U.S. patent application number 10/182297 was filed with the patent office on 2003-08-07 for display device with multiple row addressing.
Invention is credited to Bonny, Jean-Daniel, Eugster, Andreas Carl, Guitard, Patrice, Hirsch, Stefan.
Application Number | 20030147017 10/182297 |
Document ID | / |
Family ID | 8171016 |
Filed Date | 2003-08-07 |
United States Patent
Application |
20030147017 |
Kind Code |
A1 |
Bonny, Jean-Daniel ; et
al. |
August 7, 2003 |
Display device with multiple row addressing
Abstract
A device for multiple row addressing is driven with pulse
patterns based on sets of 8 (or more) orthogonal functions which
have a less varying frequency content than pulse patterns based on
a set of 8 Walsh functions.
Inventors: |
Bonny, Jean-Daniel;
(Fullinsdorf, CH) ; Eugster, Andreas Carl;
(Allschwil, CH) ; Guitard, Patrice; (Hegenheim,
FR) ; Hirsch, Stefan; (Lorrach, DE) |
Correspondence
Address: |
THOMAS HOXIE
NOVARTIS, CORPORATE INTELLECTUAL PROPERTY
ONE HEALTH PLAZA 430/2
EAST HANOVER
NJ
07936-1080
US
|
Family ID: |
8171016 |
Appl. No.: |
10/182297 |
Filed: |
July 26, 2002 |
PCT Filed: |
January 25, 2001 |
PCT NO: |
PCT/EP01/00843 |
Current U.S.
Class: |
349/34 |
Current CPC
Class: |
G09G 3/3625 20130101;
G09G 2310/06 20130101 |
Class at
Publication: |
349/34 |
International
Class: |
G02F 001/133 |
Foreign Application Data
Date |
Code |
Application Number |
Feb 15, 2000 |
EP |
00200508.0 |
Claims
1. A display device comprising a liquid crystal between a first
substrate provided with row or selection electrodes and a second
substrate provided with column or data electrodes, in which
overlapping parts of row and column electrodes define pixels, drive
means for driving the column electrodes in conformity with an image
to be displayed, and drive means for driving the row electrodes
which, in the operating condition, sequentially supply groups of p
row electrodes with p mutually orthogonal signals, characterized in
that the mutually orthogonal signals are obtained from at least two
types of orthogonal functions having four elementary units of time,
within which four elementary units of time one pulse time each time
has a polarity which is different from that of the other
pulses.
2. A display device as claimed in claim 1, characterized in that
the orthogonal signals are obtained from orthogonal functions
having four elementary units of time, within which four elementary
units of time the pulse having a polarity which differs from that
of the other pulses each time shifts by one elementary unit of
time.
3. A display device as claimed in claim 1 or 2, characterized in
that the orthogonal signals are obtained from orthogonal functions
having four elementary units of time which, viewed in a time
sequence, are situated one after the other.
4. A display device as claimed in claim 3, characterized in that at
least two orthogonal signals have opposed DC contents.
5. A display device as claimed in claim 1 or 2, characterized in
that the orthogonal signals are obtained from orthogonal functions
having four elementary units of time, in which the elementary units
of the orthogonal functions are interwoven.
6. A display device as claimed in claim 1 or 2, characterized in
that p=4, and in that four orthogonal signals have identical DC
contents and four are free from a DC voltage.
7. A display device as claimed in claim 6, characterized in that
the DC content of 2 orthogonal signals of the orthogonal signals
having an identical DC content is opposed to that of the two other
orthogonal signals.
8. A display device as claimed in claim 1 or 2, characterized in
that the drive means invert the orthogonal signals after each frame
period.
Description
[0001] The invention relates to a display device comprising a
liquid crystal material between a first substrate provided with row
or selection electrodes and a second substrate provided with column
or data electrodes, in which overlapping parts of the row and
column electrodes define pixels, drive means for driving the column
electrodes in conformity with an image to be displayed, and drive
means for driving the row electrodes which, in the operating
condition, sequentially supply groups of p row electrodes with p
mutually orthogonal signals. Such display devices are used in, for
example, portable apparatuses such as laptop computers, notebook
computers and telephones.
