`
`SAMSUNG EX. 1011 -1/18
`
`SAMSUNG EX. 1011 - 1/18
`
`

`

`Pramana — journal of physics
`
`Editor
`
`H R Krishnamurthy
`Indian Institute of Science, Bangalore
`
`Associate Editor
`
`Rohini M Godbole
`
`Indian Institute of Science, Bangalore
`
`Editorial Board
`
`I K BhattaChal‘jee, Indian Assoc. for the Cultivation of Science, Calcutta
`Neelima Gupte, Indian Institute of Technology, Chennai
`S Kailas, Bhabha Atomic Research Centre, Mumbai
`Dilip G Kanhere, University of Pune, Pune
`A V Khare. Institute of Physics, Bhulzaneswar
`C Manohar, Mumbai
`
`Deepak Mathut‘, Tata Institute of Fundamental Research, Mumbai
`S MOhan, Indian Institute of Science, Bangalore
`Debashis Mukherjee, Indian Assoc. for the Cultivation of Science. Calcutta
`Sunil Mukhi, Tata Institute of Fundamental Research, Mumbai
`Rajaram Nityananda. Raman Research Institute, Bangalore
`R Ramaswamy, Jawaharlal Nehru University, New Delhi
`A K Raychaudhuri, National Physical laboratory, New Delhi
`K C Rustagi, Centre for Advanced Technology, Indore
`E V Sampathkumaran, Tata Institute of Fundamental Research. Mumltai
`Abhijit Sen, Institute for Plasma Research, Gandhinagar
`Dinesh Sharma, Italian Institute of Technology, Mumbai
`R Simon, Institute of Mathematical Science, Chennai
`K Subramanian, National Centre for Radioastronomy, Pune
`C S Sundar. Indira Gandhi Centre forAtomic Research, Kalpakkam
`
`Editor of Publications of the Academy
`
`N Mukunda
`Indian Institute of Science, Bangalore
`
`
`Annual Subscription Rates
`
`All countries except India
`(Price includes AIR MAIL charges)
`India
`
`US$ 200
`
`Rs. 200
`
`Annual subscriptions for individuals for India and abroad are Rs. 100/— and $50 respectively.
`
`All correspondence regarding subscription should be addressed to the Circulation Department of
`the Academy
`Editorial Office
`
`Indian Academy of Sciences, C V Raman Avenue,
`RB. No. 8005. Bangalore 560 080, India
`
`Telephone: 80-334 2546
`Telefax: 91»80—334 6094
`
`Website: http://www.iisc.ernet.in/pramana/
`
`Email: pramana®ias.emet.in
`
`© I999 by the Indian Academy of Sciences. All rights reserved.
`
`“Information for contributors” is printed in the last issue of every volume.
`
`As part of the Gold
`Meeting on Liquid
`December 1998. St
`
`topics, which inclu
`water and biologic:
`delivered by Prof,
`About 100 particip.
`and post does selec
`for their benefit (11
`
`topics which wou
`proceedings contai
`Co-operation fror
`school and the disc
`Raman Research I]
`
`colleagues, in partic
`V A Raghunathan a
`organization of the
`crystal laboratory p
`Research Institute 1
`
`Our most gratei
`The lectures in the z
`
`Sadashiva, K A 811
`
`V Lakshminaraya
`M Muthukumar 01
`
`Hindu University, I
`Jawaharlal Nehru .(
`
`Finally, our than
`making this public;
`
`SAMSUNG EX. 1011- 2/18
`
`SAMSUNG EX. 1011 - 2/18
`
`

`

`
`
`R.N. 24935/73
`Regd. No. KRN—A 64
`
`Pramana journal of physics
`
`Vol. 53 (No. 1), July 1999
`
`Special Issue on Liquid Crystals and
`Other Soft Materials
`
`a-
`
`«:1
`
`Contents
`
`
`
`Quasi-one d1mens19nalelectrical conductivity and thermoelectric power studies
`on a discotic liquid crystal ........ V S K Balagurusamy, S Krishna Prasad
`'
`'1‘.
`S Chandrasekhar, Sandeep Kumar M Manickam and C V Yelarnaggaa’
`
`
`Surface controlled nematie bistability ............ I Dozov and G Durand
`Surface adsorptionandCollapse transition of linear polymer chains ........
`'
`
`‘Liquid water: A very complex fluid ................. H Eugene Stanley
`
`"
`
`.Yashwant Singh, Sanjay Kumar and Debaprasaa’ Giri
`
`Biology and the flow of molecular information ............A J Libchaber
`
`Phase transitions in Langmuir monolayers .
`
`. K A Suresh and A Bhatracharyya
`
`_
`
`i l
`'
`
`‘
`
`3
`
`13
`
`25
`
`37
`
`53
`
`85
`
`93
`
`(Continued on inside back cover)
`
`The lamellar and sponge phases of dilute surfactant systems: Structures and
`defects at equilibrium and under shear ................ Maurice Kleman
`
`107
`
`Phase separated composite films of liquid crystals ....................
`................. Valery Vorflusev, Jae-Hoon Kim and Satyendra Kumar
`
`Natural optical activity and liquid crystals ........ C Oldano and M Becchi
`
`An LCD for the multimedia network age: Polymer stabilized FLCD ........
`.............. Shunsuke Kobayashi, Hirokazu Furue and Taiju Takahashi
`
`. F Gerbal, VNoireaux, C Sykes,
`.
`.
`On the ‘Listeria’ propulsion mechanism .
`F Ju’licher, P Chaikin, A On, J Prost, R M Golsteyn, E Friederich,
`D Louvard, V Laurent and M F Carlier
`
`Structure and dynamics of charged macromolecules: Minimal representation of
`biological systems ................................ M Muthukumar
`
`121
`
`131
`
`145
`
`155
`
`171
`
`SAMSUNG EX. 1011 - 3/18
`
`SAMSUNG EX. 1011 - 3/18
`
`

