.2.7 additional processing of VALID TOUCH
by methods well known to those versed in the art are
indicated.
A further modification of the microprocessor pro- 40
gram for this alternative embodiment is to exempt from
the tolerance test of the confrrmation steps 4.3 and 4.6,
described in relation to the flow chart of FIG. 10, the
electrode referred variation values derived from the
two adjacent electrodes on each side of the electrode 45
associated with the maximum ERVV. In this alternative
embodiment the ERVVs correspond to the SSVVs in
FIG.10.
The remaining features of the embodiment of FIG. 1
are likewise applicable to this alternative preferred em- 50
bodiment.
There are several changes that can be made in the
foregoing described preferred embodiments within the
scope of the present invention. For example, in the
embodiment of FIG. 2, there are eight drive lines 55
DLl-8 and eight sense lines SLl-8 capacitively coupled
in the designated manner to the electrodes El-64. It is
apparent that by suitable analog switching, a line could
serve as a drive line at one time and as a sence line at
another time. In particular, a 2-of-N coding scheme for 60
the electrodes could be used in which there are N combined drive and sense lines and (N(N -1»/2 electrodes
with each electrode being capacitively coupled to each
of a unique pair of the combined drive and sense lines.
This 2-of-N encoding scheme is well known to those 65
versed in the art.
Another change that can be made in the foregoing
described preferred embodiments can be appreciated by

touch sensing array and could be covered by some of
the su~porting laye~ 23. In some applic~tions this might
be desrred as a deVice to protect portlO.ns of the electrodes other tha? the tabs 20,21 fr?m bemg affected by
touch. ~lte';TIatlvely the prot~ctlve layer 25 can be
made.thick.m areas .not covermg tabs ~O and 21. The
resultmg raised portl.ons of the protecttve l.a!er could
serve as well as a gUide to finger touch position.
A still further change that could be made in the foregoing described preferred embodiments is in the drive
signal patterns shown in FIG. 6 which illustrates a technique of achieving a balancing of drive signals by the
simultaneous use of the drive signals DS and the inverted drive signal IDS on a pair of drive lines. It is
apparent that other means of balancing could be found
which avoid placing a drive signal on more than one
drive line at a time and thus avoid having a simultaneous increase in the variable capacitance to ground
(YCG) of two electrodes go undetected such as the
simultaneous increase in the VCG of electrodes E1 and
E2 as is apparent upon referring to Table 3.
An alternative scanning and sense signal value processing method is described with reference to FIGS.
17-19. In this method, because of an increased number
of drive signal patterns, a distinction can be made between sense signal variations due to the onset of touch
and sense signal variations due to the cessation of touch.
Based on this distinction, an erroneous assumption
within the signal processing program that the system is
in the untouched state will not lead to a misinterpretation of a later cessation of touch.
According to this technique· the drive signals are
provided on the drive lines DLl-8 by the drive signal
generator 17 (FIG. 1) in a repetitive sequence of six
different drive signal patterns DSP1 through DSP6, as
shown in FIG. 17. During DSP1, a drive signal DS is
provided on drive line DL1 and an inverted drive signal
IDS is provided on drive line DL3. Table 4 shows the
association of the drive signal DS, the inverted drive
signal IDS, and the drive lines DL during the six different drive signal patterns DSPl-6.
TABLE 4
DSPI
LINE
RECEIVINGDS
LINE
RECEIV·
ING IDS

DSP2

DSP3

DSP4

DSP5

DSP6

DLi

DL2

DLi

DLS

DL6

DL5

DL3

DU

DL4

DL7

DL8

DL8

Location of DS and IDS on
DLl-8 during DSPI-6

4,733,222

23

Referring again to FIG. 17, it is seen that during each
drive signal pattern, the paired drive signal DS and
inverted drive signal IDS occur simultaneously, are
synchronous with each other, of approximately the
same value and of opposing polarities. During each
cycle of drive signal patterns DSPl-6, a plurality of the
drive signal packets, DS, IDS are applied sequentially
on some of the drive lines and are paired synchronously
with the drive signal packets of opposite polarity IDS,
DS on different ones of the other drive lines. For exampIe, drive signal packets DS are applied on drive line
DLl during drive signal patterns DSPI and DSP3 and
are paired synchronously with inverted drive signals on
drive lines DLJ and DL4 during drive signal patterns
DSPI and DSP3 respectively.
A complee scan of the keyboard of this embodiment
corresponds to a sequence in which the drive signal
pattern cycle DSPl-6 is repeated for each of the sense
lines SLl-8 to yield a total of 48 sense signal values
SSVl-48. Table 5 shows the sign under the polarity
choice of DS and IDS of Table 3 of the variation of the
respective sense signal values associated with the different electrodes E that is due to the onset of an operator
touching a touch responsive area overlying parts of the
respective electrodes. The SSV number is a processing
sequence identification number and not the value of the
sense signal.
TABLE 5
E

SSV

1 1,17
2
9
3
1
4 9,17
5 2,18
6 10
7
2
8 10,18
9 3,19
':10 11
3
".'11
~~12 11,19
--·'''··13 4,20
'~:"14 12
15
4
16 12,20

