JPH0711651B2 - Subtraction arithmetic unit - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a subtraction calculation device using an optical calculation device such as a spatial light modulator.

(Prior Art) The subtraction operation between two numerical values is performed by using a digital operation circuit using an integrated circuit.

In order to perform such an operation, a numerical value is transferred to an arithmetic circuit called an adder, and a complement operation is performed to perform a subtraction operation.

Since this arithmetic circuit can usually perform only one subtraction operation, it is necessary to repeat many subtraction operations or prepare a large number of arithmetic circuits in order to subtract a large number of numerical values.

In addition, the representation of numerical values that can be handled is limited to the binary system, which is a numerical value modulo 2.

(Problems to be Solved by the Invention) Due to such a large amount of subtraction operations, a large number of subtraction operations have been performed or a large amount of arithmetic circuits have been required.

Further, the subtraction of a numerical value having more digits than the number of digits prepared in the arithmetic circuit has a drawback that the number of subtraction operations is further increased.

Furthermore, there is a drawback in that the numerical expressions that can be handled by the arithmetic circuit are limited to binary numbers that are modulo 2.

The object of the present invention is to perform the subtraction of a large amount of numerical values, which is the above-mentioned drawback, by a simple operation, to minimize the increase in the number of subtraction operations due to the restriction of the number of digits, and to express the numerical value that can be handled is an arbitrary number. An object of the present invention is to provide a subtraction device having a new configuration that can take a value modulo.

(Means for Solving Problems) In order to achieve the above object, the subtraction operation device according to the present invention has an intensity corresponding to a difference between light intensities of sequentially input bright points having a storage function and a subtraction function. A first optical arithmetic unit that outputs the bright spot group that it has, and a threshold function for the incident bright spot group that has a recording function and a threshold processing function, and stores and outputs the bright spot group that has binarized light intensity. A second optical operation device and first and second input devices for converting numerical information of a subtraction target modulo an arbitrary number into light intensities proportional to the numerical values and arranging them as bright spot groups in a plane. A light source device for causing light having a uniform light intensity to be incident on the first optical operation device, and individual luminescent spots of the bright spot groups output from the first optical operation device and the second optical operation device. A photoelectric conversion device for converting the light intensity of a point into numerical information modulo an arbitrary number and outputting the numerical information, It includes optical means for forming an image by superimposing a plurality of bright spot groups output from the first and second input devices and a bright spot group output from the second optical arithmetic device.

Then, the number modulo At, Bt: n BOt: The number borrowed from the upper digit t: The index indicating the number of digits is At + n is At + n- (Bt + BOt _-1 ) BOt is At + n- (Bt + BOt _-1 ) + n. When BOt is At + n− (Bt + BOt ₋₁ ) + n · BOt−n, an output corresponding to At from the first input device and an output corresponding to n from the light source device are output to the first arithmetic unit. Writing is performed, and light corresponding to Bt and BOt ₋₁ from the second input device is applied to the first arithmetic device to obtain output light corresponding to the subtraction operation. The output is written to the second arithmetic device to control the threshold value to obtain an output, and the output corresponding to n from the light source device is applied to the first arithmetic device to perform subtraction and recording, Is output by a photoelectric conversion device.

(Example) The present invention will be described in more detail with reference to the drawings.

FIG. 1 is a diagram for explaining the basic configuration and operation of a spatial light modulator used as the first and second optical arithmetic units, which are the basic components of the embodiment of the subtraction arithmetic unit according to the present invention. It is a schematic diagram.

The image 1 incident on the photocathode 3 of the spatial light modulator via the lens 2 is converted into a photoelectron image.

The photoelectron image is multiplied by the microchannel plate 4 and then forms a charge pattern on the surface 61 of the crystal 6. The electric field across the crystal 6 changes according to the charge pattern, and the refractive index of the crystal 6 changes due to the Pockels effect.

