JP2016143037A - Spatial phase modulator - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a spatial phase modulator capable of suppressing phase cross talk relative to a liquid crystal element while maintaining the phase of a deflection pattern and its correction method.SOLUTION: An optical switch 1 includes: a computing unit 21 for generating a deflection pattern to be applied to each pixel of a liquid crystal element 10; a computing unit 23 for calculating a correction value in proportion to a voltage difference among each pixel using the voltage difference and a correction coefficient selected from a plurality of correction coefficients; and a subtractor 25 for correcting the deflection pattern based on the correction value. The computing unit 23 selects a correction coefficient to be a desired deflection pattern from a memory 24.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a liquid crystal-based spatial phase modulator used as a deflection element in a spatial optical switch used in a communication optical transmission device, a wavelength routing device, or the like.

  In recent optical communication, high-speed communication is realized by transmitting an optical signal as it is to a communication destination without converting the optical signal into an electrical signal. In addition, WDM (Wavelength Division Multiplexing) technology that multiplexes one optical signal in correspondence with one wavelength enables large-capacity optical transmission using a single optical fiber. With the development of such optical communication technology, the role of an optical switch that switches a communication path while maintaining an optical signal is becoming more important.

  As the size of optical communication networks increases, the number of wavelengths of optical signals also increases, and the wavelength selection switch that selects any wavelength from among several tens of wavelengths and outputs it from one of multiple output fibers is downsized and high. Functionalization is progressing. As such a highly functional wavelength selective switch, a spatial optical system optical switch using a liquid crystal element as a spatial phase modulator has been attracting attention (Non-Patent Document 1).

  A basic configuration of a conventional wavelength selective optical switch will be described with reference to FIG. FIG. 11 is a diagram showing a basic configuration of a wavelength selective optical switch using a reflective liquid crystal-based spatial phase modulator as a deflection element.

  The wavelength selective optical switch shown in FIG. 11 includes input / output ports 101 and 102, a dispersion spatial optical system 103 that divides a wavelength multiplexed signal into spaces, and a spatial phase that deflects a light beam in a form orthogonal to the dispersion direction of 103. And a modulator 104.

  Next, the principle of deflection realized by the wavelength selective optical switch will be described with reference to FIG. 12A and 12B are schematic diagrams showing a change in deflection angle caused by a change in phase state and a mode of occurrence of optical crosstalk, where FIG. 12A is a deflection principle, FIG. 12B is a phase shift due to disclination, and FIG. ) Indicates scattering due to phase shift.

  As shown in FIG. 12A, on the screen of the liquid crystal element which is the spatial phase modulator 104, a periodic deflection pattern and a continuous deflection pattern having a phase difference of 2π at the periodic end are displayed. With this deflection pattern, the light irradiated on the screen is reflected in a manner similar to mirror reflection. This is because an appropriate voltage is applied to each pixel of the liquid crystal element to give the incident light a spatial phase delay similar to that of mirror reflection.

  In FIG. 12A, a phase difference of 2π is provided in the deflection pattern using the characteristics of the wave that light has. In this case, although the deflection pattern itself is intermittent, the phase change seems to be continuous when viewed from the incident light, and it is deflected as if it hits an angled plane mirror. This is the same as when a Fresnel lens having a periodic structure functions as a lens.

  Usually, the periodic deflection pattern is set so that the phase changes linearly with respect to the position on the screen. In general, the maximum phase modulation amount of the spatial phase modulator 104 is finite. However, by changing the phase by 2π before and after a certain position before reaching the maximum phase modulation amount, the wavefront of light has no phase difference even though the phase at that position has returned by 2π. Since it returns to a state, it will be in the state which is changing linearly everywhere. The deflection pattern at this time is a sawtooth pattern (FIG. 12A).

  When a sawtooth-shaped deflection pattern is applied to the liquid crystal-based spatial phase modulator 104, a sharp phase change occurs at a position where the phase returns from 2π to 0, that is, at the boundary surface of one period of the sawtooth. A combination (phase crosstalk phenomenon between elements in which the rotation direction of liquid crystal molecules is different from the surroundings at a deflection pattern boundary or the like) may occur. When disclination occurs, there is a slight phase shift from a preset phase (FIG. 12B). For example, the position of 2π is less than 2π, and the position of 0 is a phase larger than 0. As a result, although the phase difference of 2π is set so that the wavefronts of the light are aligned, the wavefront is distorted by this phase shift. In the optical switch shown in FIG. 11 described above, when the wavefront is distorted by the spatial phase modulator 104, the light beam is scattered, which also causes optical crosstalk to the unconnected port, and the switch characteristics deteriorate. Can do.

  In FIG. 12B, a phase shift due to disclination occurs at a discontinuous position where the phase changes from 2π → 0 due to the deflection pattern. As a result, although the wavefront of the light is set to be continuous with a phase difference of 2π, the phase is discontinuous with respect to the wavefront of the light at the intermittent position of the deflection pattern due to the phase shift. It seems to change. This is the same situation as when the mirror has irregularities, and the light at this position is scattered in a direction different from the continuous deflection direction (FIG. 12C). As a result, the scattered light is coupled to a port that is not a connection destination, and optical crosstalk can occur.

