RU2427884C1 - Device of data sets identification - Google Patents

Abstract

FIELD: information technologies. ^ SUBSTANCE: device of data sets identification, comprising a pulse generator, a synchroniser, four scratchpad memory units, two code converters, two buffer memory units and a unit of code comparison, additionally two invariant generators and a unit of comparison to threshold are connected with according links. The invariant generator comprises four code converters, two permanent memories, four coefficient calculators, a calculator of normalisation coefficient, four invariant calculators, a unit of invariant vector generation. ^ EFFECT: device, which realises the invariant-group logic of identification, makes it possible on the basis of invariants combination to identify data sets, providing calculation accuracy and high reliability of identification procedure. ^ 2 cl, 6 dwg

Description

The invention relates to computer technology and radar and can be used in multi-position goniometric systems.

, l={1, 2, …}.The prototype of the claimed device selected device identification data sets [1]. The disadvantage of the prototype is the limitation of its functionality, since the known device for identifying data sets does not allow the identification of data sets with a high degree of reliability for the case when the azimuth α and elevation angle β are close to the value

, l = {1, 2, ...}.

The objective of the invention is the creation of an identification device for data sets with a high degree of reliability of the identification procedure.

The task is achieved in that the data set identification device contains a pulse generator, a synchronizer, four superoperative memory blocks, two invariant shapers, two code converters, two buffer memory blocks, a code comparison unit, a threshold comparison unit, and the output of the pulse generator is connected to the input synchronizer, the first output of the synchronizer is connected to the input of the first, second, third, fourth superoperative memory blocks, the second output of the synchronizer is connected to the input of the first, second code generators, the third output of the synchronizer is connected to the third input of the first buffer memory unit, the fourth output of the synchronizer is connected to the third input of the second buffer memory unit, the fifth output of the synchronizer is connected to the third input of the code comparison unit, the sixth output of the synchronizer is connected to the input of the pulse generator, the output of the first superoperative the memory block is connected to the first input of the first shaper of invariants, the output of the second superoperative memory block is connected to the second input of the first shaper In this case, the output of the third superoperative memory block is connected to the first input of the second invariant shaper, the output of the fourth superoperative memory block is connected to the second input of the second invariant shaper, the output of the first invariant shaper is connected to the first input of the first buffer memory block, the output of the second invariant shaper is connected to the first input the second buffer memory block, the output of the first code converter is connected to the second input of the first buffer memory block, the output of the second the code converter is connected to the second input of the second buffer memory block, the output of the first buffer memory block is connected to the second input of the code comparison block, the output of the second buffer memory block is connected to the first input of the code comparison block, the output of the code comparison block is connected to the input of the comparison block with a threshold, the output which is the output of the data set identification device.

The novelty of the claimed device consists in the introduction of the first and second shaper of invariants and a comparison block with a threshold and the exclusion of the first and second functional converters, which allows to implement an invariant group data processing algorithm, free from the disadvantages of the data processing algorithm implemented in the prototype [1].

Figure 1 presents a structural diagram of a device for identifying data sets, figure 2 is a structural diagram of an invariant generator, figure 3 is a structural diagram of a code comparison unit, and figure 4 is a structural diagram of a code comparison device.

The data set identification device (Fig. 1) contains a pulse generator 1, a synchronizer 2, a first 3, a second 4, a third 5, a fourth 6 super-operative memory blocks, the first 7, the second 8 shapers of invariants, the first 9, the second 10 code converters, the first 11 , the second 12 buffer memory blocks, a code comparison unit 13, a comparison unit with a threshold of 14.

Shaper invariants (figure 2) contains the first 15, second 16, third 17, fourth 18 code converters, first 19, second 23 ROM, first 20, second 21, third 22, fourth 24 factor calculators, normalization factor calculator 25, first 26 , second 27, third 28, fourth 29 calculators of invariants, the block of forming the vector of invariants 30.

Block comparison codes (figure 3) contains a device for comparing codes 31 ₁₁ , ..., 31 _NN .

The code comparison device (Fig. 4) comprises a first 32, a second 35, a third 38 subtractors, a first 33, a second 36, a third 39 multipliers, a first 34, a second 37, a third 40 dividers, an adder 41.