`

`
`
`lepending neither
`precisely as seen
`1m form T/fik4 is
`z (the analogue in
`ng the curvature-
`
`a linearly stable,
`active membrane
`idence can mimic
`f the observations
`
`partures from the
`ictions, including
`we have merely
`mes. A complete
`: behaviour in the
`
`Rao for valuable
`
`v York, 1995)
`y, New York, 1982)
`
`98)
`1356—4359 (1999)
`in such a structural
`
`iponent membranes
`
`992)
`
`Oxford, 1986)
`
`Driving matrix liquid crystal displays .............. T N Ruckmongathan
`
`,
`
`Molecular structure and chiral liquid crystalline phases ...... B K Sadashiva
`
`Linear and nonlinear rheology of wormlike micelles ...................
`................ A K Sood, Ranjini Bandyopadhyay and Geetha Basappa
`
`Nonequilibrium noise and instabilities in membranes with active pumps .....
`................... Srimm Ramaswamy, John Toner and Jacques Pros!
`
`199
`
`213
`
`223
`
`237
`
`
`
`Indexed in CURRENT CONTENTS ISSN 0304-4289
`
`Edited and published by N Mukunda for the Indian Academy of Sciences, Bangalore 560 080.
`Printed at Tholasi Prints India, Bangalore
`
`
`
`SAMSUNG EX. 1011 - 4/18
`
`SAMSUNG EX. 1011 - 4/18
`
`

`

`
`
`PRAMANA
`—journal of
`physics
`
`(9 Indian Academy of Sciences
`
`Vol. 53, No. l,
`July 1999
`pp. 199—212
`
`Driving matrix liquid crystal displays
`
`T N RUCKMONGATHAN
`
`Raman Research Institute, CV. Raman Avenue, Bangalore 560080, India
`
`Abstract. Liquid crystal displays had a humble beginning with wrist watches in the seventies.
`Continued research and development in this multi-disciplinary field have resulted in displays with
`increased size and complexity. After three decades of growth in performance, LCDs now offer a
`formidable challenge to the cathode ray tubes (CRT).
`A major contribution to the growth of LCD technology has come from the developments in
`addressing techniques used for driving matrix LCDs. There are several approaches like passive
`matrix addressing, active matrix addressing and plasma addressing to drive a matrix display.
`Passive matrix LCD has a simple construction and uses the intrinsic non—linear characteristic of
`the LCD for driving. Departure from conventional line by line addressing of a passive matrix has
`resulted in improved performance of the display. Orthogonal functions have played a crucial role in
`the development of passive matrix addressing. Simple orthogonal functions that are useful for
`driving a matrix LCD are introduced. The basics of driving several rows simultaneously (multi—line
`addressing) are discussed by drawing analogies from multiplexing in communication. The impact
`of multi-line addressing techniques on the performance of the passive matrix LCDs in comparison
`with the conventional technique will be discussed.
`
`Keywords. Addressing; multiplexing.
`
`PACS Nos
`
`42.79.kr; 85.60.Bt; 85.60.pg
`
`1. Introduction
`
`Liquid crystal displays (LCDs) are the most popular among the various flat panel displays.
`LCDs operate at low voltages and need negligible power. This has led to their use in watches
`and calculators. However, limited Viewing angle characteristics, large response times and
`limitations in the electro-optic characteristics were some of the drawbacks of LCD’s for large
`information content displays like television and Video displays. Considerable research and
`development during the last two decades were devoted towards overcoming these limitations
`in LCDs. Performance of LCDs is now comparable to that of CRT. However, they cost more
`as compared to CRT. Drive techniques are an important component in these developments.
`Objective of this paper is to discuss various methods for driving a matrix LCD.
`
`2. Matrix displays and multiplexing
`
`Any flat panel display consists of an array of picture elements (pixel) arranged as a rectan-
`gular matrix. In a matrix LCD the row and column electrodes are perpendicular to each
`other. Area of intersection of the row and column electrode defines a pixel. A row electrode
`and a column electrode uniquely address a pixel as shown in figure 1. Multiplexing or
`
`199
`
`
`
`SAMSUNG EX. 1011 - 5/18
`
`SAMSUNG EX. 1011 - 5/18
`
`

