RU2085033C1 - Method for correcting analog-to-digital conversion error - Google Patents

Abstract

FIELD: measurement technology; data-measuring systems. SUBSTANCE: reference code signal used as complement to first reference code signal in shaping second reference code signal is calculated in each iteration cycle; corrected input code signal is calculated using formula given in description of invention. EFFECT: enlarged functional capabilities, improved speed. 5 dwg

Description

 The invention relates to measuring equipment and can be used in various applications of information-measuring systems with high requirements for the accuracy of signal conversion.

 Known iterative methods for correcting errors in analog-to-digital conversion using piecewise linear approximation of sections of the converter characteristic in correction algorithms (1).

 Closest to the proposed invention is a correction method (a.s. N 1714808, 1992), in which the iterative process of finding a solution in a nonlinear problem is carried out by the Newton method.

 The disadvantage of the automatic transmission method of the prototype is that for certain characteristics of the converter, when nonlinearity is essential, the correction algorithm is not feasible. In cases where the converter characteristic is close to critical, the automatic transmission process becomes lengthy.

 This goal is achieved by the fact that in the proposed automatic transmission method, the code of the magnitude of the reference signal K, used as an additive to the code of the first reference signal when generating the code of the second reference signal, is not constant, as in the prototype, in magnitude, but variable and calculated in each cycle iterations.

 In FIG. 1a and 1b show the procedure for determining K for the convex and concave characteristics of the ADC. The implementation of the proposed automatic transmission method is illustrated on a measuring and computing complex (CPI), a block diagram of which is shown in FIG. 2.

 The CPI includes RAM 1, calculator 2, ADC 3, switch 4, DAC 5, and trunk 7.

 We show the AKP by the proposed method.

через коммутатор поступает на АЦП, и происходит цифровое измерение сигнала. Операцию аналого-цифрового преобразования входного сигнала обозначим через

Результат преобразования

через магистраль направляется в ОЗУ.From the input source 6 measured value

through the switch enters the ADC, and digital measurement of the signal. The operation of analog-to-digital conversion of the input signal is denoted by

Conversion result

through the highway goes to RAM.

, который принимается за первый нескорректированный код входного сигнала

Далее осуществляется последовательное цифроаналоговое и аналого-цифровое преобразование первого эталона, которое обозначим через F(x_нск). F_нск y₁. Код y₁ запоминается в ОЗУ.For the correction algorithm to work, it is necessary to form two reference signals. In the first correction cycle, the code is selected as the digital equivalent of the first reference signal

which is taken as the first uncorrected input code

Next, a sequential digital-to-analog and analog-to-digital conversion of the first standard is carried out, which we denote by F (x _nsc ). F _nsk y ₁ . Code y _{1 is} stored in RAM.

.In the prototype, to generate the code of the second reference signal, a constant reference signal K is used. In the proposed automatic transmission method, in each correction cycle, the signal K is calculated, for which chords are drawn through the ADC characteristic before crossing the line

.

,
здесь характеристика АЦП является выпуклой. Хорда проведена из точки x 0, y 0 до точки x x_нск, y y₁. К₁ находится на пересечении хорды с линией

(фиг. 1а).In the first correction cycle, the value of K ₁ is determined by the formulas:

,
here the characteristic of the ADC is convex. The chord is drawn from the point x 0, y 0 to the point xx _nsk , yy ₁ . K ₁ is at the intersection of the chord with the line

(Fig. 1a).

,
здесь характеристика АЦП является вогнутой. Хорда проведена из точки x x_м, y y_м до точки x x_m, y y₁. K₁ находится на пересечении хорды с линией

(фиг. 1б). Код второго эталонного сигнала обозначим через

.

,
here, the ADC characteristic is concave. The chord is drawn from the point xx _m , yy _m to the point xx _m , yy ₁ . K ₁ is at the intersection of the chord with the line

(Fig. 1b). The code of the second reference signal is denoted by

.

Запишем операцию последовательного цифроаналогового и аналого-цифрового преобразования второго эталонного сигнала:

Первый скорректированный код входного сигнала вычисляется по формуле

В последующих циклах итераций i 2.n код величины образцового сигнала К_i вычисляется по единой формуле:

Формулы для вычисления второго эталонного сигнала и скорректированного кода входного сигнала i-го цикла имеет вид:

Для i 2 К₂ вычисляется по формуле:

Хорда в этом случае проведена между точкой

и точкой x x_1ск, y y₂.

We record the operation of sequential digital-to-analog and analog-to-digital conversion of the second reference signal:

The first corrected input signal code is calculated by the formula

In subsequent iteration cycles i 2.n, the code of the magnitude of the reference signal K _{i is} calculated by a single formula:

The formulas for calculating the second reference signal and the adjusted input signal code of the i-th cycle are:

For i 2 K _{2 is} calculated by the formula:

The chord in this case is drawn between the point

and point xx _1sk , yy ₂ .

позволяет определить К₂.The intersection of the chord with the line

allows you to determine K ₂ .

Останов решения может произойти в двух случаях:

с выдачей скорректированного кода входного сигнала

с выдачей скорректированного кода входного сигнала x_nск.

Stopping a solution can occur in two cases:

with the issuance of a corrected input signal code

with the issuance of a corrected input signal code x _nsk .

The code x _is in the (i + 1) cycle is used as the first reference.

In FIG. 3 shows an example when the algorithm of the ACP of the prototype is unrealizable, because the value x _1sk is beyond the limits of the characteristic.

 FIG. 4 and 5 illustrate the performance of the ACP according to the prototype and the proposed method. The iterative process gives a solution for the prototype after 4 cycles, according to the proposed method after 2. Thus, the introduction of the calculation of K by the chord method allows to increase the speed of the automatic transmission.

Claims (1)

A method for correcting errors in analog-to-digital conversion, which consists in generating a code signal proportional to the input analog signal with its subsequent storage, performing n correction cycles, in the first of which the first reference code signal is formed, which is used as a stored code signal followed by digital-to-analog and analog - its digital conversion with storing the result, after which a second reference code signal is formed by adding an exemplary code signal to the first reference code signal with its subsequent digital-to-analog and analog-to-digital conversion with storing the result, the corrected code of the input analog signal is calculated from the code signals proportional to the input and two reference signals, it is stored and compared with the stored code signal proportional to the input signal, if the difference obtained does not exceed the predetermined value in advance; an output code signal equal to the corrected code signal is formed, otherwise suschestvlyayut following correction cycles, in which as a first reference code signal using the memorized in the previous cycle adjusted correction code signal, characterized in that the calculation of the corrected code signal carried by the formula

X $\genfrac{}{}{0ex}{}{\text{i}}{\text{(i-1) CK}}$ = X _{(i-1) CK} + K _i ;
in addition, for i = 1;

for i = 2.n;

stopping of correction cycles is performed when one of the criteria is fulfilled:

the solution X ' _{(n-1) ck} is issued;

the solution is X ' _{n with k} ;
where i is the numbering of the correction cycles;
X _to , Y _x code signals proportional to the maximum values of the input and output signals of the ADC;
X _{i with a} corrected input code;

result of digital measurement of the input signal;

results of analog-to-digital conversion of the first and second reference signals;
K _i code value of the reference signal;
Δ is the permissible error of correction.

Images