  A liquid crystal element used as the spatial phase modulator 104 has the same structure as a liquid crystal element used in an image apparatus such as a liquid crystal projector. For liquid crystal elements used as image devices, measures for disclination have been proposed from the viewpoint of improving image quality. In the case of an image device, for example, a black line appears on a white screen and a white line appears on a black screen between adjacent pixels where the difference in drive voltage is likely to appear. It is regarded as a problem that causes a decline in

  For disclination relating to an image device, Patent Document 1 discloses a method of correcting a level corresponding to the level of a pixel adjacent to the target pixel with respect to the target pixel. Patent Document 2 discloses a method of adjusting an upper limit value and a lower limit value with respect to a gradation range of an image.

JP 2000-330084 A JP 2012-237828 A

G. Baxter et al., “Highly programmable Wavelength Selective Switch based on Liquid Crystal on Silicon switching elements,” OFC 2006, OTuF2, 2006.

  The countermeasure against the above-mentioned disclination is mainly to correct the applied voltage value so as to reduce the applied voltage difference between pixels in which the lateral electric field becomes strong. However, when the liquid crystal element is used as a spatial phase modulator, it is important how to suppress the phase crosstalk from the viewpoint of image quality.

  As the phase becomes closer to the ideal phase according to the deflection pattern, unnecessary scattered light is suppressed, so that optical crosstalk can be suppressed. In that respect, increasing the electric field in the lateral direction and increasing the disclination is not a big problem.

  In the case of a spatial phase modulator, it is not only necessary to reduce disclination, but it is important to suppress optical crosstalk while maintaining the phase of the deflection pattern. However, there is no technique for suppressing phase crosstalk from such a viewpoint.

  An object of the present invention is to provide a spatial phase modulator capable of suppressing phase crosstalk with respect to a liquid crystal element while maintaining the phase of a deflection pattern under the above-described circumstances, and a correction method thereof.

  The present invention for solving the above problem is a spatial phase modulator having a liquid crystal element, a voltage pattern generation unit for generating a deflection pattern to be applied to each pixel of the liquid crystal element, and a voltage between the pixels. A correction value calculating unit that calculates a correction value proportional to the voltage difference using a difference and a correction coefficient selected from a plurality of correction coefficients, and a correction that corrects the deflection pattern based on the correction value The correction value calculation unit selects the correction coefficient that provides a desired deflection pattern.

  The correction value calculation unit may select the correction coefficient corresponding to the orientation direction of the liquid crystal, the sign of the voltage difference, the magnitude of the voltage difference, or the light attenuation rate.

  The correction value calculation unit may select the correction coefficient associated in advance according to the applied voltage of the pixel.

  The correction value calculation unit may select the correction coefficient that continuously changes according to the applied voltage of the pixel.

  Further, the present invention for solving the above-described problem is a method for correcting a driving voltage of a spatial phase modulator having a liquid crystal element, the step of generating a deflection pattern to be applied to each pixel of the liquid crystal element, Using the voltage difference between each pixel and a correction coefficient selected from a plurality of correction coefficients, calculating a correction value proportional to the voltage difference; and correcting the deflection pattern based on the correction value The step of calculating the correction value includes selecting a correction coefficient that provides a desired deflection pattern.

  According to the present invention, it is possible to suppress phase crosstalk with respect to the liquid crystal element while maintaining the phase of the deflection pattern.

It is a figure which shows an example of schematic structure of the spatial phase modulator in embodiment of this invention. It is a figure for demonstrating the appearance of the disclination at the time of the application of a rectangular tooth voltage. It is a figure for demonstrating the process for correct | amending the voltage value of an attention pixel according to the voltage difference between each pixel. It is a figure for demonstrating each correction coefficient selected according to the orientation direction and voltage difference in each pixel. It is a figure which shows a phase waveform when a correction process is performed. It is a figure which shows the example of a setting of the attention pixel and adjacent pixel at the time of applying a linear pattern and a two-dimensional pattern. It is a flowchart of the correction process of the voltage value in the spatial phase modulator of embodiment. FIG. 8 is a flowchart of remaining correction processing executed following the correction processing of FIG. 7. It is a flowchart of the correction process which made the correction coefficient change continuously according to the magnitude | size of a drive voltage difference. FIG. 10 is a flowchart of remaining correction processing executed following the correction processing of FIG. 9. It is a figure which shows schematic structure of the conventional spatial phase modulator. It is a schematic diagram which shows the generation | occurrence | production aspect of the change of the deflection angle which generate | occur | produces by the change of a phase state, or optical crosstalk.

<First Embodiment>
Hereinafter, an optical switch 1 that is a spatial phase modulator according to a first embodiment of the present invention will be described.

  FIG. 1 is a diagram illustrating an example of a schematic configuration of an optical switch 1 according to the first embodiment. The optical switch 1 is mainly intended to suppress optical crosstalk, and has a function of correcting a deflection pattern.