ROW
SIGN E

+
+
+
+

+
+
+
+

17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32

SSV

SIGN

E

SSV

5,21
13
5
13,21
6,22
14
6
14,22
7,23
15
7
15,23
8,24
16
8
16,24

+
+

33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48

25,41
33
25
33,41
26,42
34
26
34,42
27,43
35
27
35,43
28,44
36
28
36,44

+
+
+
+
+
+

5

10

15

20

25

COLUMN
SIGN E SSV

+
+
+
+

+
+

49
50
51
52
53
54
55
56
57
58
59

60

+
+

61
62
63
64

SIGN

+
+

29,45
37
29
37,45
30,46
38
30
38,46
31,47
39
31
39,47
32,48
40
32
40,48

For example, at the onset of the touch responsive area
overlying a tab complex with a tab from row electrode
El and a tab from column electrode E48 being touched
by an operator there is a positive change in both sense
signal values SSVI and SSV17 and there is a negative
change in both sense signal values SSV36 and SSV44.
It will be appreciated upon inspection of Table 5 that
the change in sense signal values SSVl-48 brought
about by the cessation of operator touch of a single
touch responsive area is distinct from the change
brought about by the onset of operator touch. For example, the cessation of operator touch of a touch responsive area overlying a tab complex associated with
electrode E5 produces opposite changes to those shown
in Table 5. Thus there are significant negative changes
produced in sense signal values SSV2 and SSV18.
These negative changes are seen to uniquely indicate a
cessation of operator touch of a touch responsive area
overlying a tab complex associated with electrode E5.
These changes are not those that occur with the onset of
touch of a touch responsive area overlying a touch
element associated with electrode E8 because the required negative change in sense signal value SSVI0 is

24

not present. Also these changes are not those that occur
with the onset of touch of a touch responsive area overlying a tab complex associated with electrode E7 since
in that case a significant negative change in sense signal
value SSV18 is not present.
The microprocessor 15 is programmed to process the
sense signal values SSVl-48 on line 43 iteratively to
ascertain the signs and values of the variations
SSVVl-48 in SSVl-48 and to process SSVVl-48 as a
group to determine whether such variations resulted
from a touch responsive area being touched by an operator. The flow charts for this program are shown in
FIGS. 7, 18 and 19.
The steps described above in relation to FIG. 7 for
the first-described embodiment are also applicable to
this program except that in step 2 a scan of the touch
sensing array 10 is carried out iteratively with the production of sense signal values SSVl-48 as the program
input. Significant program differences are shown in the
flow charts of FIGS. 18 and 19. The step-numbering
system in FIGS. 18 and 19 agrees in the first digit with
that of FIG. 7.
In the flow chart of FIG. 18, step 1.1 consists of
initializing accumulated difference values ADl-48 to
zero and a stability count SC to zero. Sense signal values SSVl-48 are input in step 2.1. Next, in step 2.2 the
sense signal variation value SSVVk is calculated and is

+
+
+
+
+
+

50

55

60

65

added to ADk, the previous sense signal value SSVk is
then discarded, and the current SSVk is saved for each
k between I and 48. In step 2.3 SC is incremented if the
magnitude of each of SSVVl-48 is found to be less than
a predetermined stability bound SB. Next, in step 2.4 SC
is compared to a predetermined stability count limit
SCL. If there is equality, there is a branch to step 2.5
where ADl-48 and SC are reset to zero. Otherwise
there is a branch to step 3.1.
In step 3.1, if the magnitude of each ADl-48 is found
to be smaller than a predetermined accumulated difference bound ADB, the touch sensing array is judged to
be in the UNTOUCHED state and there is a branch to
step 6, otherwise there is a branch to step 4.
To summarize the function of the program shown in
the flow chart of FIG. 18, during each iteration the
current sense signal values SSVl-48 are compared with
the respective sense signal values obtained during the
previous iteration to determine sense signal variation
values SSVVl-48. The SSVVl-48 are added to the
respective accumulated difference values ADl-48; and
each of ADl-48 is compared to a predetermined accu-

25

4,733,222

mulated difference bound ADB to determine whether
the UNTOUCHED state is present. The ADl-48 are
reset to an initial value ofzero whcn each of SSVVl-48
remains less than a predetermined stability bound SB
over a predetermined number of iterations. Each of 5
SSVl-48 is compared to SB; and the stability count SC
is incremented in response to each set of SSVl-48 for
which one of SSVVl-48 exceeds the predetermined
stability bound SB. ADl-48 and SC are reset to an
10
initial value of zero when SC reaches SCL.
Referring next to FIG. 19, in the course of processing
the accumulated differences ADl-48 the microprocessor 15 carries out a feature extraction stage in which it
determines ROWMAXl, the maximum magnitude of
the accumulated differences ADl-24 which are the 15
differences derived from the row electrodes El-32 and
ROWMAX2, the second largest magnitude of ADl-24.
Next, parameters for a confIrmation stage are determined; these are TOLl, a tolerance value which is a
predetermined fraction of ROWMAXI and THR2, a 20
predetermined fraction of ROWMAXI which serves as
a threshold value for ROWMAX2. In a similar fashion,
the microprocessor 15 determines features COLMAXl,
the maximum magnitude of the accumulated differences 25
AD25-48 which are the differences derived from the
column electrodes E33-64 and COLMAX2, the second
largest magnitude of AD25-48 and determines confumation parameters TOL2, a tolerance value which is a
predetermined fraction of COLMAXI and THR4, a 30
predetermined fraction of COLMAXI which serves as
a threshold value for COLMAX2. A valid touch onset
is judged to be present when separate determinations
relating to row electrodes El-32 and to column electrodes E33-64 are affIrmative.
35
ConfIrmation requirements for the row component of
a valid touch are that (a) ROWMAXI must exceed a
predetermined threshold value THRl; (b) if ROWMAXI is derived from a row electrode to which two
drive signal packets are applied during each repetitive 40
sequence of drive signal patterns DSPI-6, then ROWMAX2 must be derived from the same row electrode as
ROWMAXI and must exceed the threshold THR2; (c)
all magnitudes of ADl-24 other than ROWMAXl, and
when applicable ROWMAX2, must be less than TOLl; 45
and (d) the specific one of ADl-24 giving rise to ROWMAXI, and when applicable the specifIc one of AD1-24 giving rise to ROWMAX2, must indicate, on consulting Table 5 (which is stored in the memory of the
microprocessor 15), an appropriate response as con- 50
cerns row electrodes to the onset of touch of a single
touch responsive area.
ConfIrmation requirements for the column component of a valid touch are that (e) COLMAXI must
exceed a predetermined threshold value THR3; (t) if 55
COLMAXI is derived from a column electrode to
which two drive signal packets are applied during each
repetitive sequence of drive signal patterns DSPI-6,
then COLMAX2 must be derived from the same colUmn electrode as COLMAXI and must exceed the 60
threshold THR4; (g) all magnitudes of AD25-48 other
than COLMAXl, and when applicable COLMAX2,
must be less than TOL2; and (h) the specific one of
AD25-48 giving rise to COLMAXl, and when applicable the specific one of AD25-48 giving rise to COL- 65
MAX2, must indicate, on consulting Table 5, an appropriate response as concerns column electrodes to the
onset of touch of a single touch responsive area.