[0002] Passive-matrix displays of this type are generally known
and, for realizing a high number of lines, they are increasingly
based on the STN (Super-Twisted Nematic) effect. An article by T.
J. Scheffer and B. Clifton "Active Addressing Method for
High-Contrast Video Rate STN Displays", SID Digest 92, pp. 228-231
describes how the phenomenon of "frame response" which occurs with
rapidly switching liquid crystal materials is avoided by making use
of "Active Addressing". In this method, all rows are driven
throughout the frame period with mutually orthogonal signals, for
example, Walsh functions. The result is that each pixel is
continuously excited by pulses (in an STN LCD of 240 rows: 256
times per frame period) instead of once per frame period. In
"multiple row addressing", a (sub-)group of p rows is driven with
mutually orthogonal signals. Since a set of orthogonal signals,
such as Walsh functions, consists of a plurality of functions which
is a power of 2, i.e. 2.sup.S, p is preferably chosen to be equal
thereto as much as possible, i.e. generally p=2.sup.S (or also
p=2.sup.S-1). The orthogonal row signals F.sub.i(t) are preferably
square-wave shaped and consist of voltages +F and -F, while the row
voltage is equal to zero outside the selection period. The
elementary voltage pulses from which the orthogonal signals are
built up are regularly distributed across the frame period. In this
way, the pixels are then excited 2.sup.S (or (2.sup.S-1)) times per
frame period with regular intermissions instead of once per frame
period. Even for low values of p such as p=3 (or 4) or p=7 (or 8)
the frame response appears to be suppressed just as satisfactorily
as when driving all rows simultaneously, such as in "Active
Addressing", but it requires much less electronic hardware.
[0003] However, it appears that, notably for Walsh functions, the
frequency content of the functions from a complete set of functions
is greatly different. Since the dielectric constant of liquid
crystalline material is frequency-dependent, this may cause the
liquid crystalline material to react differently at different
positions in, for example, a matrix display, dependent on the image
contents. This leads to artefacts in the image such as different
forms of crosstalk.
[0004] It is, inter alia, an object of the invention to provide a
display device of the type described above, in which a minimal
number of artefacts occurs in the image.
[0005] To this end, a display device according to the invention is
characterized in that the mutually orthogonal signals are obtained
from at least two types of orthogonal functions having four
elementary units of time, within which four elementary units of
time one pulse each time has a polarity which is different from
that of the other pulses.
[0006] It is found that orthogonal signals can thereby be generated
which differ little in frequency content and thus do not give rise
or hardly give rise to artefacts in the image. Such orthogonal
signals are obtained, for example, from orthogonal functions having
four elementary units of time, within which four elementary units
of time the pulse having a polarity which differs from that of the
other pulses each time shifts by one elementary unit of time. The
use of four elementary units of time has the additional advantage
that the number of column voltage levels remains limited to five,
while this number is six when using, for example, three elementary
units of time, within which three elementary units of time one
pulse having a polarity which differs from that of the other pulses
shifts by only one unit of time. A larger number of column voltage
levels to be used of course leads to more expensive drive
electronics.
[0007] These and other aspects of the invention are apparent from
and will be elucidated with reference to the embodiments described
hereinafter.
[0008] In the drawings:
[0009] FIG. 1 shows diagrammatically a display device in which the
invention is used, and
[0010] FIGS. 2 and 3 show sets of 4 and 8 Walsh functions,
respectively, and orthogonal signals derived therefrom for the
purpose of multiple row addressing, while
[0011] FIG. 4 shows another set of four orthogonal functions
according to the invention, and orthogonal signals derived
therefrom for the purpose of multiple row addressing, and
[0012] FIG. 5 shows a generalization of FIG. 4, while
[0013] FIGS. 6 and 7 show some orthogonal signals according to the
invention, derived from FIG. 5, for the purpose of multiple row
addressing.