`

`Driving m¢
`
` R
`
`DHDRIVR5
`
`Figure 3. Schematic of .
`+1.
`-1
`_:l:l:l:l:F
`
`:2:
`
`Ti :25:
`+1
`-1
`+1
`.1 _
`+1 .
`.1.
`
`Figure 5.
`
`1
`
`7— +1 wal(0,t)
`+ l
`.1:l:+1 wal(1,t)
`_1 fl“ wal(2,t)
`, i *1 wal(3,t)
`
`*1: wal (0,0
`_lf wal(1.l)
`:1:L__ wal(2.t)
`:blIl: we»
`+l:l:l:l:l: WW4“)
`.l:l:l:Fl:F mus.»
`.J’LFLFL warm
`+1 77 ,,
`7
`7 wal(7,t)
`-1
`
`I ‘1
`
`Figure 6. Walsh functi01
`orthogonal matrices.
`
`A set of sine (or cosi
`to each other. However,
`
`of orthogonal function:
`Transform using these (
`of these functions is jus
`circuits. Figure 5 illustr.
`frequencies decreasing (
`
`T N Ruckmongmhan
`
`/ Column electrode
`
`Pixel
`
`STNLCD
`
`N - Rows &
`M- Columns
`N.M- Pixels
`
`V10 - 'I‘llreshold voltage
`
`
`'
`v‘
`-Sntumlion -
`l
`
`90
`..
`lomg'
`lo
`90
`
`
`
`Row electrode
`
`:lo-umm-ugmnw-q-—] —>
`Applied Voltage (RMS) —>
`
`Figure 1.
`
`Schematic of a matrix LCD.
`
`
`Figure 2. Typical electro-optic characteristic
`of TN and STN LCD’s.
`
`
`matrix addressing is the technique for driving matrix displays. The term multiplexing is
`derived from communication where several signals are multiplexed over a single channel.
`
`Matrix addressing is very similar, since the information for pixels in a column is
`
`multiplexed through the column electrode. Matrix addressing may use the intrinsic non-
`
`linear electro-optic characteristics of LCD. Figure 2 shows the typical non—linear electro-
`
`optic characteristics of twisted nematic (TN) and super twisted nematic (STN) LCDs.
`
`LCDs usually have a threshold voltage below which there is no change in its optical
`
`response. Similarly there is practically no change in transmission when voltages applied to
`
`the LCD is above the saturation voltage. Light transmission through the cell varies as the
`
`voltage applied to the LCD is increased from threshold to saturation voltages. For the sake
`
`of convenience. threshold voltage by definition is the voltage when there is 10% change in
`
`light transmission from the unexcited state. Similarly the saturation voltage is the voltage
`
`when the light transmission changes by 90%. Displays using this intrinsic non-linearity are
`
`called passive matrix LCDs, in contrast to active matrix LCDs wherein an additional non-
`
`linear element such as diode or a transistor is used in conjunction with each pixel. Plasma
`addressing is yet another approach that is becoming popular for driving matrix LCDs.
`
`
`
`3. Passive matrix LCDs
`
`
`Passive matrix displays exploit the intrinsic non-linearity of the LCD [1]. Figure 3
`
`illustrates the schematic of a passive matrix LCD. An equivalent circuit of a 2 X 2 matrix
`
`LCD, when a voltage is applied between a row and column electrode while the other two
`
`electrodes are left floating is shown in figure 4. A part of the voltage applied to the
`
`selected pixel P“ will also appear across the non-selected pixels (P12, P21, P22). Therefore
`
`floating electrodes must be avoided in a matrix display to prevent such cross—talk. LCDs
`
`are slow responding devices (response time ranges from a few to few hundred milli-
`
`seconds). Hence, the RMS voltage of the applied electric field characterises the response
`of LCD rather than the instantaneous voltage across a pixel.
`
`4. Orthogonal functions
`
`A set of functions are orthogonal to each other when:
`
`/f,~(6)_fi(6)d9 = constant for i =j and is equal to zero when i 7Aj.
`
`(l)
`
`200
`
`Pramamz — J. Phys, Vol. 53, N0. 1, July 1999
`
`
`
`SAMSUNG EX. 1011 - 6/18
`
`SAMSUNG EX. 1011 - 6/18
`
`