When the crystal 6 is irradiated with the linearly polarized light 8 here, the charge storage surface
Since the polarization state of the reflected light from 61 is changed, the light output 10 having a light intensity corresponding to the light intensity of the input light 1 can be obtained by passing through the analyzer 9.

Next, the main functions of the spatial light modulator required for the present invention will be described.

(1) Memory Function The spatial light modulator has the function of maintaining the charge distribution on the surface of the electro-optic crystal for a long time. Since the crystal 6 has a very high electric resistance value, the charge distribution on the crystal surface 61 can be maintained for several days or longer.

(2) Subtraction function The spatial light modulator can selectively form a positive or negative charge distribution on the surface of the electro-optic crystal. FIG. 2 is a graph showing the secondary electron emission characteristics of the electro-optic crystal surface.

As shown in FIG. 2, when the primary electron energy E incident on the crystal surface 61 is smaller than the first crossover point E ₁ or larger than the second crossover point E ₂ , the number of primary electrons is Since the number of secondary electrons is larger than (δ <1), the crystal surface is negatively charged. If the energy of primary electrons is between E ₁ and E ₂ , the number of secondary electrons is larger than the number of primary electrons (δ> 1) and the crystal surface is positively charged.

This positive / negative writing is executed by controlling the voltages of Vc and Vb shown in FIG.

Here, the subtraction function can be provided by the method of first writing the negative charge once and then the positive charge, and the method of writing the negative charge first and then the negative charge. The amount to be subtracted can be controlled by the following three methods.

One is a method of changing the incident light intensity at the time of subtraction, the other is a method of changing the duration of the voltage 41 applied to the microchannel plate 4, and a method of changing the voltage 41 applied to the microchannel plate 4. .

(3) Threshold Control Function As shown in FIG. 3, let us consider a case of writing an image having three brightness levels of A, B, and C.

The electron density on the charge storage surface 61 of the crystal 6 corresponding to the brightest portion B is highest, and decreases in the order of A and C.

In Figure 4, the output light intensity of the A moiety are I _0/2, B part
The I ₀ and C portions correspond to 0.

In this state, by lowering the voltage Vb shown in FIG. 1, only the portion B having the most negative charges can be set to the negative potential.

An example will be described below.

In FIG. 1, it is assumed that the input light shown in FIG. 3 is written with Vb = 2KV and Vc = 1KV. The area corresponding to the pattern B has the largest electron attachment amount, the potential of the mesh electrode reaches 1 KV, and the potential of the portion corresponding to A reaches 1.5 KV,
It is assumed that the portion corresponding to C where almost no electrons are attached is 2KV which is the same as Vb.

If Vb is 0.5 KV, the potential of the portion corresponding to B is -0.5 KV, the potential of the portion corresponding to A is 0.0 KV, and the potential of the portion corresponding to C is +0.5 KV.

When the photocathode 3 of the spatial light modulator is irradiated with uniform light with the mesh voltage Vc set to 0.5 KV, for example, the portion B repels the incident electrons at a negative potential, and the portion corresponding to A has that portion. Electrons will be injected until becomes 0.5 KV. As a result, the charges other than the portion corresponding to B can be made zero.

Setting a certain charge level and binarizing it above or below that level is called a threshold control function.

In general, the formulas for subtracting values from each other modulo an arbitrary number n are represented by the following (1) and (2).

Here, At and Bt are numerical values modulo n, SUBt is a numerical value of the subtraction result, BOt is a numerical value borrowed from the upper digit, and t is a subscript at the t-th digit.

SUBt = At-Bt-BOt- ₁ + n-BOt (1) BOt = 0 (At-Bt-BOt- ₁ ≧ 0) BOt = 1 (At-Bt-BOt- ₁ <0) (2) Here, the equations (1) and (2) are modified as follows: SUBt = (At + n)-(Bt + BOt- ₁ ) -n + n.BOt ...
(3) BOt = 0, [2n-1- {At + n- (Bt + BOt- ₁ )} <n] BOt = 1, [2n-1- {At + n- (Bt + BOt- ₁ )} ≧ n]
(4) Next, the procedure of subtraction is shown.

