KR20000077168A - A driving scheme for liquid crystal display - Google Patents

Abstract

The present invention relates to a drive system for a liquid crystal display of any arrangement and dimension consisting of a matrix building block having orthogonal and displacement orthogonality (SO) characteristics.

The driving system uses an Mx (N + 1) M-dimensional para-unitary matrix as an orthogonal building block.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a driving system for a liquid crystal display,

The present invention relates to a driving system for a liquid crystal display.

In order to drive a liquid crystal display, a passive matrix driving system is generally adopted. Use minimum crosstalk, deviations such as APT and IAPT.

However, even in these improved methods, manual driving still results in display monotonization and low contrast. In these high-mux displays with fast responding liquid crystals, the problem of contrast loss due to frame response is significant. To overcome this problem, an activity addressing using an orthogonal Hadamard matrix as a common driving signal is proposed. The frame response effect is minimized by selecting each pixel over the entire frame. However, this method results in a high degree of computation and memory loading. Even more difficult is that the sequential differences in the rows of the matrix cause the row signal frequencies to differ. This causes severe crosstalk problems. On the other hand, where there is a deviation of the active drive system, the common drive matrix is selected as a block diagonal matrix consisting of the lower Hadamard matrix. The resulting square matrix is still orthogonal, alleviating the problems of high sequencing and computation. By selecting the Hadamard building blocks in a different arrangement, a multi-line addressing (MLA) scheme adjusts between frame response, sequencing, and calculation problems. Unfortunately, the number of lines selected at a time is limited by the row order of the Hadamar building blocks, so the frame response persists.

It is an object of the present invention to provide a new matrix drive system.

Figure 1 shows a common (low) waveform < RTI ID = 0.0 >

Fig. 2 shows a repetitive unit of a drive waveform with frame overlap

Figure 3 shows a digital filter implementation of the rho-drive system.

4 is a driving waveform of heat exchange.

According to the present invention, there is provided a driving system for a liquid crystal display of arbitrary arrangement degree, comprising a matrix building block having orthogonal and shift orthogonality (SO).

Naturally, it is possible to understand the displacement orthogonality, which is a characteristic that the matrix is orthogonal to the thermally deformed version.

Therefore, the displacement orthogonality SO is imposed on the common drive signal. As a result, the building block of the matrix becomes a square paraunitary matrix. Due to the SO characteristics of these matrices, redundancy of the building blocks is allowed. Therefore, the above matrix is q x r, q and r are integers and q > r. The r may be any integer multiple of q. For example, if q = 2, we can see the para-unitary matrix of r = 4,6,8, .... r = 6, 9, 12 for q = 3. The r value is similar to the conventional MLA arrangement. The array-r para-unitary matrix can be seen to perform similar to MLA-r in terms of voltage selection and bias ratio.

It can be seen that the para-unitary matrix is orthogonal to itself and to any number of integer versions of M up to and including its own radial version. Using such a matrix, the drive scheme outperforms the MLA of the arrangement in terms of lower hardware complexity, less crosstalk, higher contrast and better viewing cone, and higher adaptability.

A drive system for a liquid crystal display for implementing the present invention will now be described with reference to the accompanying drawings.

It is further understood that the building block is redundant due to the displacement orthogonality (SO) characteristic of the matrix. On the other hand, it is also possible to obtain a row drive matrix having a non-overlapping building block by making the frame size larger.

The driving system for implementing the present invention has the following advantages. First, the number of rows in the building block of material is q, regardless of r order. Thus, rather than O (r ² ) as in the case of MLA, the computation increases linearly with order r (ie, O (r)). In addition, since the number of rows is significantly limited, the difference in the sequence between rows can be reduced by selecting an appropriate para-unitary matrix. As a result, the frame response can be removed using a matrix having a sufficiently high order (i.e., r), irrespective of the above computation and sequencing problems. In addition, since the driving matrix is now constructed by overlapping building blocks, the crosstalk problem can be further alleviated as the number of sudden voltage changes is reduced. Finally, the driving system implementing the present invention can be implemented by selecting a para-unitary matrix of the order of an integer multiple of q. For example, if q = 2, r = 4, 6, 8, 10 ... can be obtained. Thus, we obtain greater flexibility than the MLA consisting of the Hadamard matrix of the order 2, 4, 6, 8, 16, 32 .... Due to the redundancy of the para-unitary building blocks, it is always possible to realize a seamless frame with no redundancy. For example, it is assumed that the 10-way display is driven by the para-unitary matrix drive scheme (simply, p-drive) and the MLA of the present invention. For ρ-drive, we can select the para-unitary matrix of steps 4, 6, 8, and 10. In each case, the frame size is 10. However, for the MLA, the fourth order Hadamard matrix has a frame size of 12, while the 8th order Hadamard matrix has a frame size of 16. Contrary to the case of the para-unitary matrix of the present invention, increasing the frame size reduces the ratio of selection time and further reduces the contrast of the display in the Hadamard matrix.

