SU1100715A1 - Interpolating filter - Google Patents

Abstract

The interpolation filter, co-terminals, the first subtraction unit, the first input is the input of the ING filter, the analog-to-digital converter, the memory unit, and the first digital-to-analog converter, the output of which is connected to r ESEJr-JwTi iA. the first input of the functional converter, characterized in that, in order to improve filtration, the accumulating sukmator, the second digital-to-analog converter, the second subtraction unit, the amplitude analyzer, the memory register and the third digital-to-analog converter, connected in series between the output of the memory block and the second input of the functional converter, the output of which is connected to the second input of the second unit —the subtraction, with the first subtraction unit, the analog-digital converter, the memory block These and the first digital-to-analog converter are connected in series, and the second input (L of the first subtraction unit is connected to the output of the second digital-to-analog converter.

Description

SP

The invention relates to radio engineering and can be used for solving problems of smoothing and restoring discretization. bathroom signals.

. An interpolating filter is known that contains a serially connected digital register, a code to voltage converter, a summing amplifier and several integrators connected in series, with 10 between the input of the first integrator and the other input of the summing amplifier, the element is switched on, and the output is connected; each of the other integrators are connected in series; a United additional storage element and a summing amplifier, the output of the code converter in the voltage of the sub-, whose well as to the inputs of adding; 20 solid summing amplifiers Cl3. The disadvantages of the known device

 .acclusion7; with in the absence of the possibility of a smooth adjustment of the filter in. lane. -frequencies of the input signal and in limited functionality. The closest in technical essence to the invention is an in interpolation filter containing

serially connected first subtraction unit, the first input of which is a filter input, a latching unit, an analog-to-digital conversion, a tele, the first switch, a memory block, a second switch, the first digital-an-. ,

the yogic converter and the functional converter, the output of which is connected to the second input of the first. subtraction block C27,

The disadvantage of this interpolating filter is the low homogeneity, since in each clock cycle the INPUT f signal and the feedback voltage are compared in each block. Due to the delay introduced by the calculator and the analog-digital conversion-.45 .. -the telecom, the formation of the feedback signal is delayed by the time tj. Therefore, the comparison of the indicated voltages will be performed with the error.

d and the greater the ratio of s ® 50. TD - sampling period, the More

total error of the device. In addition, when rushing to zero, the conversion error also

tends to zero, and this is possible with. provided that t, therefore build

The design is only intended to operate at low sampling rates. : f And consequently, taking into account the unambiguous connection of the sampling frequency f with the upper frequency of the spectrum 50 of the input signal, only low-frequency (slowly changing v and ti}) input signals whose spectrum does not exceed one kilohertz can be smoothed. .,. ,,,.

The purpose of the invention is to improve the accuracy of filtration.

To do this, an interpolating filter containing the first subtraction unit, the first input of which is the input of the interpolating filter, an analog-to-digital converter, a memory block and the first digital-to-analog converter, the output of which is connected to the first input of the functional pre-generator, is added to the accumulator, the second digital analog, converter, second subtractor, amplitude analyzer, memory register and third D / A converter connected in series between the output of the memory block and v the primary input of the functional converter, the output of which is connected. About the second input of the second subtraction unit, with the third first subtraction unit, the analog-digital converter f the memory unit and the first digital-to-analog | Log converter connected in series, and the second input of the first digital-analog converter connected to the output of the second digital-to-analog converter.

FIG. 1 shows the structural electrical circuit of the proposed interpolating filter; in FIG. 2. The time diagrams illustrating his work, -.,. / -;::::,; The interpolating filter comprises: a first subtraction unit 1 in series, an analog-to-digital converter 2 (A / D converter), a memory block 3, a first digital-to-analog converter 4 (DAC); and a functional converter 5, successively connected accumulating adder 6. the second DAC 7, the SECOND block 8, having read the amplitude analyzer 9, the memory register 10 and the third DAC 11, inserted between the output of memory block 3 and the second input of the function converter 5, while the first input of the first readout unit 1 is input an interpolating filter, jia, which is supplied by the analog output of the second DAC 7 (signal: 0оc is connected to the second input of the first subtraction unit 1, the output of the functional converter 5 is connected to the second input of the second subtraction unit 8, the accumulator of the accumulator b and register Memory 10 is connected to the control bus, -.: - i Interpolation Filter works as follows.