`

`Driving matrix liquid crystal displays
`
`C11
`
`C12
`
`C21
`
`l @2”
`
`C12
`
`C11 I022
`
`
`
`
`
`
`
`Figure 3. Schematic of a passive matrix LCD.
`
`‘ Figure 4. Cross-talk in matrix LCD‘s.
`
`
`Ti '.mm
`
`..mm
`+1.
`-
`
`
`+1
`
`_, .__J——I_l——L
`+1
`
`1th
`
`Figure 5. A set of Rademachar functions which are orthogonal.
`
`. W...
`.7,
`+1§+1 wal(0,t)
`+1 wal(1 t)
`’ a w =
`+1
`._ 4
`wai(2t)
`‘
`fl ml 7 +1 wal(3,t)
`-1
`
`-1
`-1
`
`,
`
`W
`.1, a
`
`1
`1
`1
`l
`1
`-1-1 1
`_
`_
`1
`1
`1
`1
`1-11-1
`
`H = 1
`1
`I];
`
`:
`
`.
`1
`1
`1-1 '
`
`H2
`
`_
`_
`H,—H 38(11—
`4
`2
`2
`
`H H
`2
`H —H
`2
`
`2 _
`—
`
`2
`
`1
`1
`1
`1
`1-1 1-1
`1
`1 '1 '1
`1-1-1 1
`
`-
`
`wal(0,l)
`+1
`1.. 1,2,“.(1‘0
`11111111
`-1—1-1~11111
`:l:l:““'(2~°
`.
`_
`.
`.
`2:12:1«1115» _4 w = 1111111111
`+‘:|:F:|:F Wul(4.i) E 3
`1
`.1 .1
`1
`1
`-1«1
`1
`4% wal(5,t)
`-l
`l
`1 —1
`1 -l -l
`1
`m 11-11(61)
`11 i i i i j i j
`+11 n:n H mum
`
`
`
`“8 2 H2 381 "4 =
`
`l
`l
`l
`1
`l
`l
`l
`l
`1-11-11-11-1
`11-1-111-1-1
`1 '1'1 11'1‘1 1
`1
`1
`1
`1 '1‘1 '1 '1
`1
`-1
`1 -1-1 1 -1
`1
`1
`1
`-1 -1-1 -1
`1
`1
`1 -1 —1 1-1 1
`1 -1
`
`Figure 6. Walsh functions and the corresponding
`orthogonal matrices.
`
`Figure 7. Hadamard matrices.
`
`. are orthogonal
`.
`A set of sine (or cosine) functions for example sin(27rnft); n = 1,2 .
`to each other. However, Rademachar functions and Walsh functions are much simpler sets
`of orthogonal functions [2, 3]. These functions take just
`two values (+l and —l).
`Transform using these orthogonal functions need no multiplication since the amplitudes
`of these functions is just unity. Hence they are important for hardware realisation of the
`circuits. Figure 5 illustrates a set of Rademachar functions. They are square waves with
`frequencies decreasing (or increasing) by a factor of two. The Walsh functions are shown
`
`Pramana — J. Phys., Vol. 53, No. 1, July 1999
`
`201
`
`SAMSUNG EX. 1011 - 71/18
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`[LCD
`
`v10 - 'l‘llresllold voltage
`V90 - Saturation voltage
`
`> —
`
`optic characteristic
`
`In multiplexing is
`r a single channel.
`1
`in a column is
`the intrinsic non-
`non-linear electro-
`
`atic (STN) LCDs.
`mge in its optical
`voltages applied to
`3 cell varies as the
`
`tages. For the sake
`3 is 10% change in
`tage is the voltage
`ic non-linearity are
`an additional non-
`
`each pixel. Plasma
`g matrix LCDs.
`
`CD [1]. Figure 3
`t ofa 2 x 2 matrix
`vhile the other two
`
`age applied to the
`’21, P22). Therefore
`h cross-talk. LCDs
`ew hundred milli-
`
`erises the response
`
`
`
`
`SAMSUNG EX. 1011 - 7/18
`
`