The matrix driving of the LCD can be represented mathematically by the following simple linear equation.

Ax = b

Here, A is an m 占 m matrix representing an m-way common drive signal. x is an m x 1 vector representing the corresponding segment drive signal. b is a column of actual data to be displayed. b is not the voltage detected by the display pixel. The actual detected RMS voltage is the accumulated version b version. A is an arbitrary orthogonal matrix (the orthogonality condition will be proved later). I is an identity matrix, and for A = I, a conventional manual drive is obtained. At A, which becomes the Hadamard matrix and its derivative function, the active addressing is obtained. For example, the next 8x8 Hadamard matrix can be used as a common (row) signal for an active-drive 8-way display. To practically drive the display, the Hadamard matrix is usually selected from among all orthogonal matrices for two voltage levels. The following shows the 8th Hadamard matrix.

As described above, due to computational load and sequencing issues using active drive, MLA is proposed. To implement the 8-way drive using the 4-line MLA, we use two quadratic Hadamard matrices as the diagonal building blocks of the 8x8 drive matrix. As a result, the common drive matrix is:

The corresponding segment drive signal,

x = A ^-1 b = A ^T b

And A is an orthogonal matrix. The state of A is silent and sufficient for the existence of a unique x. However, in order to have the actual RMS voltage which becomes the b-version of the displacement accumulation, a state in which A is orthogonal must be imposed.

A method of creating a para-unitary matrix to be used as a building block of a common (row) matrix will be described below. The para-unit state is represented by a dense matrix. To achieve this, the n × n displacement matrix S _{n, m} is expressed as:

For the i = 1, 2, ..., N, the para-unitary matrix E of the array Mx (N + 1)

(I) E is orthogonal, i.e.,

EE ^T = I

(Ii) E is orthogonal to its ridge by a multiple of M,

ES _{(N + 1) M, iM} ET = 0

.

The set of para-unitary matrices can be obtained by cascaded grid representation. For the case of M = 2, the N + 1 class para-unitary scattering matrix E (z) (z is a variable of the z-transform) has an equation of standard factorization.

Where Z ^-1 is a 2x2 matrix

ego,

Ω _k is the rotation matrix of c _k = cos (α _k ), s _k = sin (α _k )

Respectively. In order to obtain a matrix representation of E without the variable z, a diagonal matrix ^{_{Λ n ∈R 2 (N-n}} + 1) × 2 (N-n + 1) and ^{Zn∈R 2 (N-n + 1} ) × 2 (N ^{-n + 1)}

and,

.

The matrix representation of the para-unitary filter bank is obtained as follows.

For N = 1, it is 2 x 4E. There are two angles that are π / 4 radians, and there is an entrance standardized para-unitary matrix.

This matrix can be used to implement an array-4 para-unitary matrix drive scheme. For M = 2, Ω _S is a 2 × 2 circulation matrix. In the general case M, each of Ω _S is derived from the result of a given rotation matrix M (M-1) / 2. (N + 1) (M-1) / 2 rotation angles for the parity unit matrix of order M × M (N + 1).

Assume the 8-way display as described above. Suppose we have 2 × 4 E with two rotation angles, such as π / 4. The common drive signal (normalized size)

(See Figure 1 of the accompanying drawings for the waveform).

Note that instead of the square as in the case of the conventional passive, active and MLA drive systems, now the frame matrix is rectangular. In the m-way display, the frame size is m + (r-M). r = M (N + 1) is the number of rows of E similar to MLA. The overall common drive matrix is orthogonal due to the displacement orthogonal (SO) characteristic of E.