  In FIG. 1, the optical switch 1 includes a liquid crystal element 10 and a controller 20 that provides screen information to the liquid crystal element 10 via a D / A converter 30. In FIG. 1, only the configuration for controlling the liquid crystal is shown, but the optical switch 1 is similar to FIG. 11 in that the input / output ports 101 and 102, the dispersion space optical system 103, and the spatial phase of the liquid crystal element. And a modulator 104. In the optical switch 1 of this embodiment, a liquid crystal element 10 having disclination is provided as a spatial phase modulator 104.

  The controller 20 includes a calculator (voltage pattern generation unit) 21, a memory 22, a calculator (correction value calculation unit) 23, a memory 24, and a subtracter (correction unit) 25. Further, the connection destination port of the optical switch 1 and the attenuation rate of the connection optical power are input to the controller 20 from the outside.

  The calculator 21 generates an arbitrary deflection pattern in the liquid crystal. The memory 22 records the deflection pattern itself and data that is the basis of the deflection pattern. The computing unit 21 refers to the deflection pattern and data in the memory 22 and processes the deflection pattern as necessary.

  When the request for the connection destination port and the attenuation factor is input from the outside, the computing unit 21 generates a deflection pattern corresponding to the request.

  The calculator 23 obtains a correction value of the drive voltage for the target pixel of the deflection pattern generated by the calculator 21.

  A correction coefficient is recorded in the memory 24 in advance, and the computing unit 23 reads a correction coefficient described later from the memory 24, and calculates a voltage difference between the target pixel and an adjacent pixel (hereinafter simply referred to as “voltage difference between pixels”). And the correction coefficient, the correction value of the voltage included in the deflection pattern is calculated for the target pixel. This process will be described in detail later.

  The subtractor 25 corrects the deflection pattern by subtracting the correction value obtained by the calculator 23 from the deflection pattern generated by the calculator 21. Depending on how the sign of the calculated correction value is taken, it may be addition instead of subtraction, but this is due to the difference in expression, and here, in either case, explanation will be made with a subtractor. The corrected deflection pattern is applied to the electrode of the liquid crystal element 10 via the D / A converter 30.

  The subtracting process of the subtracter 25 is shown in equation (1) described later.

  In the example of FIG. 1, two arithmetic units 21 and 23 are shown. However, in an actual circuit configuration, one arithmetic unit (for example, CPU or DSP) serves as the functions of the arithmetic units 21 and 23. It is envisaged to configure. Since the D / A converter 30 is often provided on the substrate side of the liquid crystal element 10, it may not be on the controller 20 side.

[Deflection pattern correction]
Next, deflection pattern correction processing realized by the optical switch 1 will be described.

In this optical switch 1, the controller 20 corrects the voltage value of the deflection pattern for the target pixel using the voltage difference between the pixels and the correction coefficient. Here, assuming that the voltage of the pixel of interest is V ₀ and the voltage of the adjacent pixel is Va, the corrected voltage value V _{1 of the} pixel of interest is expressed by the following equation (1).

_{V 1 = {V 0 - k} × (Va -V 0)} (1)
In equation (1), k represents a correction coefficient. k is a real number from −2 to +2.

In the above formula (1), for example, when V ₀ > Va and k is a positive value, V ₁ > V ₀ . That is, the corrected voltage value V ₁ is larger than the uncorrected voltage value V ₀ . Although this is not preferable from the viewpoint of suppressing disclination because the electric field in the lateral direction becomes strong, if the voltage value has a phase of the deflection pattern that makes it difficult to cause disturbance of the wavefront, This is preferable from the viewpoint of suppressing crosstalk.

  In the optical switch 1, when the correction coefficient k is optimized so as to obtain a desired deflection pattern, this k is not always a positive value. As will be described later, when obtaining the optimum correction coefficient k in the case of a complicated deflection pattern, the calculator 23 first temporarily sets the value of the correction coefficient k from all values including negative values, and then sets the correction value. Is calculated. Then, the computing unit 23 determines a correction coefficient k that becomes a desired deflection pattern from the given correction coefficient k. As to whether or not it is a desired deflection pattern, for example, whether or not the phase crosstalk is minimized so that the deflection pattern having the corrected voltage value approaches a predetermined phase (one having no distortion from the design condition) (FIG. 12). (See (b)), or the degree of optical crosstalk due to scattering (measured value) on the screen based on the deflection pattern having the corrected voltage value (see FIG. 12C). .

  As will be described later, the phase crosstalk changes according to the alignment direction of the liquid crystal, the positive / negative of the voltage difference between the pixels, and / or the magnitude of the voltage difference between the pixels. It is preferable to set in consideration of these factors.

[Orientation direction]
The liquid crystal element 10 has an alignment direction. This is a direction generated by regularly arranging liquid crystal molecules. The appearance of disclination varies depending on the orientation direction.

  FIG. 2 is a diagram for explaining the appearance of disclination when a rectangular voltage pattern P of a rectangular wave is applied on the alignment direction axis.

  The voltage value P of the rectangular voltage pattern P changes from 0 to Vm at the rise and fall of the rectangle, and the phase φ1 takes a value as shown in FIG. 2 following the voltage. The phase φ1 does not follow the rectangular voltage pattern P because of disclination, but the phase φ1 does not become symmetrical with respect to the rectangular pattern P that is further symmetrical. That is, the phase φ1 shows a different shape from the center of the rectangle on the left or right side of the drawing. In the following description, the left side of the drawing is referred to as “minus side”, and the right side is referred to as “plus side”.

