SU771651A1 - Random process generator - Google Patents

Description

one

The invention relates to the field of computer technology and can be used both for modeling random processes with specified statistical characteristics and for spectral analysis of continuously time-varying signals.

A random generator is known, comprising a pulse generator, a random pulse sensor, AND elements, a control unit, a counter. To simplify the setting of the required law, the generator contains an encoder and a register, but its functionality is limited to I 1,

Another known random generator contains a block of elements And whose outputs are connected to the inputs of the memory block, the first inputs are connected to the outputs of a random number sensor, and the second inputs are connected to the first outputs of the control unit, the second output of the control unit is connected to the first input of the readout unit, the second the input of which is connected to the output of the summation unit, and the output to the output of the generator. However, this generator does not allow

random processes with specified spectral and neutral characteristics and cannot be used for spectral analysis of signals 2,

five

It is also known a device for calculating the decomposition coefficients of a function in series, comprising a cumulative block, a pulse generator of n keys, n blocks of voltage dividers (n outputs each), n channels from a series-connected adder, an analog-digital converter and a recording unit with the corresponding links between blocks and ysjjaMH 3J. The disadvantages of the known device include a large structural complexity, due to the presence of n keys and n blocks of voltage dividers for g outputs each, as well as the presence of a cumulative unit containing a large number of integrators; slow response due to

25 that the actual values of the spectral coefficients can only be obtained after the end of the integration time (i.e., after the end of the time interval in which

30 signal analysis is performed); limitations of functionality 3.

The closest technical solution to the invention is a random process generator comprising a series-connected random number sensor, an AND block, a pp block, a switch, a sign assignment block, a summation block, and a read block. In addition, it contains a Walsh function generator, the outputs of which are connected to other inputs of the switch, and a control unit 4,

However, this generator cannot be used to form spectral coefficients from the analyzed signal.

The aim of the invention is to further expand the functionality of the generator by forming spectral coefficients over the analyzed signal and then forming a random process with the same spectral characteristics as the analyzed signal.

The goal is achieved by the fact that a random process generator containing an adder, serially connected to the first memory block, and serially connected random number sensor, a second memory block, a register and a switch, to the other inputs of which is connected with its information outputs. 1 and a discrete orthogonal generator. functions, and the output of the first memory block is the output of the generator, contains an integration block, a block of OR elements, a block of adders, two shift ring registers, a third memory block and even the latent memory block, whose input is connected to the switch output, and the output to the input of the first shift ring register, the output of which is connected to the control input of the adder, whose information input is combined with the information inputs of the block of adders and connected to the output of the third memory block, information input which is connected to the output of the second shift ring register, all but the first, the inputs of which are connected to the corresponding outputs of the block of adders, and the first input of the second shift ring th register connected to the output of OR block elements, the first input of which is connected to the first output of the adder unit, control inputs of which are connected to respective outputs of digital orthogonal function generator and the second input unit OR- elements connected to the output integrating unit input

which is connected to the output of the random number sensor.

In addition, the integration unit contains a NOT element, two comparators, a D / A converter, an OR element, an AND element, a delay element, a reversible counter, a memory register and a character selection element whose input is combined with the first input of the first comparator and is NOT connected to the first through the element the input of the second comparator is the first input of the block, the second input of which is connected to the input of a digital-to-analog converter whose output is connected to the second inputs of the comparators whose outputs are connected to the inputs of nta OR, which is connected to the first input of the AND element, the output of which is connected to the counting input of the reversible counter, the input of which is set to zero through the delay element is connected to the control input of the memory element, whose information input is connected to the output of the reversible counter, the control input of which connected to the output of the mark selection element, in which the third and fourth inputs of the block are connected respectively to the second input of the AND element and to the control input of the memory element whose output is output m block.

FIG. 1 shows a block diagram of a random process generator; in fig. 2 is a block diagram of an integration unit.

The generator contains an integration unit 1, a sensor 2. Random numbers memory block 3, register 4, generator 5 discrete orthogonal functions (GDOF), switch b, memory block 7 shift ring register 8, block 9 adders, block 10 elements OR shift ring register 11, block 12 of memory, adders 13 and block 14 of memory. The control block is not shown in the drawing; only its outputs are indicated: a, b, c, d, e, r, ic, and IR, which are connected to the control inputs of the corresponding nodes and blocks,

The integration unit 1 contains a character extraction element 15, a HE element 16, comparators 17 and 18, a digital-to-analog converter 19, an OR element 20, an AND element 21, a reversible counter 22, a delay element 23, a memory register 24.

The generator is mainly intended for generating random processes by spectral characteristics (spectral density function) that coincide with the random process being analyzed (but the realizations of the generated random processes in general do not coincide with the random process being analyzed). In particular cases, the universal generator can be used in as a spectral analyzer only or as a generator of random processes based on previously calculated spectral coefficients (noMemaeNtJx in shift ring register 11), Therefore, in the following presentation, both modes of operation (Analysis and Generation) will be considered sequentially.

The operation of the proposed device in the Analysis mode is as follows.

The analog signal under study is fed to the input of the integration unit 1, from which time SA. a code is removed that is proportional to the value of an electron-to-one integral 3, the value of this code through a block of 10 elements OR is entered into register 11 with a shift, t, e, in such a way that through n conversion steps after the start of the analysis (after T) n ji-i ) the information inputs of the memory 12 are connected to the outputs of that memory cell in which code 3 is stored,. Starting with this mapping, the generator of 5 discrete orthogonal functions is turned on, operating with a maximum frequency for a time much smaller, the calculation of spectral coefficients is performed. If the proposed device is used only as an analyzer with continuous updating of spectral coefficients, then simultaneously with their calculation in the integration unit, the next elementary integral is computed. The coefficients of the spectral coefficients are formed in the corresponding adders of the adder unit 9. To ensure a predetermined operating mode, the control unit generates the following pulse series.