The following can be seen from the above.

(I) the number of rows of building block E is 2 regardless of degree N; This results in less computation and memory overhead compared to the conventional MLA.

(Ii) The two rows of the building block have a sequence of 1 and 2, respectively. The difference in the sequence does not increase with the increase of N. For example, for r = 8 (N = 3, M = 2), the two rows of E have sequences 3 and 4. As a result, the crosstalk caused by the difference of the row frequency components can be suppressed considerably.

(Iii) the building blocks overlap each other, and as a result, the crosstalk caused by a sudden voltage change between pixels is further reduced.

(Iv) the number of columns of the driving matrix, where the frame size is r-M greater than the number of passive drives m + (r-M). For example, for a 2x4 building block, r-M = 4-2 = 2. For the old-medium display, the reduction of the selection time rate is ignored. Nonetheless, it is possible to explain how to make frames overlap, while maintaining the average frame size m, as will be explained below.

(V) Due to the redundant nature of the para-unitary building blocks, a seamless frame without redundancy can always be realized. For example, assume that a 10-way display is driven by a para-unitary matrix drive system and an MLA. For ρ-drive, a para-unitary matrix of degree 4, 6, 8, or 10 can be selected. In any case, the frame size is 12 (or by adopting a frame duplication method 10). However, for the MLA, the fourth Hammadar matrix has a frame size of 12, while the 8th Hammadar has a frame size of 16. Increasing the frame size reduces the select time rate and further reduces the contrast of the display. The following shows the increase in frame size to 16 by adopting the 8th Hadamard matrix of the MLA.

The matrix drive of the LCD is represented by the linear equation Ax = b. For manual, active and MLA, A is a square m by m matrix. x and b are m x 1 vectors. In a para-unitary matrix drive, the linear equation is still valid, except that A is now an m × (m + r-M) matrix. x and b are (m + r-M) x 1 vectors. M and r are the sizes of rows and columns of E, as described above. Let n be the number of simultaneously selected lines, and in a para-unitary matrix drive system, r = n. In the conventional driving, n = 1 and n = m are represented by a passive and active driving system respectively, while 1 <n <m represents an n-line MLA. Let the voltage levels of the common drive be -s O and s. (N-2) d / n, - (n-4) d / n, ..., 0, ... d That is,

As shown in Fig.

The linear matrix equation can be expressed by the following equation.

Here, A 'is orthogonal. Now, let ∥ _{2 be} the 2-norm of a matrix defined as the largest single value.

And this value is a constant. In fact, this explains why A must be orthogonal. The RMS voltage detected across the pixel j is &lt; RTI ID = 0.0 &gt;

.

In the conventional driving, p = m, and in the para-unitary matrix driving, p = m + r-m.

As a result,

.

If k = s / d is a bias ratio,

.

If r is differentiated with respect to k,

and

.

Regardless of whether the display is a manual, active, MLA or para-unitary system, the selectivity depends only on the number of ways m.

The number of columns of the driving matrix A becomes the frame size of m + (r-M), which is r-M larger than the size of the passive driving. For example, in a 2x4 building block E, r-M = 4-2-2. In a conventional display, the reduction of the select time ratio is ignored. In the following, the frames overlap each other so that the drive signal overtime is still orthogonal. This allows an easy implementation of the drive system by accessing the digital filter bank. Also, the average frame size is reduced to m instead of m + (r-M), which increases the selection time ratio. Note that the calculation of the optimum bias and selectivity is the same. The following shows two frames of the modified drive signal for a six-way display (see Figure 2 of the accompanying drawings for waveforms)

The modified drive scheme can be easily implemented using a discrete filter bank approach. In the case of the 2x4 building block E, the two rows [1 1 -1 1] and [-1 -1 -1 1] can be considered as low-and high-pass quadratic digital filters, respectively. A digital filter implementation is shown in FIG. However, for a fixed number of rows in the display, the frame size is currently twice as large, thereby reducing the select time ratio. For example, in a three-way display, a 1 × 4 building block results in an average frame size of 2 × 3 = 6.