  In FIG. 2, the change (gradient) from 0 to Vm of the rectangular voltage pattern P is steeper in the negative phase φ1 than in the positive phase φ1. That is, the change in the positive phase φ1 is more gradual than the change in the negative phase φ1.

  From this point of view, the correction coefficient k is preferably changed according to the orientation direction. In the optical switch 1 of this embodiment, the calculator 23 corrects the voltage value of the pixel of interest using a different correction coefficient k depending on the orientation direction. The computing unit 23 changes the value of the correction coefficient k depending on whether the pixel is located on the minus side or the plus side of the target pixel. This process will be described in detail later.

[Signal difference between pixels]
The effect of disclination differs depending on whether the voltage difference is positive or negative. When the voltage of the target pixel is high and the voltage of the adjacent pixel is low, and vice versa, the phase is different even if the magnitude of the voltage difference between the target pixel and the adjacent pixel is the same. For this reason, the calculator 23 changes the value of the correction coefficient k in accordance with the sign of the voltage difference between the pixels. This process will also be described in detail later.

[Voltage difference between pixels]
The influence of disclination also varies depending on the magnitude of the voltage difference between pixels. For example, when the voltage difference between the pixels is large, it is conceivable that the phase change becomes large, for example, the phase of the deflection pattern changes from 0 to 2π. On the other hand, when the voltage difference between the pixels is small, the phase of the deflection pattern does not change so much. As such an example, it is conceivable that the phase change is large at the sawtooth folded portion and the phase change is small at the inclined portion.

  Disclination is more susceptible to a greater electric field in the lateral direction when the voltage difference between the pixels is larger than when it is smaller.

From this viewpoint, in the optical switch 1 of this embodiment, when the voltage difference between the pixels is large, the arithmetic unit 23 corrects the correction value represented by the above formula (1), that is, {k × (Va −V ₀ ). }, A correction value is calculated using a correction coefficient k that increases the value of. On the other hand, when the voltage difference between the pixels is small, the correction value is calculated by using the correction coefficient k such that the correction value shown in Expression (1) is small. This process will be described with reference to FIG.

  The phase crosstalk changes depending on the alignment direction of the liquid crystal, the sign of the voltage difference between the pixels, and / or the magnitude of the voltage difference between the pixels. Depending on the combination of these, the disclination shown in FIG. Unlike the way you see, overshoot and undershoot may occur. In this case, in order to correct a phase larger than desired as an overshoot, the correction value k may be an optimum negative value.

  FIG. 3 is a diagram for explaining a process for correcting the voltage of the target pixel by selecting a correction coefficient corresponding to the magnitude of the voltage difference between the pixels.

In FIG. 3, the voltage of the target pixel at the target point n is “V ₀ ”, the negative side adjacent pixel (indicated by “−1” in FIG. 3) is “V ₋ ”, and the positive side adjacent pixel ( The voltage of “ ₊₁ ” in FIG. 3 is set to “V ₊ ”.

The optical switch 1 is configured to calculate a correction value of the voltage V ₀ for a target pixel using a voltage difference between pixels and a correction coefficient.

In this embodiment, the calculator 23 calculates a voltage difference between pixels. In the example of FIG. 3, the voltage difference ΔV− between the target pixel and the negative adjacent pixel is (V ₋ −V ₀ ), and the voltage difference ΔV + between the target pixel and the positive adjacent pixel is (V ₊ −V ₀ ). It is.

  Next, the computing unit 23 selects a correction coefficient k corresponding to the magnitude of the voltage difference ΔV−, ΔV + described above from the memory 24. The memory 24 stores a plurality of correction coefficients k (for example, k in the case of correcting the influence of the minus side adjacent pixel is km1, km2, km3, km4, and k in the case of correcting the influence of the plus side neighboring pixel is kp1, kp2, kp3, kp4) and a threshold value ΔVth are stored.

  The calculator 23 reads from the memory 24 the correction coefficient k corresponding to the relationship between the voltage differences ΔV− and ΔV + and the threshold value ΔVth. Which correction coefficient k (for example, km1 to 4, kp1 to 4) is read out is shown in detail in FIG.

The computing unit 23 calculates the correction value shown in the above equation (1), that is, {k × (Va−V ₀ )} for each of the minus side and the plus side. Here, assuming that the correction value for correcting the influence of the negative adjacent pixel is Vc ₋ , and the correction value for correcting the influence of the positive adjacent pixel is Vc +, they are expressed by the following equations (2) and (2): It is represented by (3).
Vc ₋ = k × ΔV− (2)
Vc + = k × ΔV + (3)

  In the above formula (2), k is set to any one of km1 to 4 corresponding to the difference value between ΔV− and ΔVth. In the above equation (3), any value of kp1 to kp4 corresponding to the difference value between ΔV + and ΔVth is set for k.

The subtracter 25 corrects the voltage V ₀ of the target pixel (for example, the target point n in FIG. 3) in the deflection pattern, as shown in the above formula (1). The corrected pixel voltage V1 is expressed by the following equation (4).
_{V1 = V 0 - (Vc -} + Vc +) (4)

  Thus, in the optical switch 1 of this embodiment, the correction coefficient corresponding to the magnitude of the voltage difference between the pixels is selected to correct the voltage of the pixel of interest.