In Fig.1, the output of the pulse generator 1 is connected to the input of the synchronizer 2, the first output of the synchronizer 2 is connected to the inputs of the first 3, second 4, third 5, fourth 6 super-operative memory blocks, the second output of the synchronizer 2 is connected to the inputs of the first 9, second 10 code converters , the third output of the synchronizer 2 is connected to the third input of the buffer memory unit 11, the fourth output of the synchronizer 2 is connected to the third input of the buffer memory unit 12, the fifth output of the synchronizer 2 is connected to the third input of the code comparison unit 13, sixth the first output of the synchronizer 2 is connected to the input of the pulse generator 1. The output of the superoperative memory unit 3 is connected to the first input of the invariant generator 7, the output of the superoperative memory unit 4 is connected to the second input of the invariant generator 7, the output of the superoperative memory unit 5 is connected to the first input of the invariant generator 8, the output of the superoperative memory block 6 is connected to the second input of the shaper of invariants 8. The output of the shaper of invariants 7 is connected to the first input of the buffer block of memory 11, the output is formed For invariants 8, it is connected to the first input of the buffer memory unit 12, the output of the code converter 9 is connected to the second input of the buffer memory unit 11, the output of the code converter 10 is connected to the second input of the buffer memory unit 12, the output of the buffer memory unit 11 is connected to the second input of the code comparison unit 13, the output of the buffer memory unit 12 is connected to the first input of the code comparison unit 13, the output of the code comparison unit 13 is connected to the input of the comparison unit with a threshold 14, the output of which is the output of the data set identification device .

In Fig.2, the inputs of the first 15 and third 17 code converters are connected to the first input of the invariant generator 7 (invariant generator 8), the inputs of the second 16 and fourth 18 code converters are connected to the second input of the invariant generator 7 (invariant generator 8). The output of the code converter 15 is connected to the third input of the coefficient calculator 20, with the third input of the coefficient calculator 22, the output of the code converter 16 is connected to the fourth input of the coefficient calculator 20, with the third input of the coefficient calculator 21, the output of the code converter 17 is connected to the fourth input of the coefficient calculator 21, with the fourth input of the coefficient calculator 22, the output of the code converter 18 is connected to the fifth input of the coefficient calculator 20, with the fifth input of the coefficient calculator 21, with the fifth input m of the coefficient calculator 22. The first output of the ROM 19 is connected to the first input of the coefficient calculator 21, with the first input of the coefficient calculator 22. The second output of the ROM 19 is connected to the second input of the coefficient calculator 20, with the second input of the coefficient calculator 22. The third output of the ROM 19 is connected to the first the input of the coefficient calculator 20, with the second input of the coefficient calculator 21. The output of the coefficient calculator 20 is connected to the first input of the coefficient calculator 24, with the third input of the normalization factor calculator 25, with the first the course of the invariant calculator 26. The output of the coefficient calculator 21 is connected to the second input of the coefficient calculator 24, to the second input of the normalization factor calculator 25, to the first input of the invariant calculator 27. The output of the coefficient calculator 22 is connected to the third input of the coefficient calculator 24, to the first input of the normalization factor calculator 25, with the first input of the invariant calculator 28. The output of the ROM 23 is connected to the fourth input of the coefficient calculator 24. The output of the coefficient calculator 24 is connected to the second input numerator of invariants 29. The output of the calculator of the normalization coefficient 25 is connected to the second input of the calculator of invariants 26, with the second input of the calculator of invariants 27, with the second input of the calculator of invariants 28, with the first input of the calculator of invariants 29. The output of the first 26, second 27, third 28, fourth 29 calculators of invariants are connected respectively to the first, second, third and fourth inputs of the block forming the vector of invariants 30, the output of which is the output of the shaper of invariants.

