RU2649291C1 - Method of cost-effective representation and transmission of bipolar data and signals - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: invention relates to telemetry, communication technology and can be used in information transmission systems over digital communication channels. Method consists in the fact that, when transmitting positive data of a transmitted parameter or signal, as an equal to 0 value, accepting the selected number (+ΔN) when transmitting positive values of transmitted data, as well as the number (Sh – ΔN) when transmitting negative values of transmitted data, where Sh is the largest value of N-bit binary code, and to increase the noise immunity of transmitted data and signals having the internal redundancy properties, using the additional noise-proof, non-redundant and low redundant coding. On the receiving side, the separating the positive and negative subscales 0 levels are defined as the values corresponding to the received parameter or signal graphic fragments points of discontinuity. At reception "hard" and "soft" decoding algorithms are used.

EFFECT: increase in the information transmission reliability, and detecting of errors that occur during transmission.

4 cl, 3 dwg

Description

The invention relates to telemetry, communication technology, information processing systems and can be used in data and signal transmission systems over communication channels. Its use allows to increase the reliability of information transfer without introducing structural redundancy in transmitted messages, to detect errors occurring during transmission, both single and multiple, to increase the efficiency of using a limited bit grid of presentation and data processing, as well as information transfer speed with restrictions on channel bandwidth communication.

This is achieved due to the following features that appear when using the alleged invention:

- economical and more efficient use of the bit grid for the representation and processing of data represented by N-bit binary, with restrictions on the choice of N values, determined by: 1) the required indicators of the accuracy of the presentation of data, messages, measurement words and the reliability of their recovery when received in conditions interference; 2) the selected schemes for the presentation and generation of data, messages, measurement words in the cyclic structure of the formation of group digital signals (GVCs), assuming that the number of bits N must be constant (N = Const) to simplify the synchronization system of received data, the presentation of which can be both unipolar and bipolar;

- structural-algorithmic transformations (SAP), involving additional redundant or low redundant noise-resistant coding of data, messages and digital signals, which are carried out when generating information on the transmitting side (in this case, the SAP is called “direct” (PSAP)) and when it is received (implemented while SAP is called "reverse" (OSAP)). Moreover, for each type of direct structural-algorithmic transformation (PSAP), which can be considered taking into account existing approaches and as an economical non-redundant noise-resistant coding ([1], Semenyuk VV Economical coding of discrete information. - SPb .: SPbGITMO (TU ), 2001. - 115 p.), And as a peculiar operation of predistorting data, measurement words and signals. Economical coding of discrete information [1] involves the simultaneous use of data compression methods and their excessive noise-resistant coding. These two operations are present in the previously announced methods of transmitting information using SAP, resulting in a redundant and low redundant noise-resistant coding of information ([2], patent RU No. 2434301 C2 "Method for transmitting discrete information" from 11/20/2011, bull. No. 32 ; [3], patent RU No. 2586605 C2 "Method for transmitting information and a system for its implementation" published on 05/17/2016, bull. No. 16; [4], patent RU No. 2586833 C2 "Method for transmitting information and a system for its implementation "Published on 06/13/2016, bull. No. 17, [5] patent RU No. 2607639 C2" The method is determined range to an object with a radiation source of signals with different frequencies ", published on January 10, 2017, bull. No. 4; [6], patent RU No. 2609747 C2" Method for transmitting information ", published on 02.02.2017, bull. No. 4 )

One of their distinctive features is that PSAPs are produced over positional binary codes, and when receiving, they use two types of OSAP, one of which is conventionally called the “hard” algorithm, and the second is the “soft” decoding algorithm. The ability to detect and correct errors in the transmission of information is realized using the soft decoding algorithm. In this case, the greater the natural redundancy of the transmitted digital information that appears as a result of using V.A. Kotelnikov on the discretization of controlled analog processes, the higher the corrective ability when using the algorithm of "soft" decoding. The use of the universal algorithm of “hard” decoding allows one to restore the received information without deteriorating its reliability indices, regardless of its properties, including in the absence of a correlation between adjacent values (X _j and X _{j + 1} ) of transmitted messages and data, where j = 1 , 2,3, ... is a countable set.

New opportunities arise when using an unconventional representation of received and transmitted messages X by their residual images b _i obtained as a result of comparisons modulo m _i (mod m _i ):

X ≡ b _i (mod m _i ), (1)

The presented analytical representation is an abbreviated form of the description of the main theorem of arithmetic:

X = m _i l _i + b _i , (2)

where m _i is the divisor (module) by which it is necessary to divide the divisible number x, l _i is the partial quotient of division, b _i is the remainder.