Calculate At + n.

At + n- (Bt + BOt- ₁ ) is calculated.

Ask for BOt.

At + n- (Bt + BOt- ₁ ) + n.BOt is calculated.

At + n- (Bt + BOt- ₁ ) + n.BOt-n is calculated.

Repeat ~ for the required number of digits.

6 and 7 are graphs showing the operation patterns in the binary and ternary systems.

The above subtraction operation will be described based on an example of a subtraction operation of a number modulo 2.

FIG. 5 is a block diagram showing an embodiment of the subtraction arithmetic unit according to the present invention.

The first input device is formed by the LED array 11, and the second input device is formed by the LED array 12. The numerical information of the subtraction target modulo an arbitrary number is converted into light intensity proportional to the numerical value, and in the plane. Are displayed as bright spot groups.

The plurality of bright spot groups output from the first and second input devices and the bright spot group output from the second optical computing device (spatial light modulation tube 38) described later are respectively transmitted via optical means. The first optical arithmetic unit (spatial light modulator 37) is input and superposed.

The first light calculation device is formed by a spatial light modulator tube 37 having a storage function and a subtraction function and capable of outputting a bright spot group having a light intensity corresponding to the difference between the light intensities of sequentially input bright spots. Has been done.

The second optical arithmetic unit has a storage function and a threshold processing function, performs threshold processing on the incident bright spot group, stores the bright spot group having a binarized light intensity, and outputs the spatial light modulator tube 38. Has been formed.

The first input device is formed by the LED array 11, and the second input device is formed by the LED array 12. The numerical information of the subtraction target modulo an arbitrary number is converted into light intensity proportional to the numerical value, and in the plane. Are displayed as bright spot groups.

The light 15 having a uniform light intensity supplied from a light source (not shown) through the shutter 23 passes through the half mirrors 30, 31 and the lens 25 and is the spatial light modulation tube which is the first optical operation device.
It is incident on 37.

The light from the spatial light modulator 37, which is the first optical operation device, is incident on the light receiving element array 48, which is a photoelectric conversion device, via the analyzer 46, and the light of each bright point of the bright point group is incident. Strength is
It is output after being converted into numerical information modulo any number.

The polarizer 44 converts light from a light source (not shown) into linearly polarized light to form readout light.

The readout light 17 split by the half mirror 36 is supplied to the spatial light modulation tube 38 through the half mirror 34. Further, the read light 16 is supplied to the spatial light modulator 37 through the mirror 39 and the half mirror 32.

The implementation of the above-mentioned subtraction operation by the apparatus of the embodiment will be described together with the detailed configuration.

At + 2 The outputs of LED arrays 11 and 12 are At and Bt, respectively.

This output takes in numerical information modulo 2 as an electric signal 0,1 and converts it into a light intensity proportional to the electric signal 0,1 and displays it as a bright spot group in the plane of the LED array 11,12. Is.

The light intensity of the light 15 has a light intensity proportional to the modulus 2 and is incident on the half mirror 30 via the shutter 23.

With the shutter 21 closed (or the LED array 12 turned off) and the shutter 22 open, the LED array 11 and the Hoof mirror 2 are turned on.
The light (At) supplied through 9 and the light 15 are superimposed on each other by the half mirror 30 and the lens 25, and are imaged on the photocathode 3 of the spatial light modulation tube 37.
Write At + 2.

At + 2− (Bt + BOt− ₁ ) LED array 11 and light 15 are turned off (or shutters 22 and 23 are closed) to cause photocathode 3 of spatial light modulator tube 37 to be (Bt +
A light having a light intensity of BOt- ₁ ) is made incident.

Here, the subtraction operation of the spatial light modulator described above is performed.

By placing the analyzer 45 in the parallel Nicol state with respect to the polarizing plate 44 so that the reflected light 18 when there is no charge on the crystal surface 50 is passed, the output light 18 is 2.2-1- {At + 2.
It has a light intensity corresponding to − (Bt + BOt− ₁ )}.