. На вторые входы устройств сравнения кодов 31₁₁, …, 31_NN поступают инварианты, вычисленные формирователем инвариантов 7, причем вектор инвариантов, соответствующий n-му пеленгу, поступает на второй вход устройств сравнения кодов 31_n1, …, 31_nN,

. Совокупность выходов устройств сравнения кодов 31₁₁, …, 31_NN образует выход блока сравнения кодов.In Fig. 3, the first inputs of the device for comparing codes 31 ₁₁ , ..., 31 _NN receive the invariants calculated by the generator of invariants 8, and the vector of invariants corresponding to the mth bearing is fed to the first input of the device for comparing codes 31 _1m , ..., 31 _Nm ,

. The second inputs of the code comparison devices 31 ₁₁ , ..., 31 _NN receive the invariants calculated by the invariant generator 7, and the vector of invariants corresponding to the nth bearing is fed to the second input of the code comparison devices 31 _n1 , ..., 31 _nN ,

. The set of outputs of the device for comparing codes 31 ₁₁ , ..., 31 _NN forms the output of the code comparison unit.

4, the output of the subtractor 32 is connected to the first and second inputs of the multiplier 33. The output of the multiplier 33 is connected to the input of the divider 34, the output of which is connected to the first input of the adder 41. The output of the subtractor 35 is connected to the first and second inputs of the multiplier 36. The output of the multiplier 36 is connected with the input of the divider 37, the output of which is connected to the second input of the adder 41. The output of the subtractor 38 is connected to the first and second inputs of the multiplier 39. The output of the multiplier 39 is connected to the input of the divider 40, the output of which is connected to the third input of the adder 41. The output the adder 41 is the output of the code comparison device.

The claimed device based on a set of invariants implements an invariant-group algorithm for identifying data sets, providing computational correctness and high reliability of the identification procedure.

(N - число целей в секторе обзора многопозиционной угломерной системы (МУС)). В декартовой системе координат XYZ положение П₁ и П₂ задается векторами

и

соответственно. С каждым П_i (i∈{1, 2}) связана местная система координат X_iY_iZ_i, по отношению к которой пеленги Ц_n задаются азимутом α_in и углом места β_in. Соответствующие оси базовой и местных систем координат будем считать коллинеарными. Геометрия задачи представлена на фиг.5.Suppose there are two direction finders (P ₁ and P ₂ ) and one target (C _n ),

(N is the number of targets in the review sector of the multi-position goniometric system (ICC)). In the Cartesian coordinate system XYZ, the position of P ₁ and P ₂ is determined by vectors

and

respectively. Each P _i (i∈ {1, 2}) is associated with a local coordinate system X _i Y _i Z _i , with respect to which bearings _{n n} are defined by the azimuth α _in and the elevation angle β _in . The corresponding axes of the base and local coordinate systems will be considered collinear. The geometry of the problem is presented in figure 5.

, характеризующий взаимную пространственную привязку П₁ и П₂, а также единичный вектор

указывающий направление на Ц_n по отношению к местной системе координат X_iY_iZ_i (см. фиг.6), где

,

.We introduce the vector

characterizing the mutual spatial reference P ₁ and P ₂ , as well as a unit vector

indicating the direction of Ts _n in relation to the local coordinate system X _i Y _i Z _i (see Fig.6), where

,

.

,

и

можно записать уравнение плоскостиKnowing

,

and

can write the equation of the plane

Where

We proceed to the normal equation of the plane

Where

, ,

и

можно рассматривать в качестве инвариантов, то есть функций, значения которых остаются неизменными при подстановке в них пеленгов Ц_n, полученных с различных точек визирования, лежащих на одной прямой П₁П₂. Однако следует помнить, что среди элементов

,

и

уравнения (3), (4) независимыми являются только два элемента (в силу условия нормировки

), поэтому максимальное число независимых инвариантов, порождаемых нормальным уравнением плоскости, не превышает трех.By analogy with [6], the quantities

, ,

and

can be considered as invariants, that is, functions whose values remain unchanged when substituting bearings C _n obtained from different points of sight lying on one straight line P ₁ P ₂ . However, it should be remembered that among the elements

,

and

equations (3), (4) only two elements are independent (due to the normalization condition

), therefore, the maximum number of independent invariants generated by the normal equation of the plane does not exceed three.

, компонентами которого могут быть любые элементы множества {

,

,

,

}, при этом L=3.In the future, to solve the problem of identification of bearings in the ICM, we will use the vector of invariants

whose components can be any elements of the set {

,

,

,

}, with L = 3.

