SU604011A1 - Statistic analyzer - Google Patents

Description

one

The invention relates to specialized computing equipment intended for instrumental determination of the characteristics of random processes.

A known analyzer for estimating the dispersion function, comprising a channel selection unit, a centering unit, a multiplier, a cyclic storage device, an amplitude analyzer, and a calculator i. The closest to the technical standard proposed is an analyzer for estimating the dispersion function of random processes, containing a matching unit, and the second inputs of which are the inputs of the analyzer, and the first and second OUTPUTS are connected to the corresponding inputs of the correlometer, the output of which is connected to the input of the first multiplier tel, the first output of which is co.cinen with the first input of the second multiplier, the second input of which is connected to the output of the first quadr, the input of which is connected to the output of the former of the decomposition coefficients, the first input of which is connected to the second input of the analyzer, the output of the second

.2 multiplier through the adder is connected to the output of the analyzer and to the input of the second quad. 2.

However, these analyzers do not allow to determine the nature of the stochastic connection between processes.

The aim of the invention is to expand the class of tasks to be solved by defining stochastic connections of random processes.

The goal is achieved by introducing into the analyzer a subtraction unit, a divider, a measuring unit, a comparison unit, a voltage source, an AND element, a non-AND element and a display block, and the second output of the matching unit is connected to the second input of the decomposition coefficient generator, and the third exit, to; the first inputs of the subtractor and the divider respectively, the second input of which is connected to the output of the second quadr, and the output is connected to the first input of the first comparison unit and kb to the second input of the subtraction unit whose output is connected to the measurement unit, the second input of the first comparison unit is connected

to the voltage source and to the nepsoXfy input of the second comparison unit, the second input of which is connected to the output of the first multiplier, the output of the second comparison unit is connected to the first input of the element I, the second j input is connected to the output of the first comparison unit and the first input of the NAND element, output which is connected to the first block of a different nation, and the second input to the output of the Inc. element to the input of the second, g block of the indicadia.

The drawing shows a structural diagram of the analyzer.

The Ialioator contains a matching block.

I, correlometer 2, driver of decomposition coefficients 3, first multiplier

4, the first quad 5, the second UK 6, the adder 7, the second quad 8, the subtraction unit 9, the divisor 10, the measuring unit

II, comparison units 12 and 13, source of voltage 20, 14, element I 15, element NOT-AND 16, display units 17 and 18.

The analyzer works as follows.

The original processes Ct} n (tl enters, 15 fOT to the inputs of matching unit 1, in which the current smoothing of processes is relative to mathematical expectations and normalization of their standard deviations, after the fourth normalized processes go to the secondary correlometer 2 and fop a co-ordinator of decomposition coefficients 3. A correlometer determines the normal correlated function of the processes Pxu (T) 35

At the same time, according to the voltage of the implementation and the output signal of the matching unit, the decomposition coefficient shaper 3 produces a voltage proportional to the decomposition coefficient of the regression function 0 df df, which after squaring the first KBaapaTopONi 5 is fed to the input of the second multiplier 6.

Signals proportional to the value of h y (T. from the output of the correlometer 2 is output; from the input to the first multiplier 4, which forms the signals proportional to (T) (Tt 1, 2, ..., K) - These signals decrease The splitter enters the second input of the second multiplication 5 bodies 6. Each of the K channels of the 1st resident multiplies the multiplication of the voltages acting on its inputs. The resultant sipgals from the output of each channel, equal to (t), go to adder 7, a signal is formed at the output, the pro-ortotopal estimate of the dispersion function y (t), 1, and l (-WITH ).

After obtaining an estimate of the dispersion function from the output of the matching unit, the voltage is equal to the variance of the process -u. (T. I: o steps at the input of the divider 10, the second input of which is supplied with a voltage equal to the value of the square of the dispersion function (t). From the output of the divider A value equal to the correlation value of the Xy (YIS (T) / Oy) ratio is fed to one input of the subtraction unit 9, to the second input of which from the first multiplier 4 a voltage is applied that is numerically equal to the square of the normalized mutual correlation function /) xi. { .

From the output of the block, subtracting w 9 is, in fact, equal to cr) (r) (t) and characterizes the degree of nonlinearity of the stochastic connection, goes to the measurement unit 11. The condition xi | (C) O indicates that the communication between the processes is linear .

One time multiplier from the first multiplier 4

Yah (

voltage equal to

. enters

the first block is compared to 12, and from the healer 10 a voltage equal to xy is given to the second block compared with 13, where 01g is compared with the constant voltage received from the voltage source 14.

In the case of equality of voltages with a reference voltage at the output of the comparison units, long duration voltage pulses are generated, which are necessary to eliminate the effect of non-duplicating operation of the comparison devices.

If the voltage appeared only at the output of the second comparator unit 12, which is equivalent to fulfilling the conditions for supervising the 4x4 C) -lifo non-linear functional link, it is through the element 11E-H1 1G P1x; it starts by block 18, according to which it is determined that the link is non-linear and strictly functional. The degree of communication linearity is determined by the measurement unit 11.

If the voltage appears on the outputs of both comparison blocks, then the element 15 triggers, i.e., at T, (y (t) l and f xuCt) 0, the output Tkey. Locks the element 11E-P 1G, turns off block 18 and connects block 17, by which it is determined that the communication of the underlying processes being monitored is linear and strictly functional.

Thus, in the case of a linear stochastic connection between the processes being analyzed, the reading of the measuring unit 11 is zero, in the case of a linear functional connection, the block 17 is turned on and the instrument readings are also equal to zero. The degree of non-linearity of communication is determined by the indications of measuring unit 11, and in the case of non-linear functional communication, block 18 is turned on.