`

`T N Ruckmongathan
`
`in figure 6. Orthogonality of these functions is not altered when some of these functions
`are multiplied by —1. Walsh functions are related to Hadamard matrices. It is easy to
`obtain the Walsh functions by interchanging the rows and columns of Hadamard matrices
`and multiplying some of the rows by — 1. It is easy to generate Hadamard matrices using
`kroneker product as illustrated in figure 7. Parsaval’s theorem of orthogonal transforms is
`of relevance in matrix addressing. In simple terms the Parsaval’s theorem states that the
`energy is conserved when an orthogonal transform is taken on any signal as shown in the
`following equations.
`
`
`
`1
`
`2
`
`'f/rx (t)dt=?Zan,
`
`C x 2
`11:0
`
`wherein a,,’s are the transform coefficients and C is a constant.
`
`5. Multiplexing and demultiplexing
`
`Multiplexing is a technique for transmitting more than one signal over a channel. Figure 8
`shows the general block diagram of multiplexing and de-multiplexing. Each signal is
`multiplied with a carrier. The sum of these multiplied signals is transmitted over a single
`channel. At the receiving end the multiplexed signal is multiplied with carriers that are
`replica of the carriers at the transmitting end. Signals are recovered back by integrating the
`output of the multipliers. The carriers are orthogonal to each other. Let f1,f2, .
`.
`.
`, fN be a
`set of N orthogonal functions. Let d1, d2, .
`.
`. ,dN be the data to be transmitted. Multiplexed
`signal is given by (f1.d1 +f2.d2 + -
`-
`- +fN.dN). At the receiving end di can be recovered
`by multiplying the multiplexed signal with f,- followed by integration. This will result in
`
`/fi.(fi.d1+f2.d2+---+fN.dN)dt:/f,.2d,-dt.
`Integral off? is unity by the definition of orthogonal function (see eq. (1)). Hence the
`signal or data d,- is recovered.
`Multiplexing and de—multiplexing may also be illustrated using matrix algebra. Let 0
`be the orthogonal matrix. Let D be the data to be multiplexed. Multiplexing is equivalent
`to taking the orthogonal transform of the data, that is M 2 0D. Data can be recovered
`
`Integrator
`
`
`
`FV—J
`blultiplgxiug
`
`\fi,_/
`Demultiplexiug
`
`Figure 8.
`
`Schematic multiplexing and demultiplexing.
`
`202
`
`Pramana — J. Phys., Vol. 53, N0. 1, July 1999
`
`Driving 1
`
`ll
`
`HHD—i—I
`
`1
`
`D—lI—IF-‘l-l
`
`lD:—
`40
`
`Figure 9.
`
`back by multiplying
`M : 0‘1.0.D = ID
`
`Here a four by four (
`of the matrix is also
`wherein n is the orde
`
`6. Multiplexing mat
`
`A matrix LCD consi:
`passive matrix LCD
`upon how the matri)
`scanning electrodes
`Repetitive periodic w
`are orthogonal to eacl
`serve the same functi
`assume that the rows
`
`Pixels are assigned a (
`pixels getting a higl
`Orthogonal transform
`
`7. Inverse transform
`
`As seen above the im
`functions are inherent
`
`. ,fN be
`Letfl,f2,. .
`displayed in a given ct
`fN-dN) Let k~f1, k -f2,-
`matrix display. The rc
`
`VRMS =
`
`
`
`VRMs =
`
`
`
`
`
`
`SAMSUNG EX. 1011 - 8/18
`
`
`
`SAMSUNG EX. 1011 - 8/18
`
`

`

`
`
`: of these functions
`
`
`
`
`
`
`
`
`
`trices. It is easy to
`Hadamard matrices
`
`nard matrices using
`:gona] transforms is
`3mm states that the
`nail as shown in the
`
`a channel. Figure 8
`ing. Each signal
`is
`mitted over a single
`ith carriers that are
`
`:k by integrating the
`:tf],f2,...,fN be a
`.mitted. Multiplexed
`di can be recovered
`This will result in
`
`eq. (1)). Hence the
`
`atrix algebra. Let 0
`ulexing is equivalent
`ta can be recovered
`
`
`
`
`
`Driving matrix liquid crystal displays
`
`1 111
`__ 11-1-1
`0" 1-1-11
`l-l l-l
`
`a
`b
`’1): c
`d
`
`a+b+c+d
`a+b-c-d
`‘M‘OD‘ a-b-c+d
`a-b+c-d
`
`a
`1111 a+b+c+d
`b
`_1.1 _111.1_1a+b-c-d
`D‘TO M_71_1_11a-b-c+d:c
`1-11-1 a-b+c-d
`(1
`
`Figure 9.
`
`Illustration of multiplexing and demultiplexing using orthogonal matrix.
`
`back by multiplying M by the inverse of the orthogonal transform matrix 0. D : 0".
`M : 0‘1.0.D = ID, where I is the unit matrix. An example is illustrated in figure 9.
`Here a four by four orthogonal matrix is used for multiplexing four signals. The inverse
`of the matrix is also the same, since the matrix is symmetric. A scaling factor (IA/H),
`wherein n is the order of the square matrix is used to ensure that 0.0—1 is unit matrix.
`
`6. Multiplexing matrix LCDs
`
`A matrix LCD consists of row electrodes and column electrodes. An RMS responding
`passive matrix LCD is a crude form of demultiplexer as explained below. Depending
`upon how the matrix is scanned one set of electrodes (row or column) is called the
`scanning electrodes while the other set of electrodes is called the data electrodes.
`Repetitive periodic waveforms are applied to the scanning electrodes. These waveforms
`are orthogonal to each other and are independent of the information to be displayed. They
`serve the same function as carrier in communication. For the sake of simplicity we can
`assume that the rows are scanned and data is multiplexed through the column electrodes.
`Pixels are assigned a data +1 for OFF pixels and —l for ON pixels. This will result in ON
`pixels getting a higher root-mean-squared voltage as compared to the OFF pixels.
`Orthogonal transform of the data in a column is applied as the column waveform.
`
`7. Inverse transform using RMS responding device
`
`As seen above the inverse transform requires multiplication and integration. Both these
`functions are inherent in a root-mean square responding device like LCD as shown below.
`Letf1,f2,. .
`. ,fN be a set ofN orthogonal functions. Let d1, d2, .
`.
`. 7alN be the data to be
`displayed in a given column. Column waveform for this column is (fl .dl +f2.d2 + -
`.
`- +
`fN.dN) Let k.fl, k .fz, .
`.
`. ,k.fN be the N orthogonal waveforms applied to the rows of the
`matrix display. The root mean squared voltage across a pixel in row i will be
`'VRMs : \/</(kfi — (fl-(1'1 +f2-d2 + ‘ ‘ +fN-lel2dt>
`
`
`VRMS = \//(k2f;2)dt ~ 2k /f;~2didl + /(fl.d1 +f2.d2 + '
`'
`' +fN.dN)2dI
`
`Pramana — J. Phys, Vol. 53, N0. 1, July 1999
`
`203
`
`
`
`SAMSUNG EX. 1011 - 9/18
`
`SAMSUNG EX. 1011 - 9/18
`
`