The decision on the bearing belonging to the same C is made on the basis of the following decision rule (excluding direction finding errors):

- некоторая норма (например, евклидова).Where

- some norm (for example, Euclidean).

The main difficulties in identifying bearings arise in the presence of measurement errors and a sufficiently dense arrangement of C in the sector of the review of the ICC. Given the random nature of the measurement errors, the bearing identification algorithm reduces to the following rule:

, ε - порог отождествления (ε>0).Where

, ε is the identification threshold (ε> 0).

For the Euclidean norm, the weighted average square of the distance characterizing the measure of identification of bearings measured by points P ₁ and P ₂ is

- корреляционная матрица ошибок вычисления вектора инвариантов

, которую для простоты примем диагональной. Конкретизируя выражение (7) для вектора инвариантов, состоящего, например, из компонент

,

,

, получаем следующее выражение для евклидовой нормыWhere

- correlation matrix of errors in calculating the vector of invariants

, which, for simplicity, we accept diagonal. Concretizing expression (7) for an invariant vector consisting, for example, of components

,

,

, we obtain the following expression for the Euclidean norm

(где l={1, 2, …}), существенно ухудшается качество отождествления, и метод, развитый в [6], становится некорректным с вычислительной точки зрения для данных условий наблюдения.A comparative analysis of the developed method with the well-known approach to identifying data sets [6], based on the use of an invariant — the tangent of the angle of inclination of the plane passing through C and the base of the ICC, shows that at angles α _in and β _in close

(where l = {1, 2, ...}), the quality of identification deteriorates significantly, and the method developed in [6] becomes computationally incorrect for these observation conditions.

,

,

,

}, то получимIf we restrict ourselves to the considered case and the first invariant of the set {

,

,

,

} then we get

и

где l∈{1, 2, …}. Данный случай возможен, когда вектора

и

коллинеарны, при этом уравнение плоскости (3), (4) становится некорректным. Поскольку для практики такой случай является скорей исключением, чем правилом, то можно утверждать, что развитый метод является корректным для всех возможных случаев функционирования МУС.From the analysis of expression (9) it follows that the developed method is inoperative only in the case when 1-sin ² α _in cos ² β _in = 0, i = 1, 2, which is possible only if two conditions are satisfied simultaneously:

and

where l∈ {1, 2, ...}. This case is possible when the vectors

and

are collinear, while the equation of the plane (3), (4) becomes incorrect. Since in practice such a case is an exception rather than a rule, it can be argued that the developed method is correct for all possible cases of the functioning of the ICC.

The use of new invariants leads to an increase in the reliability of identification. In addition, the use of the vector invariant

where L = 3, also allows you to increase this characteristic of the ICC.

A device for identifying data sets works as follows.

In the initial state, memory blocks 3 and 4 contain codes whose values correspond to the azimuth α _1n and elevation angle β _1n measured at the first receiving point, respectively, and codes whose values correspond to the values measured at the second receiving point are stored in memory blocks 5 and 6 azimuth α _2n and elevation angle β _2n, respectively. The number of pairs (α _1n , β _1n ), (α _2n , β _2n ) is equal to the number of objects N in the field of view. The order in which these codes are entered into the blocks corresponds to the order in which they are received at the receiving points (the chains for loading codes into the blocks are not shown).

The information processing cycle is performed according to the signals from the synchronizer 2 in the following order.

According to the signal from the first output of the synchronizer 2 to the inputs of blocks 3-6, the codes corresponding to the azimuth and elevation angle are called from blocks 3-6 to the shapers of invariants 7 and 8, in which the value of the invariant is calculated for the first and second points, respectively .

The shaper of invariants 7 (8) works as follows.