  Here, an example of correcting the influence of the adjacent pixels on the plus side and the minus side is shown, but the same applies to the case of correcting the influence of the two-dimensional adjacent pixels. In this case, the influence of each adjacent pixel is corrected. What is necessary is just to correct | amend by adding a correction value.

  FIG. 4 is a diagram for explaining the correction coefficients km1 to km4 and kp1 to kp4 selected according to the relationship between the voltage difference between the pixels and the threshold value. In FIG. 4, the voltage differences ΔV− and ΔV + are expressed as “ΔV”.

  As shown in FIG. 4, in the optical switch 1, when correcting the voltage, correction coefficients (km1 to km4, kp1 to kp4) corresponding to the relationship between the voltage difference between the pixels and the threshold are selected.

  For example, “km1” and “kp1” are set as correction coefficients k when ΔV ≧ ΔVth, and “km2” and “kp2” are set as correction coefficients k when ΔVth ≧ ΔV ≧ 0, respectively. Further, for example, when 0> ΔV ≧ −ΔVth, the correction coefficients k are set to “km3” and “kp3”, and when ΔV <−ΔVth, the correction coefficients k are set to “km4” and “kp4”, respectively. .

  In this way, a deflection pattern in which the voltage of each pixel is corrected is generated for all pixels. The deflection patterns and phase waveforms before and after correction are illustrated in FIG.

  FIG. 5 shows deflection patterns before and after correction. (A) is a deflection pattern before correction, (b) is a phase waveform when the deflection pattern before correction is applied as it is, and (c) is after correction. Deflection pattern (d) illustrates the state of the phase waveform when the corrected deflection pattern is applied.

  In FIG. 5A, a triangular sawtooth wave is set as a deflection pattern before correction. When this deflection pattern is applied to the liquid crystal element 10 as it is, the controller 20 has a waveform in which the triangular corner is dull due to disclination as shown in FIG. In the portion where the waveform is blunt, light does not turn to the direction of the target port, which may cause optical crosstalk. Therefore, it is preferable to remove the blunt as much as possible.

  Therefore, in the optical switch 1 of the present embodiment, the controller 20 executes the above-described deflection pattern correction processing to correct the deflection pattern as shown in FIG. That is, in the example of FIG. 5C, a correction value is added to a triangular corner in the deflection pattern, and the deflection pattern is corrected so that the corner protrudes. Then, the controller 20 applies the corrected deflection pattern to the liquid crystal element 10. When such a deflection pattern is applied, according to FIG. 5D, a phase waveform (see FIG. 5A) close to the original sawtooth waveform is obtained.

  The deflection pattern shown in FIG. 5A may be a linear pattern in which the voltage changes in a linear direction with respect to the screen, or a two-dimensional pattern in which the voltage changes in an arbitrary direction on the screen. Good.

  FIG. 6 is a diagram showing an example of such a deflection pattern, where (a) shows a linear pattern and (b) shows a two-dimensional pattern.

  In FIG. 6A, since there are two adjacent pixels (n−1, n + 1) as adjacent pixels of the target pixel n, the voltage value of the target pixel n is set according to the voltage state of each adjacent pixel. It is corrected.

  In FIG. 6B, since there are four adjacent pixels (n−h1, n + h1, n−v1, n + v1) as adjacent pixels of the target pixel n, the attention is made according to the voltage state of each adjacent pixel. The voltage value of the pixel n is corrected.

[Operation of optical switch]
The operation of the optical switch 1 that realizes the above-described correction of the deflection pattern will be described below with reference to FIGS. 7 to 8 are flowcharts of voltage value correction processing in the optical switch 1.

  In the optical switch 1, the calculator 21 sets a sawtooth waveform as shown in FIG. 7 as a deflection pattern before correction. During the correction process, this deflection pattern is input to the computing unit 23, and the computing unit 23 performs the processes of steps S1 to S3 described later.

  In step S1, the calculator 23 acquires the electrode voltage on the screen position from the input deflection pattern. Note that, according to the flowchart of FIG. 7, the electrode voltage is represented by V [N] (where N represents the number of a pixel representing a position on the screen).

  In step S2, the loop processing of the following steps S21 to S24 is repeated from N = 1 to the maximum pixel number of the deflection pattern.

In step S21, voltage information is acquired. Specifically, the arithmetic unit 23 acquires the electrode voltage of the pixel of interest and its adjacent pixels. In the example of FIG. 7, there are two adjacent pixels, the negative side and the positive side (see FIG. 3), so the calculator 23 calculates the electrode voltage V _{0 of the} target pixel and the electrode voltage V of each adjacent pixel. _- , V ₊ will be acquired.