[0014] FIG. 1 shows a display device comprising a matrix 1 of
pixels at the area of crossings of N rows 2 and M columns 3 which
are provided as row and column electrodes on facing surfaces of
substrates 4, 5, as can be seen in the cross-section shown in the
matrix 1. The liquid crystal material 6 is present between the
substrates. Other elements such as orientation layers, polarizers,
etc. are omitted for the sake of simplicity in the
cross-section.
[0015] The device further comprises a row function generator 7 in
the form of, for example, a ROM for generating orthogonal signals
F.sub.i(t) for driving the rows 2. Similarly as described in said
article by Scheffer and Clifton, row vectors driving a group of p
rows via drive circuits 8 are defined during each elementary time
interval. The row vectors are written into a row function register
9.
[0016] Information 10 to be displayed is stored in a p.times.M
buffer memory 11 and read as information vectors per elementary
unit of time. Signals for the column electrodes 3 are obtained by
multiplying the then valid values of the row vector and the
information vector during each elementary unit of time and by
subsequently adding the p obtained products. The multiplication of
the values which are valid during an elementary unit of time of the
row and column vectors is realized by comparing them in an array 12
of M exclusive ORs. The addition of the products is effected by
applying the outputs of the array of exclusive ORs to the summing
logic 13. The signals 16 from the summing logic 13 drive a column
drive circuit 14 which provides the columns 3 with voltages
G.sub.j(t) having p+1 possible voltage levels. Every time, p rows
are driven simultaneously, in which p<N ("multiple row
addressing"). The row vectors therefore only have p elements, as
well as the information vectors, which results in a saving of the
required hardware such as the number of exclusive ORs and the size
of the summing circuit, as compared with the method in which all
rows are driven simultaneously with mutually orthogonal signals
("Active Addressing").
[0017] As stated in the opening paragraph, it is possible to use
less drive electronics by choosing p to be low, for example, in the
range between 3 and 8. FIG. 2 shows a frequently used set of
orthogonal functions referred to as Walsh functions (FIG. 2a) and
the pulse patterns derived therefrom for the purpose of multiple
row addressing (FIG. 2b), with p=4. It is clear that the frequency
content of the lumped functions, or the number of sign changes
within the derived pulse patterns, greatly differs for each one of
the different functions. The first function (1) comprises DC
components, because the lumped function consists of half a period
of a square wave, whereas the other functions do not comprise any
DC component. The second function (2) comprises, within one period,
a (square) wave with the double frequency of the first function.
The fourth function (4) is doubled in frequency again with respect
to the second function, while the third function (3) is a shifted
variant of the fourth function. Even when the first function is not
used to avoid DC effects, there is a great difference in frequency
content of the three remaining functions. The dielectric constant
of the liquid crystal material is frequency-dependent so that,
dependent on the image contents, the use of such functions may lead
to artefacts such as crosstalk. The same applies when using Walsh
functions (FIG. 3a) and the pulse patterns derived therefrom for
the purpose of multiple row addressing (FIG. 3b), with p=8.
[0018] FIG. 4 shows another set of four orthogonal functions (FIG.
4a) and the pulse patterns derived therefrom for the purpose of
multiple row addressing (FIG. 4b), with p=4. The frequency content
of the lumped functions, or the number of sign changes within the
pulse patterns derived therefrom is now substantially the same for
each one of the different functions. This set is obtained by
shifting the negative pulse each time by one position in the second
and subsequent functions. Since such a set, in which the
sign-different pulse is each time shifted by one position, is very
attractive, this function is shown in a generalized form in FIG. 5
for p pulses consisting of one negative pulse and (p-1) positive
pulses, with the negative pulse being shifted each time by one
position in the second and subsequent functions. The positive
pulses have an amplitude A.sub.p and the negative pulses have an
amplitude A.sub.n. To be mutually orthogonal, it holds for the two
functions that their product, summed over a period of the duration
of the set must be zero, or:
-2 An.multidot.Ap+(p-2).multidot.A.sub.p.sup.2=0; which yields
A.sub.n=A.sub.p.multidot.(p-2)/2 (1)
[0019] In addition, the effective value of the function must be 1
(normalized for the function F). This leads to 1 A n 2 + ( p - 1 )
A p 2 p = 1 ( 2 )
[0020] It follows from (1) and (2) for A.sub.p and A.sub.n that
A.sub.p=2/{square root}{square root over (p)} and 2 A n = p - 2 p
,
[0021] respectively. For p=4 it holds that A.sub.p=A.sub.n=1 and
the number of possible column voltages is 5. This is higher for
other values; for p=3, the number of possible column voltages is 6,
namely (-5/2)A.sub.p, (-3/2)A.sub.p, (-1/2)A.sub.p, (1/2)A.sub.p,
(3/2)A.sub.p en (5/2)A.sub.p. However, when using Walsh functions,
the number of required column voltage levels would be 4 for p=3 (a
subject chosen from a set of 4 Walsh functions).