, поступают на входы вычислителей коэффициентов 20-22 таким образом, чтобы обеспечить формирование коэффициентов A_in, B_in, C_in по правилу (2). С выходов блоков 20, 21, 22 коды, пропорциональные соответственно значениям коэффициентов A_in, B_in, C_in, поступают на первый, второй, третий входы блока 24 соответственно. Коды, пропорциональные значениям вектора

, подаются из ПЗУ 23 на четвертый вход блока 24. В блоке 24 по правилу (2) происходит формирование коэффициента D_in. Сформированные в блоках 20-22 коэффициенты A_in, B_in, C_in поступают также на третий, второй и первый входы блока 25 соответственно, где происходит вычисление коэффициента нормировки

. Коды, пропорциональные значениям коэффициентов A_in, B_in, C_in, D_in, поступают также на первые входы блоков 26, 27, 28 и четвертый вход блока 29 соответственно. Код, пропорциональный значению коэффициента нормировки, с выхода блока 25 поступает на вторые входы блоков 26, 27, 28 и на первый вход блока 29. В блоках 26-29 происходит вычисление инвариантов по правилу (4). С выходов блоков 26, 27, 28, 29 сформированные инварианты поступают соответственно на первый, второй, третий и четвертый входы блока формирования вектора инвариантов 30. В блоке 30 формирователя инвариантов 7 и в блоке 30 формирователя инвариантов 8 из четырех инвариантов

,

,

,

происходит формирование трехкомпонентного вектора инвариантов

Правило формирования трехкомпонентного вектора инвариантов может быть любым из трех возможных:

но обязательно одновременно одинаковым в обоих блоках 30 формирователя инвариантов 7 и формирователя инвариантов 8. Выход блока 30 образует выходную шину формирователя инвариантов, по которой передается код, пропорциональный значению вектора инвариантов

.Codes proportional to the azimuth α _in received at the first input of the invariant generator from the output of the superoperative memory block are sent to the inputs of the code converters 15, 17, and codes proportional to the elevation angle β _in received at the second input of the invariant generator from the output of the superoperative memory block are fed to the inputs of code converters 16, 18. Code converters 15, 17 implement the mathematical operation sinα and cosα, respectively, and code converters 16, 18 - the mathematical operation sinβ and cosβ, respectively . Codes received at the outputs of blocks 15-18, as well as codes written in ROM 19, are proportional to the values of the vector

, arrive at the inputs of calculators of coefficients 20-22 in such a way as to ensure the formation of the coefficients A _in , B _in , C _in according to rule (2). From the outputs of blocks 20, 21, 22, codes proportional to the values of the coefficients A _in , B _in , C _in , respectively, are received at the first, second, third inputs of block 24, respectively. Codes proportional to vector values

are supplied from ROM 23 to the fourth input of block 24. In block 24, by rule (2), the coefficient D _{in is formed} . The coefficients A _in , B _in , C _in generated in blocks 20-22 also go to the third, second and first inputs of block 25, respectively, where the normalization coefficient is calculated

. Codes proportional to the values of the coefficients A _in , B _in , C _in , D _in also arrive at the first inputs of blocks 26, 27, 28 and the fourth input of block 29, respectively. A code proportional to the value of the normalization coefficient is output from the output of block 25 to the second inputs of blocks 26, 27, 28 and to the first input of block 29. In blocks 26-29, the invariants are calculated according to rule (4). From the outputs of blocks 26, 27, 28, 29, the generated invariants arrive respectively at the first, second, third, and fourth inputs of the invariant vector generation unit 30. In the invariant generator unit 30 and the invariant generator unit 30 and 8 of the four invariants

,

,

,

the formation of a three-component vector of invariants

The rule for the formation of a three-component vector of invariants can be any of three possible ones:

but it must be simultaneously identical in both blocks 30 of the shaper of invariants 7 and the shaper of invariants 8. The output of block 30 forms the output bus of the shaper of invariants, along which a code is transmitted proportional to the value of the vector of invariants

.

из формирователей инвариантов 7 и 8 записываются соответственно в блоки 11 и 12 по адресам, поступающим по второму входу каждого из них от блоков 9, 10 соответственно. Запись в блоки 11, 12 происходит по тактовому импульсу, поступающему с третьего и четвертого выходов синхронизатора на третьи входы блоков 11 и 12 соответственно.The calculated values of the vector of invariants

from shapers of invariants 7 and 8 are written respectively in blocks 11 and 12 at the addresses arriving at the second input of each of them from blocks 9, 10, respectively. Writing to blocks 11, 12 occurs on a clock pulse from the third and fourth outputs of the synchronizer to the third inputs of blocks 11 and 12, respectively.