26

All of such of the eight immediately preceding recited determinations (a)-(h) that are applicable being
affirmative indicates the onset of only a single one of the
touch responsive areas being touched instantaneously
by an operator.
Tests for a valid touch response are carried out in
steps 4.1 through 4.14, as shown in the flow chart of
FIG. 19. In step 4.1 ROMAXI is compared to the first
predetermined threshold value THRI. If ROWMAXI
exceeds THRI there is a branch to step 4.2 where the
tolerance value TOLl is set to some fixed fraction ofthe
value of ROWMAXI. Next, in step 4.3 if it is determined that ROWMAXI is derived from a row electrode to which two drive signal packets are applied
during each repetitive sequence of drive signal patterns
DSPl-6, whereby a valid touch response requires that
the second largest magnitude of ADl-24, ROWMAX2,
be over the threshold value THR2, there is a branch to
step 4.5. Otherwise there is a branch to step 4.4 where
each of ADl.,..24 other than the one chosen as ROWMAXI is compared to the tolerance value TOLl. If
none of these values exceeds TOLl then the program
branches to step 4.7. In step 4.5 ROWMAX2 is compared to the threshold value THR2. If ROWMAX2 is
found to be greater than THR2 there is a branch to step
4.6 where each of ADl-24 other than the ones chosen
as ROWMAXI and ROWMAX2 is compared to
TOLl. If none of these values exceeds TOLl there is a
branch to step 4.7 where the accumulated difference
associated with ROWMAXI and when applicable the
accumulated difference associated with ROWMAX2
are checked using the information in Table 5. If the
result is found to be valid, then a VALID ROW
TOUCH is judged to be present and the program
branches to step 4.8.
In step 4.8, the variable COLMAXI is compared to
the threshold value THR3. If COLMAXI exceeds
THR3 then TOL2 is calculated as a fixed fraction of
COLMAXI in step 4.9. In step 4.10, if it is determined
that COLMAXI is derived from a column electrode to
which two drive signal packets are applied during each
repetitive sequence of drive signal patterns DSPl-6,
whereby a valid touch response requires that the second
largest of AD25-48, COLMAX2, be over the threshold
value THR2, there is a branch to step 4.12. Otherwise
there is a branch to step 4.11 where each of AD25-48
other than the one closed as COLMAXI is compared to
the tolerance value TOL2. If none of these values exceeds TOL2, then the program branches to step 4.14. In
step 4.12, COLMAX2 is compared to the threshold
value THR4. If COLMAX2 is found to be greater than
THR4, there is a branch to step 4.13 where each of
AD25-48 other than the ones chosen as COLMAXI
and COLMAX2 is compared to TOL2. If none of these
values exceeds TOL2, there is a branch to step 4.14
where the accumulated difference associated with
COLMAXI and when applicable the accumulated difference associated with COLMAX2 are checked using
the information in Table 5. If the result is found to be
valid, then the tests for a VALID TOUCH have been
passed and there is an exit to step 5.1.
Failing any of the tests of steps 4.1, 4.4, 4.5, 4.6, 4.7,
4.8, 4.11, 4.12, 4.13, or 4.14 causes a branch to step 7
where an INYALID RESPONSE is processed. In step
5.1 processing of a VALID TOUCH response is carried
out including the appropriate encoding of the touched
touch res~onsive area by row number and by column