In this case, one digital filter is required instead of the two digital filters. The columns of the drive matrix can be rearranged to distribute the selection more evenly. This arrangement can further improve the contrast of the display by further suppressing the frame response. The following shows the case of 8-way drive. Without rearrangement, the next iteration unit of the drive signal is obtained.

After the rearrangement,

.

Note that there is another possibility that the orthogonal property of the driving signal is maintained by the exchange of rows and columns.

Four methods for the implementation of the concentration are described below.

First, the three methods are the methods for completing the concentration in passive and MLA drives. These three methods can be employed as complementary concentrations in a unitary drive system. The fourth method is based on the use of a multi-order orthogonal / para-unitary building block for the common signal.

In the first method, the frame ratio control is employed for the implementation of manual drive and MLA concentration. This method is relatively simple to apply to the new drive system. (n-1) m, where m is the original frame size for monochrome display, the implementation n gray level is typically the number of display ways. Briefly, the common signal is only made by the connection of the n-1 original frame for the B / W display. This shade is determined by the number of ONs between these n-1 sub-frames in the range from 0 to n-1.

The second method is a voltage replenishment method and can be adopted in the P-drive system. As described above, the RMS voltage was applied to a pixel which can be expressed as follows.

here,

to be.

For b _j to be ± b, as in the case of a non-gray shade drive, the second term on the right-hand side, such as the square of 2-gambling of b, is a constant independent of signal b. This is not the case if the shade is displayed on the grid. In this case, b? B _j? B, the term depends on signal b and is not a constant. The term is optimal when all entries satisfying b are ± b. To make the RMS voltage dependent only on the third term, an extra time slot is added to the frame. In the extra time slot, blank signals are commonly applied. For the segment, the supplemental voltage v,

.

The voltage?

Lt; / RTI &gt;

finally,

.

The supplemental voltage v is calculated for each row.

The voltage replenishing method described above is based on amplitude modulation. However, the method depends on the calculation of the intensive row complement voltage for each frame. To overcome this problem, the method that has been derived for the MLA is based on a one time extended frame.

This method can be applied to the proposed P-drive system. In order to display the signal b≤d _j ≤b, sub-signal b for the frame is

Lt; / RTI &gt;

On the other hand, for the second sub-frame,

The third term

. &Lt; / RTI &gt;

This is irrelevant to signal b as desired. It is also possible to increase the number of gray levels by mixing the frame ratio control and the amplitude modulation. On the other hand, by adopting the irregular variance of d _j and a plurality of sub-frames, the number of grayscale levels can be increased. By adopting 1, 2, and 3 sub-frames and 4 levels in a frame, 4, 9, and 25 grayscale levels are realized. This method can be applied to the proposed P-drive system.

In a fourth method, a concentration method based on a multi-order, orthogonal / para unitary building block is provided. This method creates a trade-off between the frame size and the number of voltage levels and balances circuit complexity and LCD band inverse demand. If the 4-way display is driven at 8 concentrations, the next orthogonal common drive matrix is introduced.

In a higher multiplexed display drive, the matrix may be used as a diagonal building block in a higher order common drive matrix. The matrix is a cascade of orthogonal square matrices of orders 2 ² , 2 ^1, and 2 ⁰ . If the otherwise orthogonal building blocks are n, a 2 ⁿ concentration is possible and the common drive matrix will be 2 ^n-1 x n2 ^n-1 . Multiplexing the display,

x = A ^T b

, The number x of voltage levels is 2 ^{&lt; n &gt; -1} +3. The RMS voltage is

, Where d is the current smallest segment voltage.

k _opt =

and

, Usually achieving the same performance as manual / active drive. The selectivity per frame of the proposed scheme is usually (2 ⁿ -1) / n larger than the manual drive. Para units can also be used as building blocks. For this embodiment, then,

.

As noted above, the scheme can be modulated to reduce the number of segment voltage levels at the expense of a larger frame. The 2-way LCD is driven at 8 levels. The common drive signal matrix

. A '

.

By calculating x by x = A ' ^T b, the number of voltage levels can be limited to 5 regardless of n. However, the frame size is now increased to m2 ^n-1 . To achieve this level of 8, the frame is magnified 4 times. The per-frame selection is now (2 ⁿ -1) / 2 ^n-1 higher than normal manual drive. The RMS voltage,

, Where d is the minimum segment voltage, as described above.

and

.