`

`T N Ruckmongathan
`
`VRMS z . i/szizdt—Zk/diffdt+N since /(d,—f,—)2dr is 1.
`
`First and last term in the above expression are constants and the middle term is the product
`term useful for demultiplexing. Ratio of root-mean—square voltage across an ON pixel to
`that across an OFF pixel is called the selection ratio. This is a measure of the discrimination
`between the ON and OFF pixels and is an indirect method of estimating the contrast in the
`display. A high selection ratio is desirable to obtain a high contrast in the display.
`For the maximum selection ratio the first term is equal to the last term. Hence k = x/N
`RMS voltage across the pixel will be proportional to (2N — 2\/Nd,-) and the maximum
`(SR) is
`
`SR:
`
`(WT/+1)/(\/1VA1).
`
`(2)
`
`Selection ratio is infinite for N = l and rapidly falls as N increases. For example
`SR 2 1.105 when N = 100, which means that
`there is just 10% difference in RMS
`voltages between ON and OFF pixels.
`
`8. Line by line addressing
`
`One of the early driving scheme is the line by line addressing [4] where in the rows of a
`matrix display are sequentially selected one at a time. Figure 10 illustrates the conven-
`tional line by line addressing of passive matrix LCD. Orthogonal matrix used here is a
`unit matrix with 1 as the diagonal elements. Column signal is the data itself as the
`orthogonal transform of the data is same as the data. The corresponding row and column
`waveforms are shown in figure 10. Both the row and column waveforms are inverted after
`a time period T, when addressing is complete. This ensures that no dc field is present
`across the pixels in the display. While a display may be driven by dc fields the life of the
`display will be reduced due to migration of ions towards the electrodes and slow
`electrochemical reactions in the display cell. The line by line addressing technique gives
`
`
`
`oooecoet—
`
`OOOOOOHO
`
`ccoccwco
`
`9963666:
`
`cor—coco:
`
`61-1606950
`
`Orthogonal
`matrix
`
`
`ocear—teoo
`
`>1
`+1
`- l
`+1
`+1
`
`+1
`
`+1
`
`|
`_
`
`-1
`+1
`- 1
`+1
`+1
`
`+1
`
`+1
`
`Image
`matrix
`
`/ Row waveforms
`
`n
`
`U
`
`
`HGGOOOOO
`
`Column waveform J
`
`Figure 10.
`LCD.
`
`Illustration of the conventional line by line addressing of a passive matrix
`
`204
`
`Pramana — J. Phys, Vol. 53, No. 1, July 1999
`
`
`
`Driving
`
`a maximum selectic
`waveform to amplitt
`that are multiplexed
`The instantaneous V(
`of the liquid crystal
`tained below thresho
`LCDs are slow de‘
`milliseconds. Hence
`Period (T) of the addr
`times of the LCD. H
`times the period may
`tional line by line add
`contrast in the displaj
`
`9. Frame response
`
`Under optimum cond
`waveforms deliver or
`
`delivered through tht
`energy from the row
`
`threshold voltage). Tl
`poor contrast. Frame
`
`One of the techniq
`technique, described 1
`
`10. Active addressin
`
`Active addressing use:
`all
`the rows are siml
`functions. The columr
`
`data to be displayed. E
`
`Waveform a(
`a pixel.
`
`in
`
`L‘
`
`'1
`
`*1
`
`Figure 11.
`
`
`
`SAMSUNG EX. 1011 -10/18
`
`SAMSUNG EX. 1011 - 10/18
`
`