It is known that binary coding of data has advantages over their decimal representation. This is due not only to the fact that binary coding, which assumes the presence of two stable states, conditionally defined as the symbol “0” and “1” of a binary code, can most easily be implemented in electronic systems. In binary coding, there is no need for additional representation and transmission of comparison modules m _i . If we take the value of the byte word X = <116> ₁₀ = m l + b (2 ^* ), then the result of its binary coding is X = < 0111 0100> ₂ . Here subscripts s (<> ₁₀ and <> ₂ ) define the number system - decimal and binary, respectively. It is easy to notice that the underlined meanings of the upper half-word are the partial quotient l = < 0111 > ₂ = < 7 > ₁₀ , and the lower half-word <0100> ₂ is the remainder b obtained from dividing X by the comparison module m = 2 ⁿ = 2 ⁴ = 16. Therefore, the data about the comparison module m = 16 (2 ^* ), which were originally present (2 ^* ), were excluded from the transmission.

In the initial premise of the theory of finite fields E. Galois is the simplest system of equations

X = m ₁ l ₁ + b ₁

X = m ₂ l ₂ + b ₂ , (2 ^** )

replaced by a comparison system:

X ≡ b ₁ (mod m ₁ )

X ≡ b ₂ (mod m ₂ ). (3)

But it is an even more concise form of representing the value of X when it is encoded using residual images, because in addition to comparison modules, for example, m ₁ = 2 ⁿ - 1 and m ₂ = 2 ⁿ +1, where 2n = N is the bit depth presentations of words and messages in binary code, the data on incomplete quotients l _{1 =} 7 and l _{2 =} 6 are not transmitted to the communication channel either. The theory of finite fields E. Galois claims that incomplete quotients absent during transmission l ₁ = 7 and l ₂ = 6 It can be restored based on the Chinese remainder Theorem, provided that a common divisor of m ₁ and m ₂ will only be 1: (m _1, m ₂₎ = 1.

Therefore, using the theory of E. Galois, one can come to the most concise form of representation of the transmitted information.

At present, data compression during their presentation and transmission forms the basis of many new information-measuring technologies that are used to resolve existing contradictions in the field of transmission of information via high-speed radio links (VRL). However, fundamentally new opportunities for their synthesis and implementation appear when using mathematical methods.

If, at the same time, it is assumed that the amount of transmitted data remains unchanged, then the effect of syntactic data compression is transformed into an increase in the minimum code distance between the transmitted values. As a result of this, it is possible to create an internal structure of data, messages and signals (S ^(internal) ). With this approach, the previously formed structure S, which does not have divisions into constituent elements, includes two components, conditionally called external (S ^(external) ) and internal (S ⁽ internal ⁾ ):

S → S ^{( ext} . ⁾ + S ^(ext . ⁾ . (4)

Such a view allows us to more accurately determine those internal reserves for improving the efficiency of information transfer systems that have not been used.

The first invention, which uses PSAP and OSAP, is a method of discrete transmission of information ([2]). It also, as a result of structural-algorithmic transformations (SAP) preceding the transmission of information, forms a sequence of measurement words or messages called “samples of primary signals”, which are converted into samples with a lower bit depth of representation of the initial values. As a result of this, a method of preliminary compression of the transmitted data at the syntactic level is implemented, since the number of binary code characters to be transmitted is half that of their original number. The generated samples with lower bit depth representations of the initial values are residual images b _i . The basis of the invention [2] is the replacement of the traditional positional representation of binary words N = 2n - bit measurement words X by their display by residual images b _i . According to the mathematical model, residual images b _i . get as a result of operations corresponding to the division of X in a certain way the selected comparison modules m _i . As a result of this, the requirements of the identity equality of the original message X and its image-remainder b _i are met, resulting in an operation equivalent to the arithmetic operation of dividing X by the comparison module m _i :

X ≡ b _i (mod m _i ). (5)

In the second method of transmitting information and the system for its implementation ([3]), based on the effect of reducing data redundancy when replacing the original message X with the value of its residual image b _i having a lower bit depth, they provide increased noise immunity of the transmitted data. The mathematical basis of this effect is the model presented in the form of a system of residual classes (RNS) (a system of comparisons defined by formula (2)):

X _j ≡ b _1j (mod m ₁ )

X _j ≡ b _2j (mod m ₂ ), (6)

where X _j - j-th dimension word (message);

m ₁ , m ₂ are comparison modules, if n is half the initial bit depth N = 2n (bit grid) of the representation of the original traditional measurement words, then the case of their optimal choice is represented by the values m ₁ = 2 ⁿ - 1, m ₂ = 2 ⁿ + one;

b _1j, b _2j - residual images of the measurement word (message) X _j obtained as a result of the division of X _j into comparison modules m ₁ and m ₂ , respectively.