The internal structure and functioning order of blocks 9, 10, as well as blocks 11, 12 of the claimed device for identifying data sets are similar to the prototype. Information is written to the blocks sequentially, and reading is parallel. This cyclic process is repeated until the data arrays recorded in blocks 3-6 for all N objects are processed.

, соответствующего пеленгам на первом приемном пункте, записанное в блоке 11, сравнивается с каждым значением каждого вектора инвариантов

, соответствующего пеленгам на втором приемном пункте, записанным в блоке 12, в соответствующих устройствах сравнения кодов 31₁₁, …, 31_NN.Each value of each vector of invariants

corresponding to the bearings at the first receiving point, recorded in block 11, is compared with each value of each vector of invariants

corresponding to the bearings on the second receiving point recorded in block 12, in the respective devices for comparing codes 31 ₁₁ , ..., 31 _NN .

,

,

), работает следующим образом.A code comparison device that implements an Euclidean norm estimate (for an invariant vector consisting, for example, of components

,

,

), works as follows.

, сформированному в блоке 8, причем первая, вторая и третья компонента вектора инвариантов поступает на первые входы блоков 32, 35, 38 соответственно. На вторые входы первого 32, второго 35, третьего 38 вычитателей поступает код, пропорциональный трехкомпонентному вектору инвариантов

, сформированному в блоке 7, причем первая, вторая и третья компонента вектора инвариантов поступает на вторые входы блоков 32, 35, 38 соответственно. Таким образом, на выходе вычитателей 32, 35, 38 имеем коды, пропорциональные соответственно

,

,

. В блоках 33, 36, 39 происходит возведение в квадрат величин

,

,

соответственно. Коды, пропорциональные

,

,

, с выходов перемножителей 33, 36, 39 поступают на входы делителей 34, 37, 40 соответственно. В блоках 33, 36, 39 происходит деление соответственно на величины

,

,

, характеризующие ошибки вычисления вектора Δ

. Коды, пропорциональные величинам

,

,

, поступают с выходов блоков 34, 37, 40 на первый, второй, третий входы сумматора 41 соответственно, на выходе которого формируется код, пропорциональный величине меры отождествления пеленгов в соответствии с (8). С выхода сумматора 41, который является выходом блока сравнения кодов 13, сформированный код поступает на вход блока сравнения с порогом 14.The first inputs of the first 32, second 35, and third 38 subtractors receive a code proportional to the three-component vector of invariants

formed in block 8, the first, second and third components of the vector of invariants arriving at the first inputs of blocks 32, 35, 38, respectively. The second inputs of the first 32, second 35, and third 38 subtractors receive a code proportional to the three-component vector of invariants

formed in block 7, the first, second and third components of the vector of invariants arriving at the second inputs of blocks 32, 35, 38, respectively. Thus, at the output of the subtractors 32, 35, 38, we have codes proportional to, respectively

,

,

. In blocks 33, 36, 39, the squaring of values occurs

,

,

respectively. Proportional Codes

,

,

, from the outputs of the multipliers 33, 36, 39 go to the inputs of the dividers 34, 37, 40, respectively. In blocks 33, 36, 39 there is a division, respectively, into values

,

,

characterizing the errors in the calculation of the vector Δ

. Codes proportional to

,

,

come from the outputs of blocks 34, 37, 40 to the first, second, third inputs of the adder 41, respectively, at the output of which a code is generated proportional to the value of the identification measure of bearings in accordance with (8). From the output of the adder 41, which is the output of the code comparison unit 13, the generated code is input to the comparison unit with a threshold of 14.

или логический «0» при

), которые поступают на выход блока сравнения с порогом 14. Выход блока сравнения с порогом 14 является выходом устройства отождествления наборов данных.In block 14, for each pair of bearings, in accordance with decision rule (6), identification signs are formed (logical “1” at

or logical "0" when

) that go to the output of the comparison unit with threshold 14. The output of the comparison unit with threshold 14 is the output of the data set identification device.

INFORMATION SOURCES

1. USSR author's certificate No. 1654810, G06F, 7/06, 1991.

2. Chernyak B.C., Zaslavsky L.P., Osipov L.V. Multiposition radar stations and systems // Foreign Radio Electronics. 1987. No. 1. S.9.