`

`dt is 1.
`
`le term is the product
`:ross an ON pixel to
`of the discrimination
`
`
`
`ng the contrast in the
`in the display.
`
`erm. Hence k = W.
`,-) and the maximum
`
`
`
`(2)
`
`reases. For example
`difference in RMS
`
`
`
`here in the rows of a
`
`lustrates the conven-
`
`natrix used here is a
`he data itself as the
`
`
`
`
`10 dc field is present
`: fields the life of the
`electrodes and slow
`
`ssing technique gives
`
`iing row and column
`rms are inverted after
`
`
`
`
`sing of a passive matrix
`
`Driving matrix liquid crystal displays
`
`a maximum selection ratio when the ratio of the amplitude of the pulses in the row
`waveform to amplitude of the column waveform is x/IV, where N is the number of rows
`that are multiplexed in a matrix display. The maximum selection ratio is given by eq. (2).
`The instantaneous voltage across even an OFF pixel is higher than the threshold voltage
`of the liquid crystal display. However the RMS voltage across the OFF pixels is main-
`tained below threshold voltage.
`LCDs are slow devices with response times in the range of a few tens to few hundred
`milliseconds. Hence the RMS voltage is important in determining the state of the pixel.
`Period (T) of the addressing waveforms is assumed to be small as compared to the response
`times of the LCD. However in a large matrix display or in a display with fast response
`times the period may become comparable to the response time of the LCD. The conven-
`tional line by line addressing is no longer suitable to drive such a display since the resulting
`contrast in the display is poor(low) due to frame response phenomenon explained below.
`
`9. Frame response
`
`Under optimum conditions for multiplexing (when the selection ratio is maximum), row
`waveforms deliver one half of the energy to the pixel while the other half of energy is
`delivered through the column waveforms. In a line by line addressing technique the
`energy from the row waveform is delivered by a single pulse (which is larger than the
`threshold voltage). This results in turning even the OFF pixels partially ON resulting in
`poor contrast. Frame response phenomenon [5] is illustrated in figure 11.
`One of the techniques proposed for suppressing frame response is active addressing
`technique, described below.
`
`10. Active addressing
`
`Active addressing uses Walsh functions as row waveforms [6]. As illustrated in figure 12
`all
`the rows are simultaneously selected by row waveforms corresponding to Walsh
`functions. The column waveform is generated by taking the orthogonal transform of the
`data to be displayed. Selection ratio is maximum when amplitude of the row waveforms
`
`Waveform across
`a pixel.
`
`Pulse amplitude for an ON pixel
`Pulse amplitude for an OFF
`
`pixel
`l
`
`
`
`Light transmission through ON pixel
`Light transmission through OFF pixel
`
`
`
`Time —>
`
`Figure 11.
`
`Frame response phenomenon associated with line by line addressing.
`
`Pramana — J. Phys., Vol. 53, No. 1, July 1999
`
`205
`
`
`
`SAMSUNG EX. 1011 - 11/18
`
`
`
`
`SAMSUNG EX. 1011 - 11/18
`
`

`

`Driving
`
`11. Multi-line add]
`
`A more elegant solt
`simultaneously ratht
`selection or multi-
`proposed in 1988 is
`is treated to be a var
`been shown that L =
`
`row waveform is eq]
`lower than Lopt ther
`waveforms. The col]
`Thus L = x/IV is t
`electronics. Selectioi
`waveforms is \/1V/L
`the number of lines
`selected and driven :
`an ON pixel to that
`However, selecting s
`of the instantaneous
`
`amplitude pulse in t
`amplitude pulses in 3
`lines that are driver
`figures 13 and 14. S
`response as shown it
`
`J'l
`
`/ J1
`Row
`J
`waveforms J
`\
`
`Column —)
`waveform
`
`:EJ‘
`
`Figure 13.
`time (L =
`
`T N Ruckmongathan
`
`Row waveforms
`7171‘]
`+1 0
`+lrl .1
`‘2 0
`Jun
`18
`”+1.1, 720
`+141“
`+2 0
`rl+l+1
`+2 0
`tl~l+1
`+6 ll
`1
`ll
`«1441
`
`+|l++ll+
`
`l++|+ll+
`
`+|+II+I+
`
`|+l+l+l+
`
`++++++++
`
`IIII++++
`
`++llll++
`
`II++II++
`
`
`
`L
`Orthogonal
`
`
`matrix
`Column waveforms/
`\Dfl_l:ln—
`
`'ON'mpixel
`
`Figure 12.
`
`Illustration of active addressing of a passive matrix LCD.
`
`
`is (1 / W) Va, wherein VC is the maximum amplitude in the column Waveforms and N is
`
`the number of rows that are being multiplexed. Frame response is generally suppressed
`
`the rows are selected. However as one can see in figure 12, when image
`when all
`
`displayed in a column is same as on one of the orthogonal functions, then the column
`
`waveform has a maximum amplitude for one time interval and zero for rest of the time
`
`intervals. Then waveform across the pixel will be similar to line by line addressing. This
`
`condition is avoided by inverting some of the orthogonal functions and limiting the
`
`amplitude of the column voltage. While these practical measures are available for
`
`suppressing the frame response, selecting all the rows simultaneously is not preferable for
`
`the following reasons:
`
`— Computation necessary for generating the column signal is very high (N.M additions or
`subtractions for one row select time and approximately N .N .M additions for a frame).
`
`— The. number of voltage levels in the column waveforms is (N + 1), hence the hardWare
`
`complexity of column drivers is high.
`
`— A frame buffer is necessary for computing the column waveforms.
`
`— Access time of such a buffer has to be very short, since the whole image has to be
`accessed within a row select time (IO—100 microseconds).
`
`
`
`
`While active addressing is not the best solution for suppressing frame response, selecting
`all the rows simultaneously is advantageous in some applications discussed below:
`
`Addressing waveforms may be simplified to have just two voltage levels. The binary
`
`addressing technique (BAT) [7] is one such example, where in all the rows are selected
`
`simultaneously when the number of rows is small (N < 9). Although the selection ratio is
`
`lower than that of eq. (2) it is adequate to have good contrast in small matrix displays.
`
`Binary addressing technique needs the lowest power supply as compared to all
`the
`
`techniques known so far. Hence it is an attractive option to be used in applications like
`
`mobile telephones and calculators. A high contrast ratio and low supply voltage (both
`
`independent of matrix size) has been obtained by selecting all the rows simultaneously
`
`[13]. Here, the information displayed is restricted such that the number of pixels carrying
`
`information in each column is a constant. This approach is useful for addressing displays
`
`in oscilloscopes and logic analysers.
`
`206
`
`Pramana — J. Phys., Vol. 53, N0. 1, July 1999
`
`
`
`SAMSUNG EX. 1011 -12/18
`
`SAMSUNG EX. 1011 - 12/18
`
`