4,733,222
27

28

initializing step, step 1 of FIG. 7, the microprocessor
number which numbers may be calculated using the
creates and stores a table of drive signal control parameinformation in Table 5.
As an example of program parameter choices, supters DSCPI-32. For each k between 1 and 32, the papose that accumulated difference values of over sixteen
rameter DSCPk controls the amplitude of DS during
units are reliably produced by operator touch under 5 the generation of the sense signal value SSVk of Table
typical operating conditions. THRI and THR3 may be
3. Methods of controlling the amplitude of a drive sigset equal to eight. SCL may be chosen so that the accunal packet such as DS based on the value of a stored
mulated differences ADI-48 are reset to zero after 0.5
parameter are well known to those skilled in the art.
seconds of stable operation. In step 4.2 TOLl may be
The drive signal control parameter DSCPk is chosen
assigned the value of ROWMAXI/4 and in step 4.9 10 during initialization so that the sense signal value SSVk
TOL2 may be assigned the value of COLMAXI/4. In
is as near as possible to the mid-range value of 128.
step 4.5 THR2 may be assigned the value of ROWProcedures for optimaly selecting the value of a control
MAXI/2 and in step 4.12 THR4 may be assigned the
parameter such as DSCPk are well known to those
value of COLMAXI/2. It is thus a feature of this proskilled in the art. Step 1.1 of FIG. 8 is modified so that
gram that ADI-48 may be reset to zero when operator 15 the value of the drive signal control parameter resulting
touch of a touch responsive area is present for an exin the value of SSVk nearest 128 is stored as DSCPk
tended dwell time without danger of misinterpretation
and this value ofSSVk nearest 128 is stored as BVk and
of the subsequent ADI-48 that occur when operator
as RVk. Step 2.1 of FIG. 9 is modified to include an
touch is removed from the tab complex. This is in conupdating procedure for the drive signal control parametrast to the situation described above in conjunction 20 ters DSCPI-32. This updating procedure, which is
with the first-described preferred embodiment discarried out for each k between 1 and 32 when the basecussed in relation to FIGS. 6-12, where resetting baseline drift compensation count lies in a predetermined
line values during touch is prohibited since misinterprerange and when the stored value of SSVk differs from
tation of the cessation of touch would then occur.
the mid-range value 128 by more than a predetermined
Alternatives to the method of resetting of the accu- 25 tolerance value, is described in Table 6.
mulated differences ADI-48 as described above will be
TABLE 6
obvious to one skilled in the art. For example, resetting
Step Procedure
could be delayed when valid touch is judged to be
present to allow for an extended dwell period and resetA. Store SSVk as OLD_SSV
; ting could be hastened in the period immediately fol- 30
B. Store DSCPk as OLD_DSCP
C. Choose the optimal value NEW_DSCP for DSCPk
-·lowing a response indicative of the cessation of touch.
which results in a value of SSVk as near as
As a further example, each ADk could be replaced
possible to the mid-range value of 128
_ periodically with a fixed fraction of its value for each k
D. Store the value of SSVk nearest to 128
between 1 and 48 and the rate of replacement could
obtained in step C as NEW_SSV
E. Test the stability of the system:
depend on such factors as stability and the state of the 35
I. use OLD_DSCP as DSCPk and obtain a
system with respect to touch.
further SSVk
Except for the differences discussed above in the
2. compare this further SSVk to OLD_SSV
description of the preferred embodiment of FIGS.
F. If the system is found to be sufficiently
stable in Step E then replace DSCPk with
_17-19, this preferred embodiment (FIGS. 17-19) is conNEW_DSCP and replace BVk and RVk with
structed and functions in the same manner as the pre- 40
NEW_SSV. otherwise DSCPk. BVk. and RVk are
__ ferred embodiment first described above in relation to
not revised
-;;FIGS. 1-12. It will be apparent to one skilled in the art
that the optional capacitor trimming step of manufacAn addition to Step C in Table 6 may be that the
ture discussed in conjunction with Table 3 must be
modified if drive signal patterns DSPI-6 of Table 4 are 45 value NEW_DSCP be constrained to differ by no more
than a set amount from OLD_DSCP, thereby simplifyemployed, since electrodes associated with two sense
ing the search for the optimal value of NEW_DSCP.
signal values will necessarily be involved in two trimIn an alternative mode of operation of the system of
ming and balancing operations. It will be apparent to
FIG. I, the operation of the synchronous demodulator
one skilled in the art that the needed algebraic independence of trimming effects is present for simultaneously 50 13 is modified. In this alternate mode, the phase of operation of the synchronous demodulator 13 is controlled
achieving balance for all of SSVI-48.
by the microprocessor 15 using stored phase control
An alternative method of attaining the desired balparameters. Phase effects in synchronous demodulation
ance of DS and IDS on each of the sense lines SLI-8
are well known to those skilled in the art. Control of the
during each of DSPI-4 which replaces the trimming
procedure described in conjunction with Table 3 in- 55 phase of demodulation may be complicated by phase
changes in the drive signals which accumulate as the
volves modifications of the features described above
drive signals propagate through the conductor arrays
with reference to FIGS. I, 7, 8, and 9. In this alternative
and the band pass amplifier 12 of FIG. 1. The phase of
method, the microprocessor 15 and the drive signal
demodulation can be fixed throughout all processing
generator 17 of FIG. 1 are configured so that the amplitude of one or both of the drive signals DS and IDS of 60 periods of the touch sensing array system or alternatively the phase can be independently regulated during
FIG. 6 are under program control. Preferably, the amthe period that each sense signal value SSVk is obtained
plitude of DS is under program control and the amplitude of IDS is fixed. The range of possible amplitudes of
for each k between 1 and 32. Phase control parameters
DS that can be obtained under program control is such
may be stored in the read only memory (ROM) of the
that, in the absence of operator touch, balancing of DS 65 microprocessor 15 or alternatively the phase control
and IDS on each of the sense lines SLI-8 during each of
parameters may be determined during initialization of
the drive signal periods DSPI-4 can be achieved
the touch sensing array system in step 1 of FIG. 6 and
through the control of the amplitude of DS. During the
stored in the random access memory (RAM) of the

29

4,733,222

microprocessor 15. In the presence of conditions of
operation of the touch sensing array system that alter
the phase of the drive signals at the synchronous demodulator, dynamic updating of phase control parameters may be employed.
5
I claim:
1. A capacitance-variation-sensitive touch sensing
array system comprising
an array of electrodes with each electrode including a
connected series of conductive tabs;
10
an array of keys with each key overlying at least one
tab of at least one of said electrodes for varying the
capacitance between said electrode and ground
upon the key being touched by an operator;
an array of drive lines capacitively coupled to the 15
electrodes, with individual drive lines being coupled to a plurality of the electrodes;
drive means for applying drive signals to the drive
lines to drive the electrodes coupled thereto;
an array of sense lines capacitively coupled to the 20
electrodes, to derive sense signals from the electrodes when drive signals are applied to the drive
lines, with each of said sense lines being so coupled
to a plurality of the electrodes as to be interconnected via unique combinations of electrodes to 25
said drive lines such that the sense lines derive
unique combined patterns of variations in said
sense signals in response to actuation of each of the
keys;
means coupled to the sense lines for sensing said pat- 30
terns of variations in said sense signals; and
means coupled to the sensing means for processing
the sense signals to provide information associated
with any particular key that is touched by an operator;
35
whereby the number of keys exceeds the product of
the number of drive lines and the number of sense
lines.
2. A system according to claim 1,
wherein the electrodes are arrayed in rows and col- 40
umns; and
wherein each key overlies a tab complex made up of
a tab of one of the row electrodes and a tab of one
of the column electrodes.
3. A system according to claim 1,
45
wherein each electrode is capacitively coupled to
only one drive line and to only one sense line; and
wherein no given drive line and sense line pair is
connected to more than one electrode in common.
4. A system according to claim 1,
50
wherein the drive means generate drive signals comprising alternating signal packets, with simultaneously occurring signal packets being generated
synchronously on different drive lines; and
wherein the means for sensing the sense signals on the 55
sense lines includes means for synchronously demodulating the sensed signals to cause the demodulation of the sensed signals to be synchronous with
the generation of the corresponding drive signal
packets by the drive means.
60
5. A system according to claim 4, wherein the processing means uses stored phase control parameters to
control the phase of demodulation of the sensed signals
by the demodulating means.
6. A system according to claim 3, wherein the differ- 65
ent drive signals in combination include alternating
signal packets applied at the same time to pairs of drive
lines with the simultaneously occuring signal packets

