SU590739A1 - Optoelectronic computer - Google Patents

Description

The invention relates to the field of optoelectronics and can be used in automation and computer technology.

Known ontoelectronic computing devices. One of them performs the functions of a shift register and is a transparent substrate, on one side of which light-emitting elements 1 are deposited. Particularity of the register is the presence of an encoder optically associated with the elements of the electroluminescent layer. The encoder is made in the form of parallel-arranged supply and output conductive shells connected by photoresistive jumpers in accordance with the adopted coding system.

The closest technical solution to the invention is an optoelectronic computing device containing optical fibers, a plane-parallel inverting transducer and light sources connected to the input of device 2. The disadvantage of such a device is its limited functionality.

The aim of the invention is to expand the functionality of the device by performing the operation of the summation of numbers. The goal is achieved by introducing a decoder made on optical fibers, one ends of which are combined into three binary focal inputs of the device, and others into the main focal outputs optically connected with the corresponding inputs of an iloconconverse inverting inverter, and the plane-parallel inverting converter is provided with additional optical input and output connected by optical fibers with a binary output "device transfer, summing up the output of which the optical fiber. th discharge group Fauconnier outputs, as additional input Fauconnier - a second group of bit Fauconnier plane-inverting output transducer.

The drawing shows a diagram of the proposed device.

Part of the light guides / forms the decoder 2. Some of the ends of these light guides, combined into focons, form binary input terminals of 3, 4 V 5 devices. Each binary input contains two harnesses, one of which consists of three, and the second of four light guides. The other ends of the light guide of the decoder are combined into three to seven main focal outputs 6 of the decoder. Along with these outputs, there are bit-type focal outputs 7. Some of these outputs are combined into a focon 8 by another optical fiber system, which is the summing output of the device. The main outputs 6 are optically coupled to the corresponding inputs of the nloscalone inverting inverter 9, the outputs of which are connected to the discharge focal outputs 7.

The plane-parallel inverting transducer 9 is a multilayer structure enclosed between glass plates W. Transparent electrodes are applied to the inner sides of the plates. One of these electrodes is in contact with the electroluminescent layer 12, the other side of which is covered by a light-impermeable partition 13. A thin layer of a transparent dielectric 44 is adjacent to the other electrode 11. control grid 16. Grid 16 is a system of tungsten wires with a diameter of about 10 microns with a pinch of 250-500 microns. The grid 16 is galvanically connected to a transparent electrode and in contact with the electroluminescent layer 12.

Both transparent electrodes are connected to the power source / 7 terminals, for which the alternating voltage generator is used. The inverting converter 9 also has an additional focon input / 5 and a corresponding additional focon output 19. The additional input and output are optically coupled to a binary output device transfer, which consists of focons 20 and 21. In addition, the additional focal input 18 is optically connected (e.g., optical fibers) with a part of the bit outputs 7. The registers of the terms are represented by light sources 22 and 23 connected to binary inputs 4 and 5, respectively.

On one inverting converter 9, several of the same type of computing devices can be mounted, the number of which is determined by the number of binary digits of the numbers being summed. At the same time, the "transfer" output (focons 20 and 21) is connected to the binary output 5 of the next high-resolution computing device.

The unit is encoded by illuminating the odd-numbered binary inputs 3, 4, and 5, and zero is illuminated by illuminating the even-numbered focon of these inputs. On the binary input 3, the unit is transferred from the younger digit, and the binary inputs 4 and 5 receive signals from the corresponding bits of the registers of the components 22 and. 23. With any combination of zeros and ones on the three binary inputs, one of the main focal outputs 6 will be darkened. Only in the case when the unit is absent on all binary inputs, all the main exits 6 of Ocep1, en. The numbers of the darkened focal exits 6 are determined by the formula

L 3Wi + 2W2 + V, + 2 (-l) (2-C) (3-E).

where L is the darkened exit number, counting

from left to right in the drawing: Wi is the weight of binary input 3; Wz is the weight of binary input 4; Wy is the weight of binary input 5;

With Wi + Wz + Ws.

Alternating voltage applied to

transparent electrodes // excites a capacitive current of mis1, enn, flowing 1Y through the layers enclosed between the electrodes //. The passage through the electroluminescent layer 12, the current causes radiation from its surfaces, which is supplied to the discharge focal outputs 7 and the additional focal output 9. If, and any portion of the photoconductive layer / 5 is illuminated from the main focal outputs 6 or the secondary input M then, due to the induced conductivity, this section is galvanically connected through the grid 16 to the transparent electrode 11. Therefore, the capacitive current, branching to the il6 grid, bypasses the corresponding section

electroluminescent layer 12. As a result, a non-emitting output section of the inverting converter 9 turns out against the illuminated part of the input. The jumbled image has an enhanced congruity that favorably affects the characteristics of the device as a whole.

Consider as an example the addition of the following numbers: О,, 1, 1. This will highlight the even tint of binary input 3 and odd

Focons of binary inputs 4 and 5, as shown in the drawing. Substituting in the expression (1) the values IF, 0, 2 1, Wz I. we get N 5, that is, it will be darkened fifth of the main outputs 6. The corresponding one of the bit outputs 7 will be radiating. Radiation by means of optical fibers is directed to the additional input 18 of the binary input transfer (focon 20). Since the focon 20 emits light and the focon 21 is dark, according to the received coding, 1 is recorded for transfer to the high-order bit. None — OF discharge outputs 7; connected to the focon 8, does not emit, which corresponds to the value “On the sum.

A crossover converter 9 allows 20-40 channels per centimeter to be placed in one direction and 100 channels per centimeter in another orthogonal direction. So on the size converter

10X10 cm can accommodate 4-LO channels. For this computing device you need 8 channels. Consequently, with the help of converter 9 of the indicated sizes, it is possible to add binary

numbers with 5-10 bits. The number of bits is not limited by optical loss, since the converter 9 operating in the image reversal mode can amplify the light c-signals. Since the light flux on binary