Analog input signal 0 or a discdiased MeiA input signal. (.fig, 2) is fed to the input of in: terpolyruptsogo filter, t, e. on the input of the first block. 1 read The second input of the first subtraction unit 1 receives a step voltage Feedback Od, At time t (Fig. 2)} (c-0. Then the difference of the input voltages Iex-Uoc ix is converted by the ADC 2 into a code, N through time - t h is written to memory block 3 by the control signal at time i.,. Further, the signal is converted in parallel over two channels, in the first of which the code is again converted by a DAC to an analog signal at the input of a functional converter. 5 / (the output signal of the converter 5 may, in the 66th case, represent any function: piecewise linear, parabolic, sinusoidal, etc., in the considered example, the output signal of the functional converter 3 is a piecewise linear Function.) In the second channel, the code N is converted to a step voltage equal to the input signal ( () V. The time instant (Fig. 2) and the instant of time tj N reaching this value in time i2 ® (B this moment can be set to the next value of the input singnal if it is sampled, fig 2, d, d | The next difference in voltage is time ij, i.e. at time / s, the A / D converter 2 is converted to the N code and written to the memory block 3 instead of the N code. By this time (tj), the output linearly varying voltage HaeUg ,,, the functional transform (el 5 becomes equal. (i) J Yapr UBUX is fed to the second input of the second subtraction unit 8, if equality (1) is not fulfilled, i.e., then after the arrival of the control signal, the third analog digital-to-analog converter at the time t appears at the time t which is summed up in the Functional Converter 5 with the main signal, compensated by deciding Others of self-accumulation of capacitance G, so that at the next moment tj of equality of equality (1 is fulfilled., -.;; in the following measures the operation of the device is similar. Thus, the output signal of the interpolating filter is a smoothed input signal, and the smoothing function set by functional converter 5, and in the given description of the filter operation this function is piecewise linear. The first digital-analog converter 4 and functional converter 5 are made respectively in the form of a transducer The code consumer to current 4 and accumulative capacitance C. Such realization allows, firstly, to simplify the summation circuit of analog signals at the input of the functional converter 5, D summing two currents in the node, and, secondly, to increase the speed by eliminating the inertial element - an operational amplifier, which would be necessary in the case of a summation of voltages. Codes (N, Nj, etc.) from the output of memory block 3 are fed through an accumulating sukmator b to the input of the second DAC 7, at the output of which a step feedback voltage is formed and, c. This voltage simultaneously arrives at the second inputs of the first subtraction unit 1 and the second subtraction unit 8, the first input of which is supplied with the voltage Ugbiy from the output of the function converter 5. Thus, the voltage difference Ap Uop-Ug is formed at the output BToppio of the unit 8, Then, it is converted by an amplitude analyzer 9. into a code which, according to a control signal, is written into memory register 10. The third DAC 11 converts this code into a 3c correction current, which is added to the main current, in the event that equality (1) is not fulfilled, thereby compensating for the undercharge of the accumulating capacitance C. The output codes of the accumulating adder 6 are the digital equivalent of the input analog When working with output 2 (Fig. 1), the interpolating filter is also an analog-to-digital converter, which differs in its functionality from the prototype device. . The proposed device also compares favorably with the prototype in accuracy. Indeed, the delay of the signal in the first block 1 of the subtraction and analog-digital converter 2 leads to a distortion of the output signal of the interpolating filter and the appearance of an error (Fig. 2, d) the difference of the output signal of the filter p ° ® of the real (OUT and) Maximal error can be calculated --17 :. -, where di „is the maximum increment of the input signal / ia over the sampling period .J is the total delay time

blocks 1 and 2, which is reduced to zero, since the input filter in the interpolating filter, the voltage Ugy is not compared with the output voltage of the Functional Converter 5, but

step voltage U

Therefore, in this case.

Kg

Claims (1)

INTERPOLING FILTER WITH? · Holding the first subtraction block, the first input of which is the input of the Interpol and Hands-Free Filter, analog-digital - a level converter, a memory block, and the first: a digital-to-analog converter, the output of which is connected to the first input of the functional converter, with the exception that, in order to increase the filtering accuracy, an accumulating adder is introduced, a second digital-to-analog converter, a second subtraction unit, an amplitude analyzer, a memory register, and a third digital-to-analog converter, connected in series between the output of the memory block. and the second input of the functional converter, the output of which is connected to the second input of the second subtraction unit, wherein the first subtraction unit, the analog-to-digital converter, the memory unit and the first digital-to-analog converter are connected in series, and the second input of the 'First subtraction unit is connected' to the output of the second digital -analog converter. sizooirns **