30

being synchronous with each other, of approximately
the same amplitude, and of opposing polarities to
thereby cause the amplitude of the sense signal appearing on a sense line connected to separate electrodes
driven by each of the simultaneously occurring drive
signals to be low in relation to the variation in the amplitude of said sense signal resulting from a key being
touched by an operator.
7. A system according to claim 6, wherein the capacitors coupling the drive lines to the electrodes and the
capacitors coupling the electrodes to the sense lines are
trimmed in order to balance each pair of the transmitted
drive signals of opposing polarities appearing on a sense
line.
8. A system according to claim 6, wherein the processing means controls the amplitudes of one or both of
the individual drive signals of each pair of the simultaneously occuring drive signals of opposing polarities in
order to balance the transmitted drive signals appearing
on a sense line.
9. A system according to claim 6,
wherein the drive means comprise a drive signal generator for applying sequentially during each repetitive sequence a plurality of said signal packets on
some of the drive lines paired synchronously with
said signal packets of opposite polarity on different
ones of the other drive lines; and
wherein the processing means include means for determining whether a sensed variation in the capacitance between an electrode and circuit ground is
due to the onset or the cessation of touching a key
by an operator.
10. A system according to claim 1, further comprising
fIrst, second, and third dielectric substrates;
wherein a fIrst portion of the electrode array is supported by one side of the fIrst dielectric substrate
and a second portion of the electrode array is supported by the opposite side of the fIrst dielectric
substrate;
wherein an individual key of the array of keys overlies a fIrst tab supported by the one side of the fIrst
dielectric substrate and extending from an electrode of the fIrst portion of the array and a second
tab supported by the opposite side of the fIrst dielectric substrate and extending from an electrode
of the second portion of the array adjacent to the
position underlying the fIrst tab so that the capacitance between both electrodes and ground is varied
when the key is touched by an operator;
wherein the drive lines and sense lines are supported
by opposite sides of the second dielectric substrate;
wherein the third dielectric substrate separates the
second dielectric substrate supporting the drive
lines and the sense lines from the fIrst dielectric
substrate supporting the electrode arrays; and
wherein the drive lines and sense lines on the second
dielectric substrate are dimensioned and disposed
in spatial relation to the electrodes on the fIrst
dielectric substrate so that the required capacitive
coupling of drive lines to electrodes and of sense
lines to electrodes is obtained.
11. A system according to claim 1, further comprising
a thin dielectric substrate;
wherein a fIrst portion of the electrode array is supported by one side of the dielectric substrate and a
second portion of the electrode array is supported
by the opposite side of the dielectric substrate;

31

4,733,222

wherein an individual key of the array of keys overlies a f11"st tab supported by the one side of the
dielectric substrate and extending from an electrode of the f11"st portion of the array and a second
tab supported by the opposite side of the dielectric 5
substrate and extending from an electrode of the
second portion of the array adjacent to the position
underlying the f11"st tab so that the capacitance
between both electrodes and ground is varied
10
when the key is touched by an operator;
wherein the drive lines and sense lines are supported
by the dielectric substrate; and
wherein a drive line or sense line on one side of the
substrate is coupled to an electrode on the opposite
15
side of the substrate by a capacitor comprising
a f11"st plate supported by the one side of the substrate and extending from the coupled drive line
or sense line; and
a second oppositely disposed plate supported by
the opposite side of the substrate and extending 20
from the coupled electrode.
12. A system according to claim 1, wherein the processing means comprise
means for providing sense signal values in accordance
with the amplitudes of the respective sense signals; 25
means for processing the sense signal values iteratively to ascertain the values of variations in the
sense signal values; and
means for processing the sense signal variation values
to determine that an operator touch is present at a 30
specific location when the sense signal variation
values match a pattern of variation values characteristically associated with touch at that specific
location.
13. A system according to claim 12, wherein the 35
means for processing the variation values process the
.. variation values as one or more groups in two stages,
'co. the first stage being a feature extraction stage in which
a search is made for features indicative of touch and the
second stage being a confirmation stage which is carried 40
out when an affrrmative determination is made in the
···.,.f11"st stage and in which tests are carried out to confirm
that a pattern of sense signal value variation characteristic of a key being touched by an operator is or is not
present.
45
14. A system according to claim 13,
wherein the electrode array defmes rows and columns of electrodes; and
wherein the means for processing the variation values
comprises
50
means for determining in the feature extraction
stage whether the maximum of the group of
variation values derived from the row electrodes
exceeds a predetermined threshold value;
means for determining in the confirmation stage 55
whether no other of the group of variation values derived from the row electrodes exceeds a
tolerance value that may depend on said maximum variation value derived from the row electrodes;
60
means for determining in the feature extraction
stage whether the maximum of the group of
variation values derived from the column electrodes exceeds a predetermined threshold value;
means for determining in the confirmation stage 65
whether no other of the group of variation values derived from the column electrodes exceeds
a tolerance value that may depend on said maxi-