In step S22, a correction value for the minus side pixel is calculated. At this time, the computing unit 23 selects a correction coefficient associated in advance according to the applied voltage of the pixel, but the specific processing of step S22 is shown in steps S221 to S228 described later. . Referring to FIG. 4, first, the computing unit 23 determines whether or not V ₋ −V ₀ (= ΔV−) ≧ ΔVth (step S221), and in the case of V ₋ −V ₀ ≧ ΔVth (step S221). YES), the correction coefficient k = km1 is read from the memory 24 (step S222), and the correction value Vc ₋ is calculated (step S228). According to the above equation (2), the correction value Vc ₋ is calculated from k · (V ₋ −V ₀ ) (= k × ΔV−).

When V ₋ −V ₀ (= ΔV−) ≧ ΔVth is not satisfied (NO in step S221), the arithmetic unit 23 next determines whether V ₋ −V ₀ (= ΔV−) ≧ 0 (step S223). , V ₋ −V ₀ ≧ 0 (YES in step S223), the correction coefficient k = km2 is read from the memory 24 (step S224), and the correction value Vc ₋ is calculated (step S228).

When V ₋ −V ₀ (= ΔV−) ≧ 0 is not satisfied (NO in step S223), the computing unit 23 next determines whether or not V ₋ −V ₀ (= ΔV −) <− ΔVth (step S225). ), If V ₋ −V ₀ <−ΔVth (YES in step S225), the correction coefficient k = km4 is read from the memory 24 (step S224), and the correction value Vc ₋ is calculated (step S228).

When V ₋ −V ₀ (= ΔV −) <− ΔVth is not satisfied (NO in step S225), the arithmetic unit 23 next reads the correction coefficient k = km3 from the memory 24 (step S227), and calculates the correction value Vc ₋ . Calculate (step S228).

Referring to the flowchart of FIG. 8, in step S23, the correction value by the plus side pixel is calculated as in step S22 described above. Specific processing in step S23 is shown in steps S231 to S238 described later. Referring to FIG. 4, first, the computing unit 23 determines whether or not V ₊ −V ₀ (= ΔV +) ≧ ΔVth (step S231). If V ₊ −V ₀ ≧ ΔVth (YES in step S221), The correction coefficient k = kp1 is read from the memory 24 (step S232), and the correction value Vc ₊ is calculated (step S238). According to the above equation (3), the correction value Vc ₊ is calculated from k · (V ₊ −V ₀ ) (= k × ΔV +).

When V ₊ −V ₀ (= ΔV +) ≧ ΔVth is not satisfied (NO in step S231), the computing unit 23 next determines whether V ₊ −V ₀ (= ΔV +) ≧ 0 (step S233). _{If +} −V ₀ ≧ 0 (YES in step S233), the correction coefficient k = kp2 is read from the memory 24 (step S234), and the correction value Vc ₊ is calculated (step S238).

When V ₊ −V ₀ (= ΔV +) ≧ 0 is not satisfied (NO in step S233), the computing unit 23 next determines whether or not V ₊ −V ₀ (= ΔV +) <− ΔVth (step S235). If V ₊ −V ₀ <−ΔVth (YES in step S235), the correction coefficient k = kp4 is read from the memory 24 (step S234), and the correction value Vc ₊ is calculated (step S238).

When V ₊ −V ₀ (= ΔV +) <− ΔVth is not satisfied (NO in step S235), the arithmetic unit 23 next reads the correction coefficient k = kp3 from the memory 24 (step S237) and calculates the correction value Vc ₊ . (Step S238).

In step S24, the subtractor 25 calculates the electrode voltage corrected in steps S22 and S23. According to the above equation (4), the corrected voltage V1 is represented by V ₀ − (Vc ₋ + Vc +).

  The above-described processing of step S2 (S21 to S24) is performed on all the pixels V [N] (N = 1, 2,...). As a result, a correction coefficient k that provides a desired deflection pattern is selected, and the voltage value of the deflection pattern is corrected.

  As a result, as shown in FIG. 5C, a deflection pattern in which the corrected electrode voltage is set is generated (step S3). As a result, the corrected deflection pattern is corrected in the voltage value while maintaining a state close to the ideal phase with little distortion at the folded portion of the periodic pattern, and as a result, optical crosstalk is suppressed.

  As described above, according to the optical switch 1 of the present embodiment, a deflection pattern to be applied to each pixel of the liquid crystal element is generated, and the correction coefficient selected from the voltage difference between each pixel and a plurality of correction coefficients. Is used to calculate a correction value proportional to the voltage difference, and the deflection pattern is corrected based on this correction value. Here, a coefficient that minimizes the phase crosstalk between the pixels is selected as the correction coefficient. Thereby, optical crosstalk is suppressed.

  Next, a modification of the optical switch 1 of the present embodiment will be described.

(Modification 1)
In the above, the case where a preset correction coefficient is selected from the relationship between the voltage difference between pixels and the threshold value ΔVth has been described with reference to FIGS. 1, 4, 7, and 8. However, the calculator 23 may calculate the correction value using a correction coefficient proportional to the voltage difference between the pixels.

  The entire processing executed to realize the correction of the deflection pattern in the first modification will be described with reference to FIGS. 1, 9, and 10. FIG.