3. Usachev V. A., Fedorov I. B. Identification of the flow of objects in the system of meters // Izv. universities. Radio Electronics 1980. No. 11. S.32.

4. Usachev V. A., Fedorov I. B. Identification of objects in an independent combined system of meters // Izv. universities. Radio Electronics 1982. No. 1. S.89.

5. Bulychev Yu.G., Manin A.P. Mathematical aspects of determining the movement of aircraft. M .: Engineering, 2000.

6. Bulychev Yu.G., Taran V.N. The invariant-group method for identifying target bearings in triangulating multi-position systems of passive location // RE. 1987.V. 32. Number 4. S.755.

7. Gilbo EP, Chelpanov IB Signal processing based on ordered selection. M .: Sov. radio, 1975.

Claims (2)

1. A data set identification device containing a pulse generator, a synchronizer, four superoperative memory blocks, two code converters, two buffer memory blocks, a code comparison unit, characterized in that two additional invariant shapers and a comparison unit with a threshold are additionally introduced into it, while the output of the pulse generator is connected to the input of the synchronizer, the first output of the synchronizer is connected to the inputs of the first, second, third, fourth superoperative memory blocks, the second output of the synchronizer is connected with the input of the first, second code converters, the third synchronizer output is connected to the third input of the first buffer memory block, the fourth synchronizer output is connected to the third input of the second memory buffer block, the fifth synchronizer output is connected to the third input of the code comparison block, the sixth synchronizer output is connected to the generator input pulses, the output of the first superoperative memory unit is connected to the first input of the first shaper of invariants, the output of the second superoperative memory unit is connected to w by the first input of the first shaper of invariants, the output of the third superoperative memory block is connected to the first input of the second shaper of invariants, the output of the fourth superoperative memory block is connected to the second input of the second shaper of invariants, the output of the first shaper of invariants is connected to the first input of the first buffer memory block, the output of the second shaper of invariants is connected with the first input of the second buffer memory block, the output of the first code converter is connected to the second input of the first buffer about the memory block, the output of the second code converter is connected to the second input of the second buffer memory block, the output of the first buffer memory block is connected to the second input of the code comparison block, the output of the second buffer memory block is connected to the first input of the code comparison block, the output of the code comparison block is connected to the input block comparison with the threshold, the output of which is the output of the device. 2. The device according to claim 1, characterized in that each invariant generator contains four code converters, two ROMs, four coefficient calculators, a normalization coefficient calculator, four invariant calculators, an invariant vector generator, the first input of the invariant generator is connected to the inputs of the first and third code converters, the second input of the invariant generator is connected to the inputs of the second and fourth code converters, the output of the first code converter is connected to the third inputs the first and third coefficient calculators, the output of the second code converter is connected to the fourth input of the first and third input of the second coefficient calculator, the output of the third code converter is connected to the fourth inputs of the second and third coefficient calculators, the output of the fourth code converter is connected to the fifth inputs of the first, second and third coefficient calculators; the first output of the first ROM is connected to the first inputs of the second and third coefficient calculators, the second output of the first ROM is connected to the second inputs of the first and third coefficient calculators, the third output of the first ROM is connected to the first input of the first and second input of the second coefficient calculator; the output of the second ROM is connected to the fourth input of the fourth coefficient calculator; the output of the first coefficient calculator is connected to the first input of the fourth coefficient calculator, to the third input of the normalization coefficient calculator and to the first input of the first invariant calculator; the output of the second coefficient calculator is connected to the second input of the fourth coefficient calculator, to the second input of the normalization coefficient calculator and to the first input of the second invariant calculator; the output of the third coefficient calculator is connected to the third input of the fourth coefficient calculator, to the first input of the normalization coefficient calculator and to the first input of the third invariant calculator; the output of the fourth coefficient calculator is connected to the second input of the fourth invariant calculator; the output of the normalization factor calculator is connected to the second inputs of the first, second, third invariant calculators and to the first input of the fourth invariant calculator; the outputs of the first, second, third and fourth calculators of invariants are connected, respectively, with the first, second, third and fourth inputs of the unit for generating the vector of invariants, the output of which forms the output of the invariant generator.

Images