32

mum variation value derived from the column
electrodes; and
means for determining in response to all of said
four immediately preceding recited determinations being affirmative that the maximum variation value derived from the row electrodes and
the maximum variation value derived from the
column electrodes each resulted from only a
single one of the keys being touched instantaneously by an operator.
15. A system according to claim 1, wherein the processing means comprise
means for providing sense signal values in accordance
with the amplitudes of the respective sense signals;
and
means for processing the sense signal values iteratively by comparing the sense signal values with
baseline values derived from sense signal values
provided when the touch sensing array is untouched by an operator to determine which one of
three states is present, the first state being an untouched state, the second state being a valid touch
state in which a valid touch by an operator is present, and the third state being an invalid response
state which occurs when neither said first state nor
said second state is present.
16. A system according to claim 15, wherein the
processing means revises the baseline values in response
to sensed variations in the sense signal values which are
stable over a predetermined number of iterations and
which indicate the invalid response state of the touch
sensing array.
17. A system according to claim 16, wherein the
means for revising the baseline values comprises
means for storing a set of reserve values derived from
a set including each of the respective sense signal
values;
means for iteratively comparing each sense signal
value to each corresponding reserve value;
means for revising the stored set of reserve values to
correspond to a current set of sense signal values
whenever the difference between any sense signal
value and the corresponding reserve value exceeds
a predetermined bound value; and
means for revising the baseline values to correspond
to the reserve values when the invalid response
state is present for a predetermined number of iterations during which time not revision of the reserve values occurs, no revision of baseline values
to correspond to reserve values occurs and the
valid touch state is not present.
18. A system according to claim 15, wherein the
processing means comprise
means for compensating for drift in the sense signal
values that are provided when the keys are in the
untouched state, wherein the baseline values are
updated by either incrementing or decrementing
each of the respective baseline values in accordance with whether the baseline value is less or
greater than the respective last-provided sense
signal value when the untouched state is present for
a predetermined number of consecutive iterations
during which time no updating of the baseline values occurs.
19. A system according to claim 18, wherein the
compensating means further comprises

33

4,733,222

means for delaying by degree the updating of baseline
values when the invalid response state occurs only
intermittently.
20. A system according to claim 1, wherein the pro5
cessing means comprise
.
means for providing sense signal values in accordance
with the amplitude of the· respective sense signals;
means for processing the sense signal values iteratively to determine whether any of the keys has
been touched by an operator by comparing the 10
current sense signal values during each iteration
with the respective sense signal values during the
previous iteration, by adding the same signal variation value derived from said comparison to an
accumulated difference value for each sense signal, 15
and by determining whether a key has been
touched by processing the accumulated difference
values; and
means for resetting the accumulated difference values
to an initial value when the sense signal variation 20
values remain less than a predetermined bound
value over a predetermined number of consecutive
iterations during which no resetting of the accumu. lated difference values to the initial value occurs.
21. A capacitance-variation-sensitive touch sensing 25
array system comprising
an array of electrodes with each electrode including a
connected series of conductive tabs;
a touch responsive panel overlying and coextensive
with the electrode array for varying the capaci- 30
tance between a plurality of the electrodes and
ground simultaneously upon the interior area of the
touch responsive panel being touched by an operator;
an array of drive lines capacitively coupled to the 35
electrodes, with individual drive lines being coupled to a plurality of the electrodes;
drive means for applying drive signals to the drive
lines to drive the electrodes coupled thereto;
an array of sense lines capacitively coupled to the 40
electrodes to derive sense signals from the electrodes when drive signals are applied to the drive
lines, with each of said sense lines being so coupled
to a plurality of the electrodes as to be interconnected via unique combinations of electrodes to 45
said drive lines such that the sense lines derive
unique combined patterns of variations in said
sense signals in response to touching distinct locations on the panel;
means coupled to the sense lines for sensing said pat- 50
terns of variations in said sense signals; and
means coupled to the sensing means for processing
the sense signals to provide an estimate of the location of operator touch;
whereby the number of said distinct locations of op- 55
erator touch that can be estimated exceeds twice
the product of the number of drive lines and the
number of sense lines.
22. A system according to claim 21,
wherein each electrode is capacitively coupled to 60
only one drive line and to only one sense line; and
wherein no given drive line and sense line pair is
connected to more than one electrode in common.
23. A system according to claim 21,
wherein the drive means generate drive signals com- 65
prising alternating signal packets, with simultaneously occurring signal packets being generated
synchronously on different drive lines; and

34

wherein the means for sensing the sense signal on the
sense lines includes means for synchronously demodulating the sensed signals to cause the demodulation of the sensed signals to be synchronous with
the generation of the corresponding drive signal
packets by the drive mea~s.
24. A system according to claim 23, wherein the
processing means uses stored phase control parameters
to control the phase of demodulating of the sensed
signals by the demodulating means.
25. A system according to claim 22, wherein the
different drive signals in combination include alternating signal packets applied at the same time to pairs of
drive lines with the simultaneously occuring signal
packets being synchronous with each other, of approximately the same amplitude, and of opposing polarities
to thereby cause the amplitude of the sense signal appearing on a sense line connected to separate electrodes
driven by each of the simultaneously occurring drive
signals to be low in relation to the variation in the amplitude of said sense signal resulting from the touch responsive area being touched by an operator.
26. A system according to claim 25, wherein the
capacitors coupling the drive lines to the electrodes and
the capacitors coupling the electrodes to the sense lines
are trimmed in order to balance each pair of the transmitted drive signals of opposing polarities appearing on
a sense line.
27. A system according to claim 25, wherein the
processing means controls the amplitudes of one or both
of the individual drive signals of each pair of the simultaneously occuring drive signals of opposing polarities
in order to balance the transmitted drive signals appearing on a sense line.
28. A system according to claim 21, further comprising
fIrst, second, and third dielectric substrates;
wherein the electrode array defines rows and columns of electrodes;
wherein the row electrodes are supported by one side
of the fIrst dielectric substrate and the column electrodes are supported by the opposite side of the
first dielectric substrate;
wherein the tabs of the column electrodes on one side
of the fIrst substrate are disposed so as to lie between the tabs of the row electrodes on the opposite side of the first substrate;
wherein the drive lines and sense lines are supported
by opposite sides of the second dielectric substrate;
wherein the third dielectric substrate separates the
second dielectric substrate supporting the drive
lines and the sense lines from the fIrst dielectric
substrate supporting the electrode arrays; and
wherein the drive lines and sense lines on the second
dielectric substrate are dimensioned and disposed
in spatial relation to the electrodes on the first
dielectric substrate so that the required capacitive
coupling of drive lines to electrodes and of sense
lines to electrodes is obtained.
29. A system according to claim 21, further comprising
a thin dielectric substrate;
wherein the electrode array defInes rows and columns of electrodes;
wherein the row electrodes are supported by one side
of the dielectric substrate and the column electrodes are supported by the opposite side of the
dielectric substrate~