Series a is the maximum reference (from the point of view of the operability of the element base) the frequency supplied to the sensor 2 random numbers. With this frequency, random numbers appear at the output of the RNG 2.

Seri O is the same as Seri O, but delayed by a time interval (the time of transient processes in elements 16, 17, 18, 19 and 20 of the integral converter 1),

Seri c - pulses with frequency I / At and delayed with respect to corresponding pulses from series b on cTfo (time of transient processes in element 21 and in the reversing counter 22). With this frequency the information C is written from the integral converter of information with writing to register 11 and resetting cyMMaijopoa in block 9,

Series 6- is the same as series, but turned on after the next pulse of series C (through time At) and switched on after the calculation of the spectral coefficients has been completed (which signals the pulse from the output of the generator 5 to the control unit that appears when zeroing). generator counters 5, i.e., after the completion of the generation of all orthogonally func- tion functions). The serial e controls the operation of blocks 5 and 14. Due to the fact that

Analysis

what's in mode

exits

the devices are the outputs of the adders of block 9, then disconnecting the series from block 14 is not necessary (which simplifies the control block), - Seri d - information transfer pulses from block 9 to register 11 and

Analysis

transition from mode

to

to the Generation mode (the first of the pulses comes no earlier than after T-time d1 after the start of the analysis),

Seri f is the same as seri 6, but with a delay of (tj, (transient time in blocks 9 and 11),

5 Seri -vi is the same as the b series. After the impulse from the d series is received from the control unit, the device enters Generation mode. The operation of the device in this mode is based on an example of using Walsh as basic functions. The generated m -discharge random number is transmitted to-block 4 by -block 3 and stored

5 in it.

The values of the Walsh functions are for-. using the generator 5 Walsh functions, and changing the values of the Walsh functions by (n + yy) in accordance with

0 characters of random units — in switch 6; In the device, the following rule is accepted between the values of the Walsh functions and their binary equivalent: the value of VQ (t} +1 corresponds to the state of the ci-th

5 GDOF outputs, and the value of Vct (), -17 sro is a single state, | П + т) - the input switch b includes m groups of modulo-summation blocks, each

0 К-а group contains modulo-two summation blocks. To the first inputs of the modulo-sum block of the two ith groups, only the outputs of those Walsh functions are connected |

5 which correspond to the spectral coefficients combined into the K-th group in accordance with the rule of combining the coefficients into the same type groups. Since the change

If 0 characters are randomly produced simultaneously for all Walsh functions combined into a group, then the second outputs of the modulo modules are two moduli of each of the Kth groups combined and connected

to the output of the corresponding bit of block 4.

Thus, each output (each bit) of switch 6 is a product) by the value of the corresponding Sab function or 5 ° at the current time point I. To obtain the value of a random process, (1) at a given current time, r, it is necessary to add or subtract the value of the corresponding spectral coefficient {which are rewritten to register 11 with a pulse of d series). To implement this algorithm, the switch's contents are rewritten at every current discrete time. to register 8 and then, to advance the contents of the shift ring registers (to 8 — posterior, to 11 according to the words); summation is performed using adder 13 — subtraction of spectral coefficients from the register 11 (in accordance with the value of the next bit of the register 8). After summing up - subtracting all spectral coefficients, the value of the random process at the time instant is output to the generator output through block 14. At the next step, the discrete time {v + 1), the state of generator 5 changes, the content of register B and the formation process: the next value (1 + 1) of the random process is repeated,

To ensure the Generation mode of operation by the control unit, the following pulse series is generated: serie d - the same as in the Analysis mode, the t series and C are absent, serie d - transmission pulses (without erasing) information from block 9 to register 11 and transmitting random numbers in register 4. All other pulse series in E generation mode are included only after the passage of the d-pulse

The e series is the same as the series with in the analysis, i.e. with a frequency of 1 / db, but with a delay relative to the impulses of the series a for a while (i.e., without the first impulse, since GDOP 5 is already in the initial state), Seri p is the same / as the ser. in the analysis, but with a delay cft relative to the pulses of the di series (for the transient time in register 4 and in switch 6),

The k and 6 series are the same as the series, but the next impulse of the p series with the skip of the first impulse is switched on and the trigger is turned off.

Seri f is the same as seri k. with the first impulse and with the delay (, (time of transient processes in the slingsky ring register 11),

To reproduce each new implementation of a random process on the interval T with the same spectral characteristics by a pulse of series d, the calculated (or established) spectral coefficients are transferred from block 9 to register 11 and the next random number is transmitted from sensor 2 random numbers to register 4, then the operation of the universal random process generator proceeds in the same way as described above.

If it is necessary to form a random process with new characteristics, it is enough to either add the contents of block 9 of adders or switch the device to the operation mode. Analyze and input the test signal.

Thus, the proposed universal generator allows one to perform all the functions of a known device and, moreover, makes it possible to implement a random process similar to the signal under investigation without first determining and approximating the statistical characteristics of the signal under study. In addition, the proposed universal generator can be used as a spectrum analyzer.

Claims (4)

1. USSR author's certificate number 344431, cl. G 06 F 1/02, 1970. 2. Authors certificate of the USSR 370717, cl. And 03 K 13/02, 1970. 3. USSR author's certificate number 470812, cl. G 06 F 15/34, 1975. 4. USSR author's certificate (532873, class G 07 C 15/02, G 06 F 1/02, 1977 (prototype).