BOt Output light from the first spatial light modulator 37 to the second spatial light modulator 38
Write 18.

Further, using the threshold control function described in 3 above, only the brightness corresponding to the brightness having a light intensity of 2 or more is output.

Here, the output light 20 of the spatial light modulator 38 has a light intensity of BOt.

The output light 20 can be held by the storage function of 1, and if necessary, the light receiving element array 49 can obtain borrowed information from the upper digit as an electric signal.

At + 2- (Bt + BOt- ₁ ) + 2.BOt The spatial light modulator 37 stores At + 2- (Bt + BOt- ₁ ).

Here, the shutters 21 and 22 are closed, or the LED arrays 11 and 12 and the light 15 are turned off, and only the light 20 of BOt is incident on the spatial light modulation tube 37 for writing (spatial light modulation tube 38 → half mirror). 34 → mirror 43, 42 → analyzer 47 → lens 27 → half mirror 35 → lens 28 → mirror 40 → shutter 24 → half mirror 31 → lens
25 → spatial light modulator 37).

As a result, At + 2- (Bt + BOt- ₁ ) + 2.BOt is obtained.

The addition of the doubled BOt can be obtained by controlling the light intensity of the read light 17, controlling the microchannel plate drive voltage 41, or controlling the duration thereof.

At + 2- (Bt + BOt- ₁ ) + 2.BOt-2 Shutters 21, 22 and 24 are closed (or LED arrays 11 and 12 are turned off) and shutter 24 is closed. It is incident on the light modulation tube 37.

Here, the subtraction function of the spatial modulation tube 37 is used to perform subtraction of -2, and the analyzer 46 is in a crossed Nicol state so as to block the output light 19 when the crystal surface 50 has no electric charge. The output light 19 is At + 2- (Bt-BOt-
₁ ) Has a light intensity corresponding to the subtraction result of + 2 · BOt−2. The output light 19 on which the analyzer 46 has dropped can be converted into an electric signal by the light receiving element array 48, and the numerical information of the subtraction result can be obtained.

The above operation is repeated by the number of digits to be subtracted from the lower digits.

The above subtraction is performed not only for one numerical value to be subtracted but also commonly for all bright points arranged in the plane.

That is, the subtraction of a plurality of numerical values is being processed at the same time.

In the above embodiment, the subtraction operation is performed on the numerical value modulo 2, but the numerical information to be set in the input At and Bt and the output of the light receiving element are analog values, and the addition of n times BOt is performed by the microchannel. Control the plate drive voltage 41, control its duration or read light 17 from device 2.
By controlling the light intensity of, the subtraction can be performed on a numerical value modulo an arbitrary number n by means of adding n times.

(Effects of the Invention) As described in detail above, the subtraction arithmetic device according to the present invention has a bright point group having a light storage function and a subtraction function and having a light intensity corresponding to the difference between the light intensities of sequentially input bright points. A second optical calculation device that outputs a first optical calculation device, and has a recording function and a threshold processing function, performs threshold processing on an incident bright spot group, and stores and outputs a bright spot group having a binarized light intensity. A device and numerical information to be subtracted modulo an arbitrary number are converted into light intensities proportional to the numerical values, respectively, and are arranged as a bright spot group in a plane.
And a second input device, a light source device that causes light having a uniform light intensity to enter the first optical operation device, and a brightness output from the first optical operation device and the second optical operation device. A photoelectric conversion device for converting the light intensity of each bright point of the point group into numerical information modulo an arbitrary number and outputting the information, and a plurality of bright point groups output from the first and second input devices. It includes optical means for forming an image by superposing each other and the bright spot group output from the second optical arithmetic unit.