35

4,733,222

wherein the tabs of the column electrodes on one side
of the substrate are disposed so as to lie between
the tabs of the row electrodes on the opposite side
of the substrate;
wherein the drive lines and sense lines are supported 5
by the dielec;tric substrate; and
wherein a drive line or sense line on one side of the
substrate is coupled to an electrode on the opposite
side of the substrate by a capacitor comprising
a first plate supported by the one side of the sub- 10
strate and extending from the coupled drive line
or sense line; and
a second oppositely disposed plate supported by
the opposite side of the substrate and extending
from the coupled electrode.
15
30. A system according to claim 21, wherein the
processing means comprise
means for providing sense signal values in accordance
with the amplitudes of the respective sense signals;
means for processing the sense signal values itera- 20
tively to ascertain the values of variations in the
sense signal values; and
means for processing the sense signal variation values
to determine that an operator touch is present at a 25
specific location when the sense signal variation
values match a pattern of variation values characteristically associated with touch at that specific
location.
31. A system according to claim 30, wherein the 30
means for processing the variation values process the
variation values as one or more groups in three stages,
the first stage being a feature extraction stage in which
a search is made for features indicative of touch, the
second stage being a confirmation stage which is carried 35
out when an affIrmative determination is made in the
frrst stage and in which tests are carried out to confirm
that a pattern of sense signal value variation characteristic of touch by an operator is or is not present, and the
third stage being an interpolation stage which is carried 40
-. out when affIrmative determinations are made in the
/; first and second stages and in which an interpolation
procedure using the variation values of a group supplies
a coordinate position for the location of touch.
32. A system according to claim 31,
45
wherein the electrode array defines rows and columns of electrodes; and
wherein the means for processing the variation values
comprises
means for determining in the feature extraction 50
stage whether the maximum of the group of
variation values derived from the row electrodes
exceeds a predetermined threshold value;
means for determining in the confIrmation stage
whether none of the other variation values de- 55
rived from the row electrodes other than a given
number of row electrodes on each side of and
adjacent to the row electrode from which said
maximum variation value is derived exceeds a
tolerance value that is a predetermined fraction 60
of said maximum variation value derived from
the row electrodes;
means for estimating in the interpolation stage by
an interpolation formula the y coordinate of
operator touch from the group of variation val- 65
ues derived from the row electrodes;
means for determining in the feature extraction
stage_ whether the maximum of the group of

36

variation values derived from the column electrodes exceeds a predetermined threshold value;
means for determining in the confirmation stage
whether none of the other variation values derived from the column electrodes other than a
given number of column electrodes on each side
of and adjacent to the column electrode from
which said maximum variation value is derived
exceeds a tolerance value that is a predetermined
fraction of said maximum variation value derived from the column electrodes; and
means for estimating in the interpolation stage by
an interpolation formula the x coordinate of operator touch from the group of variation values
derived from the column electrodes.
33. A system according to claim 21, wherein the
processing means comprises
means for providing sense signal values in accordance
with the amplitudes of the respective sense signals;
and
means for processing the sense signal values iteratively by comparing the sense signal values with
baseline values derived from sense signal values
provided when the touch sensing array is untouched by an operator to determine which one of
three states is present, the first state being an untouched state, the second state being a valid touch
state in which a valid touch by an operator is present, and the third state being an invalid response
state which occurs when neither said first state nor
said second state is present.
34. A system according to claim 33, wherein the
processing means revises the baseline values in response
to sensed variations in the sense signal values which are
stable over a predetermined number of iterations and
which indicate the invalid response state of the touch
sensing array.
35. A system according to claim 34, wherein the
means for revising the baseline values comprises
means for storing a set of reserve values derived from
a set including each of the respective sense signal
values;
means for iteratively comparing each sense signal
value to each corresponding reserve value;
means for revising the stored set of reserve values to
correspond to a current set of sense signal values
whenever the difference between any sense signal
value and the corresponding reserve value exceeds
a predetermined bound value; and
means for revising the baseline values to correspond
to the reserve values when the invalid response
state is present for a predetermined number of iterations during which time no revision of the reserve
values occurs, no revision of baseline values to
correspond to reserve values occurs and the valid
touch state is not present.
36. A system according to claim 33, wherein the
processing means comprise
means for compensating for drift in the sense signal
values that are provided when the touch sensing
array is in the untouched state, wherein the baseline values are updated by either incrementing or
decrementing each of the respective baseline values in accordance with whether the baseline value
is less or greater than the respective last-provided
sense signal value when the untouched state is
prese'lt for a predetermined number of consecutive