This is like a normal passive / active drive system. In a higher multiplexed display drive, a matrix A 'can be used as a diagonal building block for a higher order common drive matrix (i.e., diagonally (A', A ', ..., A')). The number of rows of the product driving matrix is n times the multiplexing m. As described above, 2 ⁿ is the number of gray levels. The data b must be represented in binary format so that the first frame is responsive to the most significant bit, while the last frame is responsive to the least significant bit.

The drive system of the present invention provides the following.

(1) A new driving system for liquid crystal displays is a rectangular building block, which uses a para-unitary matrix of order Mx (N + 1) M. The new drive scheme outperforms the same order of MLA by lower hardware complexity, lower crosstalk, higher contrast and higher viewing cone, and higher flexibility of implementation.

(2) The para-unitary matrices have a displacement orthogonality (SO) property that enables overlapping of the building blocks. As a result, r = (N + 1) M can implement a general-order r-driving system that selects an appropriate N and M to be any positive integer.

(3) The new drive scheme can be implemented by a valid digital filter bank approach (FIG. 3).

(4) rows can be selected in the Mx (N + 1) para unitary matrix to realize a driving signal having an increased frame size. This reduces hardware complexity. For example, by selecting one row out of the 2 rows of the 2x2 (N + 1) pixel unit matrix, the driving system can be realized by one digital filter. However, the frame size is doubled, which is the reduced selection time ratio.

(5) The rows and columns of the driving matrix can also be rearranged so that the selection is more uniformly distributed. This arrangement further suppresses the frame response, resulting in a higher contrast of the display. Various configurations are possible as the orthogonal characteristics of the driving signal are maintained by exchange of rows and columns.

(6) Hardware complexity can be reduced by using the number of rows in the para-unitary matrix.

(7) By using the more evenly distributed rows and columns of the driving matrix, this arrangement can further suppress the frame response, resulting in a higher contrast of the display.

(8) A gray seal method is used to provide a tradeoff between circuit complexity and LCD band inverse demand while providing a trade-off between frame size and number of voltage levels.

(9) With an increased frame size, the para-unitary matrix may have the property of displacement orthogonality, and the pair of fairing building blocks do not overlap or overlap each other without affecting the orthogonality of the driving matrix.

(10) Thus, a driver for a liquid crystal display of arbitrary order and magnitude comprising a matrix building block with orthogonal and displacement orthogonality (SO) characteristics is achieved. The orthogonal property of the matrix to the heat-displacement version itself is called SO.

Claims (11)

Wherein the matrix building block has orthogonal and displacement orthogonal (SO) characteristics. 2. The driving system for a liquid crystal display of any one of claims 1 to 3, wherein a parallel unitary matrix of a general Mx (N + 1) M degree is used as the orthogonal building block. 3. The driving system as claimed in claim 1 or 2, characterized in that a row-driving matrix is formed using a parabolic matrix as a building block. 4. A drive system according to any one of claims 1 to 3, for any arrangement and dimension of a liquid crystal display featuring a digital filter bank. 6. A method according to any one of claims 2 to 5, characterized in that the number of arbitrary rows is selected from an Mx (N + 1) M para unitary matrix to realize a drive signal whose frame size is increased Driving system for liquid crystal display of array and dimension. 6. The method of claim 5, wherein one of the 2 rows of the 2x2 (N + 1) para unitary matrix is selected to provide a driving scheme by one digital filter. Drive system for display. 7. The drive system according to any one of claims 1 to 6, wherein the rows and columns of the drive matrix are rearranged so that the selection is more evenly distributed. 8. The driving system for a liquid crystal display according to any one of claims 1 to 7, characterized by a gray-scale addressing method based on a multi-order orthogonal / para unitary building block. 9. A method according to any one of claims 2 to 8, characterized in that, without affecting the orthogonality of the row-driving matrix, the par- &lt; RTI ID = 0.0 &gt; unitary matrix building block &lt; / RTI &gt; Dimensional driving system for liquid crystal displays. 9. A drive system according to any one of claims 2 to 8, wherein the para-unitary matrix building block is characterized by a para-unitary matrix with non-overlapping displacement orthogonality. A liquid crystal display characterized by the drive system according to any one of the claims 1 to 10.