  FIGS. 9 and 10 are flowcharts similar to FIGS. 7 and 8 and show a case where a correction coefficient proportional to the voltage difference between pixels is used. The electrode voltage acquisition process (step S1) and the voltage information acquisition process (step S21) in FIG. 9 are the same as those shown in FIG. Also, the electrode voltage correction process (step S24) and the deflection pattern correction process (step S3) in FIG. 10 are the same as those shown in FIG. In the following, processing in steps S22A and S23A in FIG. 9 different from those shown in FIGS. 7 and 8 will be described.

In step S22A, a correction value for the minus side pixel is calculated. At this time, the computing unit 23 selects a correction coefficient that continuously changes in accordance with the applied voltage of the pixel. The specific processing in step S22A is shown in steps S221A to S224A described later. First, the arithmetic unit _{23, V - -V 0 (= ΔV-} ) to determine whether ≧ 0 (step S221A), _V - (YES in step S221A) For -V ₀ ≧ 0, the correction coefficient k, αm1 is read from the memory 24, and is multiplied by | V ₋ −V ₀ | to be k (step S222A). Here, | V ₋ −V ₀ | represents the absolute value of the voltage difference ΔV−, and αm1 represents a preset coefficient.

The computing unit 23 calculates a correction value Vc ₋ (step S224A). According to the flowchart of FIG. 9, as indicated by the above equation (2), the correction value Vc ₋ is calculated from k · (V ₋ −V ₀ ) (= k × ΔV−).

When V ₋ −V ₀ (= ΔV−) ≧ 0 is not satisfied (NO in step S221A), the arithmetic unit 23 reads αm2 from the memory 24 as the correction coefficient k and multiplies it by | V ₋ −V ₀ |. K (step S223A), the process proceeds to step S224A. Αm2 represents a preset coefficient.

In step S23A, similarly to step S22A described above, a correction value for the plus side pixel is calculated. Specific processing in step S23A is shown in steps S231A to S234A described later. First, the computing unit 23 determines whether V ₊ −V ₀ (= ΔV +) ≧ 0 (step S231A). If V ₊ −V ₀ ≧ 0 (YES in step S231A), αp1 is used as the correction coefficient k. Is read from the memory 24 and multiplied by | V ₊ −V ₀ | to obtain k (step S222A). Here, | V ₊ −V ₀ | represents the absolute value of the voltage difference ΔV +, and αp1 represents a preset coefficient.

Then, the calculator 23 calculates the correction value Vc ₊ (step S234A). According to the flowchart of FIG. 9, as indicated by the above equation (3), the correction value Vc ₊ is calculated from k · (V ₊ −V ₀ ) (= k × ΔV +).

When V ₊ −V ₀ (= ΔV +) ≧ 0 is not satisfied (NO in step S231A), the arithmetic unit 23 reads αp2 from the memory 24 as the correction coefficient k, and multiplies it by | V ₊ −V ₀ |. k (step S233A), the process proceeds to step S234A. Αp2 represents a preset coefficient.

  Even in this case, in the process shown in FIG. 10, the deflection pattern is corrected in the same manner as in step S3 in FIG. 8 (step S3). Therefore, optical crosstalk can be suppressed.

(Modification 2)
In the above, the correction coefficient k has been treated as being selected in consideration of the orientation direction, the positive / negative of the voltage difference and / or the magnitude of the voltage difference. However, the correction coefficient k is determined according to the light attenuation rate. May be.

  In general, when an attenuation pattern is added to a deflection pattern, the size and phase of the deflection pattern become complicated. The attenuation pattern is a pattern that is superimposed on the deflection pattern and has a cycle shorter than the cycle of the deflection pattern. When the attenuation pattern is superimposed on the deflection pattern, the main signal coupled to a certain port is modulated with the period of the attenuation pattern, and a part of the power of the main signal is transferred to the modulated light. As a result, the main signal is attenuated. When the modulated light is coupled to another port, large optical crosstalk occurs. Since the attenuation pattern has a shorter period than the deflection pattern, it is easily affected by disclination.

  For this reason, in the case of the orientation pattern in which the attenuation pattern is superimposed, it is considered that the optimum correction coefficient k cannot be selected if only the deflection pattern having a regular period is considered. Therefore, in the optical switch 1 in Modification 2 of this embodiment, the calculator 23 selects the correction coefficient k corresponding to the attenuation rate.

  In the optical switch 1 according to the modified example 2, first, in order to obtain the optimum correction coefficient k corresponding to the attenuation rate, the correction coefficient k is applied in a state where a deflection pattern with an attenuation rate is applied to the liquid crystal element 10 at a desired port. The value is changed within a certain range, and the value of the correction coefficient k that minimizes the optical crosstalk is obtained from the range. The correction coefficient k is not always expressed by a mathematical function of the attenuation rate. Therefore, in the optical switch 1 of the second modification, a table in which the value of the correction coefficient k is recorded for each attenuation rate and the table is referred to, or the correction coefficient k is obtained as an approximate function of the attenuation rate. The correction coefficient k corresponding to the attenuation rate is calculated by recording the coefficient of the approximate function.

  Hereinafter, a case where an attenuation pattern is superimposed on a deflection pattern will be described. In this case, a specific connection destination port is set in the optical switch 1, and an attenuation state is generated by superimposing an attenuation pattern for achieving a specific attenuation rate in the state on the deflection pattern. In this state, when the deflection pattern corrected in the flowcharts of FIGS. 7 and 8 described above or the flowcharts of FIGS. 9 and 10 is displayed on the liquid crystal element 10, this deflection pattern has some influence on the optical crosstalk.