`

`
`
`x LCD.
`
`waveforms and N is
`
`enerally suppressed
`'e 12, when image
`IS, then the column
`for rest of the time
`
`,ne addressing. This
`is and limiting the
`s are available for
`
`is not preferable for
`
`h (N .M additions or
`iitions for a frame).
`hence the hardWare
`
`9.
`
`)le image has to be
`
`6 response, selecting
`:ussed below:
`
`3 levels. The binary
`ie rows are selected
`the selection ratio is
`
`tall matrix displays.
`ompared to all
`the
`in applications like
`upply voltage (both
`'ows simultaneously
`er of pixels carrying
`addressing displays
`
`Driving matrix liquid crystal displays
`
`11. Multi-line addressing
`
`A more elegant solution to suppress frame response is to select and address a few rows
`simultaneously rather than all the rows [8—12]. This approach is referred to as multi—line
`selection or multi—line addressing. The improved hybrid addressing technique [8]
`proposed in 1988 is one such scheme wherein the number of rows simultaneously driven
`is treated to be a variable (L) and can range from two to any desired value (L > 1). It has
`been shown that L : x/lT/ is an optimum condition when the maximum amplitude of the
`row waveform is equal to the maximum amplitude of the column waveforms. When L is
`lower than Lopt then the row waveforms have a higher amplitude as compared to column
`waveforms. The column waveforms have a higher amplitude when L is greater than WV.
`Thus L = MN is
`the condition for minimum power supply voltage of the drive
`electronics. Selection ratio is a maximum (given by eq. (2)) when amplitude of the row
`waveforms is x/N/L times the maximum amplitude in the column waveforms, where N is
`the number of lines that are being multiplexed and L is the number of rows that are
`selected and driven simultaneously. The selection ratio, the ratio of RMS voltage across
`an ON pixel to that across an OFF pixel is the same as that of line by line addressing.
`However, selecting several rows simultaneously has the advantage of reducing amplitudes
`of the instantaneous voltages across the pixel. In other words, the energy in one large
`amplitude pulse in the line by line addressing is now distributed in a number of small
`amplitude pulses in a frame. The actual number of pulses will depend upon the number of
`lines that are driven simultaneously. Multi-line addressing technique is illustrated in
`figures 13 and 14. Selecting a few rows simultaneously is adequate to suppress frame
`response as shown in table 1.
`
`I F
`L.
`L
`
`I
`r
`
`L
`L
`_
`
`[:1
`
`
`
`L
`/I
`J '1
`Ru“.
`_
`waveforms
`
`_J
`\ 1
`l—
`
`_|
`
`L
`J‘L‘ __,J‘.
`
`Column —)
`waveform
`
`fl
`
`J
`J
`‘Lr‘: “LL
`
`ON Pixel
`(high rnis voltage)
`‘1.
`
`Mbgrf’q‘l
`0”le
`‘ (luwrms voltage)
`
`[Subgroupl
`-
`7
`
`7|
`lSubgroUNiH)
`
`10..010..u
`10 .
`. 0.1 o. .0
`0104:0104;
`0 l 0 '0 0_10 _ 0
`.
`I
`'
`I
`'
`.
`‘
`.
`.
`.
`.
`.
`.
`.
`.
`0 0 0 010 0 00 1
`0 0 o 010 0 00 1
`Sparse orthogonal matrix
`used for driving two rows
`simultaneously
`
`Figure 13.
`time (L = 2).
`
`Improved hybrid addressing technique (IHAT), for driving two rows at a
`
`VI'—
`
`n n n
`
`n
`
`W‘JUL’Lu—‘Lml—
`0* U‘LM—u—"nru—u-
`‘U‘u‘u—MUHI—‘Hr
`-Vr _ wLn—n—u—le—
`Vc
`U
`U
`u It U
`
`Vc/2\~j:
`,.,
`_
`-Vc/2 /
`-Vc
`
`]
`
`Subgroup i
`
`Subgroup (i+1)
`
`
`firs/E
`
`
`Vc
`3
`
`Illustration of row and c