[0022] The invention is based on the recognition that orthogonal
functions are selected as starting points based on mutually
orthogonal signals obtained from at least two types of orthogonal
functions with four elementary units of time, as is shown in FIG.
4. Starting from the functions of FIG. 4, these are repeated, for
example, after 4 elementary units of time (patterns (1), (2), (3)
and (4) in FIG. 6) or inverted and repeated (patterns (5), (6), (7)
and (8) in FIG. 6). Although there is still some variation of the
frequency content, these functions surprisingly appear to give less
rise t6 artefacts than the set of 8 Walsh functions, while the
number of required column voltages remains the same, namely 9.
[0023] The pulse patterns derived from (1), (2), (3) and (4)
comprise a DC component. To reduce its influence, preferably 2 of
these pulse patterns in a set to be chosen are inverted (the DC
content is now opposed). For a completely DC-free drive, all
signals from the used set are inverted after each frame period.
[0024] This set is denoted as K8(5,r) (Kuijk function) because in
the fifth (5,*) pattern, the negative pulse starts in the second
half period with a negative pulse (at the fifth position) which
shifts to the right (5,r) in the subsequent patterns. FIG. 7 shows
the Kuijk function K8(7,r). It holds for both Figures that the
pulse patterns derived from the patterns (5), (6), (7) and (8) are
DC-free. Overall, 8 of these sets can be formed in this way, namely
K8(5,r), K8(6,r), K8(7,r), K8(8,r), K8(5,1), K8(6,1), K8(7,1) and
K8(8,1) in which 1 indicates that the negative pulse starts in the
second half period with a negative pulse (at the specified
position) which shifts to the left in the subsequent patterns.
[0025] The set of K(uijk) functions can be further extended by
mixing, as it were, the two types of orthogonal functions shown in
FIG. 4 with four elementary units of time. FIG. 8 shows such a set
K8(3,r). The pattern (1), in FIG. 8 is obtained by inserting
pattern (1) of FIG. 4a again from the third position of pattern 1
(indicated as b in FIG. 8) and by subsequently completing pattern
(1). The patterns (5), (6), (7) and (8) in FIG. 8 are obtained by
inserting into the patterns (1), (2), (3) and (4) of FIG. 4 in the
inverted form a pattern b. In this way pattern b and pattern a are
interwoven, as if were. Patterns (2), (3) and (4) are obtained by
shifting a negative pulse to the right within both part b and part
a (formel by the two other parts). The pulse patterns derived from
the patterns (5), (6), (7) and (8) in FIG. 8 are now again DC-free.
Since this insertion can take place at four positions (elementary
units of time) and the negative pulse can shift to the right and to
the left, the possible number of functions based on pattern (1) of
FIG. 3 is multiplied by 8. Since said inversion is also possible
for the functions (2), (3) and (4) of FIG. 3, the total possible
number of K(uijk) functions is 840.
[0026] The invention is of course not limited to the embodiments
shown. Similarly as described above, more than 2 functions of FIG.
4 can be combined to obtain drive patterns with, for example,
p=16.
[0027] The protective scope of the invention is not limited to the
embodiments described. The invention resides in each and every
novel characteristic feature and each and every combination of
characteristic features. Reference numerals in the claims do not
limit their protective scope. The use of the verb "to comprise" and
its conjugations does not exclude the presence of elements other
than those stated in the claims. The use of the article "a" or "an"
preceding an element does not exclude the presence of a plurality
of such elements.
* * * * *