Then, the number modulo At, Bt: n BOt: The number borrowed from the upper digit t: The index indicating the number of digits is At + n is At + n- (Bt + BOt _-1 ) BOt is At + n- (Bt + BOt _-1 ) + n. When BOt is At + n− (Bt + BOt ₋₁ ) + n · BOt−n, an output corresponding to At from the first input device and an output corresponding to n from the light source device are output to the first arithmetic unit. Writing is performed, and light corresponding to Bt and BOt ₋₁ from the second input device is applied to the first arithmetic device to obtain output light corresponding to the subtraction operation. The output is written to the second arithmetic device to control the threshold value to obtain an output, and the output corresponding to n from the light source device is applied to the first arithmetic device to perform subtraction and recording, Is output by a photoelectric conversion device.

Therefore, a large number of subtractions modulo an arbitrary number can be executed with a small number of operations.

Therefore, the present invention can be widely used as one of the arithmetic units of an optical computer.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram for explaining the configuration and operation of a spatial light modulator which is a basic component of a subtraction arithmetic unit according to the present invention. FIG. 2 is a graph showing secondary electron emission characteristics on the crystal surface of the spatial light modulator. FIG. 3 is a schematic diagram for explaining threshold control of the spatial light modulator. FIG. 4 is a graph showing the relationship between the crystal surface charge amount δ, the output light intensity I, and the output light phase difference φ. FIG. 5 is a block diagram of the subtraction arithmetic unit according to the present invention. FIG. 6 is a graph showing a logical table of a subtraction arithmetic unit modulo 2. FIG. 7 is a graph showing a logical table of a subtraction arithmetic unit modulo 3. 1 …… Input light 2,25 ～ 28 …… Lens 3 …… Photocathode 4 …… Microchannel plate 5 …… Mesh 6 …… Crystal 61,50 …… Crystal surface 41 …… Microchannel plate drive voltage 7,29 ~ 36 …… Half mirror 8,16,17 …… Readout light 10,18,19 …… Output light 11,12 …… LED array 13,14 …… Subtracted bright spots 15 …… Uniform light intensity With light 16,17 …… Reading light 20 …… Output light (BORROW bright spot group) 21 ～ 24 …… Shutter 37,38 …… Spatial light modulator 39 ～ 40,42,43 …… Mirror 44 ...... Polarizers 45 to 47 …… Analyzer 48,49 …… Receiving element array

Claims (2)

[Claims] 1. A first optical arithmetic unit having a storage function and a subtraction function, which outputs a bright spot group having a light intensity corresponding to a difference between the light intensities of sequentially inputted bright spots, a recording function and a threshold processing. A second optical arithmetic unit that has a function and performs threshold processing on an incident bright spot group and stores and outputs a bright spot group having a binarized light intensity, and numerical information to be subtracted modulo an arbitrary number. Respectively into a light intensity proportional to the numerical value and arranged in a plane as a bright spot group, and a light having a uniform light intensity is incident on the first light calculation device. And a light source device for converting the light intensity of each bright spot of the bright spot group output from the first light computing device and the second light computing device into numerical information modulo an arbitrary number and outputting And a plurality of bright spot groups output from the first and second input devices, and 2 is composed of optical means for superimposing an image with the bright spot group output from the second optical arithmetic unit, and outputs an output corresponding to At from the first input device and an output corresponding to n from the light source device. Writing to the first arithmetic unit, irradiating the first arithmetic unit with light corresponding to Bt and BOt ₋₁ from the second input unit to obtain output light corresponding to the subtraction operation, The output of the first arithmetic device is written to the second arithmetic device to control the threshold value to obtain an output, and an output corresponding to n from the light source device is applied to the first arithmetic device to perform subtraction. A subtraction arithmetic device for recording and performing mutual subtraction of numerical values modulo an arbitrary number configured to output the output by a photoelectric conversion device. However: At, Bt: Numeric value modulo BOt: Numeric value borrowed from the upper digit t: Subscript indicating the number of digits At + n is At + n- (Bt + BOt _-1 ) BOt is At + n- (Bt + BOt _-1 ) + n. BOt is At + n- (Bt + BOt _-1 ) + n.BOt-n 2. The subtraction arithmetic unit according to claim 1, wherein the first and second optical arithmetic units are spatial light modulators.