37

4,733,222

iterations during which time no updating of the
baseline values occurs.
37. A system according to claim 36, wherein the
compensating means further comprises
means for delaying by degree the updating ofbaseline 5
values when the invalid response state occurs only
intermittently.
38. A capacitance-variation-sensitive touch sensing
array system comprising
an array of electrodes with each electrode including 10
at least one conductive tab;
an array of keys with each key overlying at least one
tab of at least one electrode of the electrode array
, for varying the capacitance between each said
overlaid electrode and ground upon the key being 15
touched by an operator;
drive means for applying drive signals to drive the
electrodes;
0
an array of sense lines coupled to the electrodes;
sensing means coupled to each of the sense lines to 2
derive a digital sense value from each of the respective sense lines when drive signals are applied to
the electrodes, which digital sense value is dependent on variations in the electrode to ground ca- 25
pacitance of at least one of the electrodes, with said
sense lines being so coupled to the electrodes as to
be interconnected via the electrodes to said drive
lines such that said sense lines derive unique combined patterns of variations in said sense signals in 30
response to actuation of each of the keys; and
means for processing the digital sense values iteratively to ascertain the values of variations in the
digital sense values and for processing the variation
values in two stages, the fIrst stage being a feature 35
extraction stage in which a search is made for said
digital sense value variations that may be indicative
of a given key being touched and the second stage
being a confIrmation stage which is carried out
when an affirmative determination is made in the 40
ftrst stage and in which said digital sense value
variations for which said afftrmative determination
was made during the feature extraction stage are
processed together with a predetermined group of
digital sense value variations to determine whether 45
said unique combined pattern of digital sense value
variations characteristic of said given key being
touched by an operator is or is not present.
39. A system according to claim 38,
wherein the keys of the touch sensing array deftnes 50
rows and columns of keys;
wherein the digital sense values include two groups
with the fIrst group providing int:ormation associated with the row of a key and with the second
group providing information associated with the 55
column of a key;
wherein the effect of operator touch of a single key of
the touch sensing array is to predominantly vary
one digital sense value in the ftrst group of digital
sense values associated with rows of keys and one 60
digital sense value in the second group of digital
sense values associated with columns of keys; and
wherein the means for processing the variations in the
digital sense values comprises
means for determining in the feature extraction 65
stage whether the maximum variation in the
values associated with rows of keys exceeds a
predetermined threshold value;

38

means for determining in the conftrmation stage
whether none of the other variations in the digital sense values associated with rows of keys
exceeds a tolerance value that may depend on
said maximum variation value derived from the
rows of keys;
means for determining in the feature extraction
stage whether the maximum variation in the
values associated with columns of keys exceeds a
predetermined threshold value;
means for determining in the confirmation stage
whether none of the other variations in the digital sense values associated with columns of keys
exceeds a tolerance value that may depend on
said maximum variation value derived from the
columns of keys; and
means for determining in response to all of said
four immediately preceding recited determinations being affirmative that the maximum variation value derived from the rows of keys and the
maximum variation value derived from the columns of keys each resulted from only a single
key being touched instantaneously by an operator.
40. A system according to claim 38, wherein the
processing means comprises
means for processing the digital sense values iteratively by comparing the digital sense values with
baSeline digital sense values derived from digital
sense values provided when the touch sensing
array is untouched by an operator to determine
which one of three states is present, the ftrst state
being an untouched state, the second state being a
valid touch state in which a valid touch by an operator is present, and the third state being an invalid
response state which occurs when neither said ftrst
state nor said second state is present.
41. A system according to claim 40, wherein the
processing means revises the baseline digital sense values in response to sensed variations in the digital sense
values which are stable over a predetermined number of
iterations and which indicate the invalid response state
of the touch sensing array.
42. A system according to claim 41, wherein the
means for revising the baseline digital sense values comprises
means for storing a set of reserve values derived from
a set including each of the respective digital sense
values;
means for iteratively comparing each digital sense
value to each corresponding reserve value;
means for revising the stored set of reserve values to
correspond to a current set of digital sense values
whenever the difference between any digital sense
value and the corresponding reserve value exceeds
a predetermined bound value; and
means for revising the baseline digital sense values to
correspond to the reserve values when the invalid
response state is present for a predetermined number of iterations during which time no revision of
the reserve values occurs, no revision of baseline
digital sense values to correspond to reserve values
occurs and the valid touch state is not present.
43. A system according to claim 40, wherein the
processing means comprises
means for compensating for drift in the baseline digital sense values that are provided when the touch
sensing array is in the untouched state

39

4,733,222

wherein the baseline digital sense values are updated
by either incrementing or decrementing each of the
respective baseline digital sense values in accordance with whether the baseline digital sense value
is less or greater than the respective last-provided 5
digital sense value when the untouched state is
present for a predetermined number of consecutive
iterations during which time no updating of the
baseline digital sense values occurs.
44. A system according to claim 43, wherein the 10
compensating means further comprises
means for delaying by degree the updating of baseline
digital sense values when the invalid response state
occurs only intermittently.
45. A system according to claim 9, wherein the pro- 15
cessing means comprise
means for providing sense signal values in accordance
with the amplitudes of the respective sense signals;
and
means for processing the sense signal values itera- 20
tively to ascertain the values of accumulated differences in the sense signal values and for processing
the accumulated differences as a group to determine whether any said differences resulted from a
key being touched by an operator;
25
wherein the electrode array dermes rows and columns of electrodes; and
wherein the means for processing the accumulated
differences comprises
means for determining whether the maximum ac- 30
cumulated difference derived from the row electrodes exceeds a predetermined threshold value;
means for determining whether all other accumulated differences derived from the same row
electrode as said maximum accumulated differ- 35
ence derived from the row electrodes exceed a

40

45

50

55

60

65

40
threshold value which may depend on said maximum accumulated difference derived from the
row electrodes;
means for determining that no accumulated difference derived from a row electrode other than
said same row electrode exceeds a tolerance
value which may depend on said maximum accumulated difference derived from the row electrodes;
means for determining whether the maximum accumulated difference derived from the column
electrodes exceeds a predetermined threshold
value;
means for determining whether all other accumulated differences derived from the same column
electrode as said maximum accumulated difference derived from the column electrodes exceed
a threshold value which may depend on said
maximum accumulated difference derived from
the column electrodes;
means for determining that no accumulated difference derived from a column electrode other than
said same column electrode exceeds a tolerance
value which may depend on said maximum accumulated difference derived from the column
electrodes; and
means for determining, in response to such of said
six immediately preceding recited determinations that are applicable being affirmative, that
the accumulated differences derived from the
row electrodes and the accumulated differences
derived from the column electrodes resulted
from the onset of only a single one of the keys
being touched instantaneously by an operator.

* * * * *