  In the optical switch 1 in the modified example 2, the value of the correction coefficient k is changed in a certain range in advance, and the dependence of optical crosstalk is obtained with respect to the given correction coefficient k. The dependency of optical crosstalk is whether or not a minimum phase crosstalk can be obtained (for example, whether a deflection pattern having a corrected voltage value has a predetermined phase (no distortion from the design conditions)). Whether or not (see FIG. 12B), or the degree of optical crosstalk due to scattering (measured value) on the screen based on the deflection pattern having the corrected voltage value (see FIG. 12C). Judgment)). Based on the dependency, a correction coefficient k that minimizes optical crosstalk is determined in advance. The correction coefficient k determined in this way is previously tabulated on the memory 24 in association with the specific port and the specific attenuation rate.

  For example, when an optical switch is set to a certain connection destination port and a certain attenuation factor, the computing unit 23 refers to the above-described table and calculates the correction coefficient k associated with the connection destination port and the attenuation factor. Using the correction coefficient k, the correction processing based on the calculation formula (1) is performed. This suppresses optical crosstalk even in the attenuated state.

  Note that the above table can be huge if it has correction coefficients associated with all connected ports and all attenuation factors. Therefore, for example, assuming that the correction coefficient k is an approximate function of the attenuation rate, the correction coefficient in this approximate function may be stored in the memory. In this way, the memory capacity can be reduced although an error due to the approximation function is included.

  In this way, the spatial phase modulator of the present embodiment calculates the optimum correction coefficient k that minimizes the optical crosstalk according to the connection destination port and / or the attenuation rate, and sets the deflection pattern or / and the attenuation pattern. to correct. Therefore, an optical switch using this spatial phase modulator can suppress optical crosstalk caused by liquid crystal disclination to a low level.

  In the above description, for ease of understanding, the deflection pattern has been described as the voltage applied to the electrode. However, in actuality, the deflection pattern is given with a gray scale value of 8 to 12 bits instead of a physical voltage. Is common. However, since this gray scale value is a value proportional to the applied voltage or can be converted to the applied voltage through γ correction, the present invention is applied to the deflection pattern given in gray scale. Easy.

1 Optical switch (spatial phase modulator)
10 Liquid crystal element 20 Controller 21, 23 Operation unit 22, 24 Memory 25 Subtractor

The present invention for solving the above-described problem is a spatial phase modulator having a liquid crystal element, a voltage pattern generation unit for generating a deflection pattern to be applied to each pixel of the liquid crystal element, and a voltage between the pixels. A correction value calculating unit that calculates a correction value proportional to the voltage difference using a difference and a correction coefficient selected from a plurality of correction coefficients, and a correction that corrects the deflection pattern based on the correction value The correction value calculation unit calculates the correction coefficient that provides a desired deflection pattern according to the alignment direction of the liquid crystal, the sign of the voltage difference, the magnitude of the voltage difference, or the light attenuation factor. select.

Further, the present invention for solving the above-described problem is a method for correcting a driving voltage of a spatial phase modulator having a liquid crystal element, the step of generating a deflection pattern to be applied to each pixel of the liquid crystal element, Using the voltage difference between each pixel and a correction coefficient selected from a plurality of correction coefficients, calculating a correction value proportional to the voltage difference; and correcting the deflection pattern based on the correction value The step of calculating the correction value includes a step of calculating a correction value according to an orientation direction of the liquid crystal, a sign of the voltage difference, a magnitude of the voltage difference, or a light attenuation factor. Select a correction factor.

Claims (5)

A spatial phase modulator having a liquid crystal element,
A voltage pattern generation unit that generates a deflection pattern to be applied to each pixel of the liquid crystal element;
A correction value calculation unit that calculates a correction value proportional to the voltage difference using a voltage difference between the pixels and a correction coefficient selected from a plurality of correction coefficients;
A correction unit that corrects the deflection pattern based on the correction value,
The spatial phase modulator, wherein the correction value calculation unit selects the correction coefficient that provides a desired deflection pattern.   The correction value calculation unit selects the correction coefficient corresponding to the orientation direction of the liquid crystal, the sign of the voltage difference, the magnitude of the voltage difference, or the light attenuation rate. The spatial phase modulator according to 1.   The spatial phase modulator according to claim 2, wherein the correction value calculation unit selects the correction coefficient associated in advance according to an applied voltage of a pixel.   The spatial phase modulator according to claim 2, wherein the correction value calculation unit selects the correction coefficient that continuously changes in accordance with an applied voltage of a pixel. A method for correcting a driving voltage of a spatial phase modulator having a liquid crystal element,
Generating a deflection pattern to be applied to each pixel of the liquid crystal element;
Calculating a correction value proportional to the voltage difference using a voltage difference between the pixels and a correction coefficient selected from a plurality of correction coefficients;
Correcting the deflection pattern based on the correction value,
The step of calculating the correction value comprises selecting the correction coefficient that provides a desired deflection pattern.