SU830430A1 - Function generator - Google Patents

Description

(54) FUNCTIONAL CONVERTER

The invention relates to automation and computing and can be used, in particular, in hybrid computing systems and devices. A functional converter is known that contains a control unit, a multistable element, switches and blocks for specifying abscissas and ordinates, a pulse-width modulator, a block of buffer amplifiers, and an output non-linear interpolator 1. Also known is a functional converter containing an argument register, digital-to-analog converters, controllable dividers voltages, descrambler, functional conductivity, and operational amplifier 2. The disadvantage of these devices is structural complexity. The closest in technical essence and the achieved result to the proposed is a functional converter containing an analog-to-digital converter connected by an input to an argument input bus and to a signal input of a controlled inverter, and outputs to inputs of a memory block whose outputs are connected to inputs of the first and second registers connected to the outputs of the bits to the digital inputs of the first and second digital-to-analog converters, respectively, connected by outputs to the inputs of the output adder, with The analog input of the first digital-to-analog converter is connected to the output of a controlled inverter connected by a control input to the output of a sign bit of the first register, and the analog inputs of the second digital-analog converter are connected to the reference voltage 3. The disadvantage of this converter with a fixed memory block is the reduced accuracy of the functional converter, which is due to the use of only the piecewise linear nature of the approximation of functions. The purpose of the invention is to increase the accuracy of the functional transformation. The goal is achieved in that a functional converter containing an analog-to-digital converter connected by input to the input argument bus, and outputs to the inputs of a memory block whose outputs are connected to the inputs of the first and second registers connected to the outputs of the bits to digital inputs, respectively the first and second digital-to-analog converters connected by outputs to the inputs of the output adder, the analog input of the first digital-to-analog converter connected to the output of the controlled inverter the main input connected to the control input with the output of the sign bit of the first register, and the analog inputs of the second digital-to-analog converter are connected to the reference voltage buses; a multiplying unit, a trigger and a control switch connected by the first signal input to the multiplication unit output are added, the second signal input - with the argument input bus and with the inputs of the multiplying unit, the output with the signal input of the controlled inverter, and the control input with the output of the trigger, whose input is connected to Exit feature memory block These functions.

The drawing shows a block diagram of the functional Converter.

The converter contains an analog-to-digital converter 1 connected by an input to bus 2 of the input argument x, and outputs to the inputs of memory block 3. The outputs of block 3 are connected to the inputs of the first and second registers 4 and 5 connected by bits to digital outputs. inputs, respectively, of the first and second digital-to-analog converters 6 and 7, connected by outputs to the inputs of the output adder 8. The analog input of the first converter 6 is connected to the output of a controlled inverter 9, connected by a control input with an output of sign digit 4. Analog inputs of converter 7 are connected to the reference voltage buses ± Uo Controllable switch 10 is connected by the first signal input to the output of the multiplying unit 11, the second signal input to the bus 2 input of the argument and to the inputs of the multiplying unit 11, the output to the signal input the inverter 9, and the control input with the trigger output 12. The trigger input 12 is connected to the output of the sign of the function of the memory block 3. The Converter operates as follows.

The argument X of the reproduced function at f (x) comes from bus 2 to the input of converter 1, which converts the input alternating voltage into the code of the approximation interval number. According to this code, the memory unit 3, equipped with adresa chains for reflection to the corresponding memory cells, at its outputs issues the codes of the approximation coefficients aj, bj and the code of the feature (O or 1) of the type of the reproduced function in the given j-OM approximation interval. Codes from the outputs of memory block 3 are received respectively on registers 4 and 5 and trigger input 12 (performed, for example, in the form of a D-flip-flop). With the help of register 4, the transmission coefficient of converter 6 is controlled, and

5, the coefficient code entered into register 4 controls the operation mode of the inverter 9, switching the latter either to inversion mode or to the mode of direct transmission of the voltage supplied to its signal input. Register 5, which is entered

The coefficient code controls the transfer coefficient of the converter 7. The output voltage of the converters B and 7 is summed on the adder 8, the voltage from the output of which goes to the output of the converter.

If the feature code of the function j O, then the trigger 12 is in the zero state, in which the output of the controlled switch 10 is supplied with voltage from its second signal input, i.e. from level 2. 0 in this case, the expression for the approximating function on j-OM approximation interval has the form

Y ± ajx ± bj. If 1, then trigger 12 passes

5 into one state and switches the switch 10, connected to its output the output of the copying unit 11 operating in the squaring mode of the input argument. The expression for the approximating function takes the form

30y ± aj X 2 ± bj

Thus, the proposed converter, in comparison with the known, increases the accuracy of the functional transformation, since depending on the features of the initial approximated function, each approximation interval uses the approximation (lymphatic or quadratic), which allows to obtain a smaller conversion error.

Claims (3)

1. USSR author's certificate number 696490, cl. G 06 G 7/26 1977. 2. USSR author's certificate number 367543, cl. G 06 G 7/26, 1971. 3. USSR author's certificate No. 698010, cl. G 06 G 7/26, 1978 (prototype).