So, in the case of the original byte representation of the measurement words (N = 2n = 8) m_one= 2^four - 1 = 15, and m₂= 2^four+ 1 = 17. If N = 2n = 10, which corresponds to the case of a 10-bit representation of the values of transmitted words or messages, then m_one= 2⁵ - 1 = 31, and m₂= 2⁵+ 1 = 33. The residual images generated by this approach for their unambiguous mapping can have an n - bit positional representation structure, which is half the number of bits N of the original message X_j. However, some exceptions to this rule are the results of encoding residual images obtained when comparing the modules m₂= 2ⁿ+1 With the proposed additional coding, new messages are received as a result of replacing the original values of X_j, j = 1,2,3, .... to message C_j, j = 1,2,3, ..., composed, for example, of the values of image residues

<b_1j (mod 15),b_2j (mod 17)>_s, s=2,10. (7)C_j

<b_1j(mod 15), b_2j(mod 17)>_ss = 2.10.  (7)

Subscripts s (<> ₁₀ and <> ₂ ), as noted earlier, define the number system - decimal and binary, respectively.

<b₁₁ (mod 15),b₂₁ (mod 17)>_s, s=2,10. В рассматриваемом случае это значение C₁=<10111110>₂=<190>₁₀. Первый подчеркнутый образ-остаток равен: b₁₁=<1011>₂=<11>₁₀, что соответствует результату деления значения 116 на модуль сравнения 15: 116=15 × 7+11. Второй четырехразрядный образ-остаток (n=4) b₂₁=<1110>₂=<14>₁₀, что соответствует результату деления значения 116 на модуль сравнения 17: 116=17 × 6+14.For example, if the first value of the measurement word is: X ₁ = <116> ₁₀ = <01110100> ₂ with eight-bit representation of the binary code (2n = 8), then with the comparison modules mod 15 and mod 17, the result of the additional noise-resistant coding will be the value C ₁

<b ₁₁ (mod 15), b ₂₁ (mod 17)> _s, s = 2.10. In the case under consideration, this value is C ₁ = < 1011 1110> ₂ = <190> ₁₀ . The first underlined image-residue is: b ₁₁ = < 1011 > ₂ = <11> ₁₀ , which corresponds to the result of dividing the value 116 by the comparison module 15: 116 = 15 × 7 + 11. The second four-bit image-remainder (n = 4) b ₂₁ = <1110> ₂ = <14> ₁₀ , which corresponds to the result of dividing the value 116 by the comparison module 17: 116 = 17 × 6 + 14.

Thus, with the traditional method of transmitting information to the value of X _j, its equivalent can be fully matched in the form: X _{j =} m ₁ × l ₁₊ b _1j or in a duplicate form: X _{j =} m ₂ × l ₂₊ b _2j , where m ₁ , m ₂ are the comparison modules, and l ₁ , l ₂ are the values of incomplete quotients resulting from division. Moreover, the duplication of the results of the restoration of X _j following from identity (4):

X _{j =} m ₁ × l ₁₊ b _{1j =} m ₂ × l ₂₊ b _2j (8)

forms the basis of additional control of the reliability of the received message X _j .

When data are presented by their residual images (6), only the values b_1jand b_2j, with information about m_one, m₂and about l_one, l₂ excluded from transmission. In this case, m_one, m₂can be considered as key data known to the user, and l_one, l₂, as the data that is restored upon receipt based on the received values of the residual images b_1jand b_2j.

However, the invention [3] has the disadvantage that the individual values of the residual image b _2j obtained by comparing modulo m _{2 =} 2 ^{n +} 1 cannot be unambiguously identified. Therefore, if, for example, for N = 2n = 8, additional redundancy is not introduced in the form of an additional 9 binary digit and is limited to n = 4 when representing the values of b _2i , then the code constructions <0> ₁₀ and <135> ₁₀ , and also <16> ₁₀ and <136> ₁₀ . To distinguish between them, it is necessary to introduce another additional symbol (the fifth symbol) to represent the remainder b _2j modulo 17 equal to and b _2j = <16> ₁₀ = <10000> ₂ . This is due to the fact that for the considered case of byte words (N = 2n = 8) with 4 binary digits allocated for non-redundant noise-resistant coding, b _{2j the} leading binary symbol in the code construction b _2j = <16> ₁₀ = <10000> ₂ , equal to "1".

This will lead to additional errors introduced. Although, in the end, this drawback will be covered by the achieved technical effect obtained when receiving by detecting and correcting errors in the transmission of information in the soft decoding mode, however, the potential for improving the noise immunity of data distorted by interference will not be achieved.

Therefore, the use of the invention [3] leads, in fact, to an additional economical low-redundant noise-resistant coding of word values in a given bit grid N + 1, with the addition of the symbol "bit parity control", which is currently used in many information transfer systems to control the reliability of received messages. When applying the invention [3], the symbol "bit parity control" is used for a new purpose - it is loaded with additional information that allows, in addition to checking the parity of bits, to eliminate the ambiguity of identification of two situations that appear in residual images b _2j obtained as a result of comparison modulo m ₂ = 2 ^{n +} 1. Given the existing additional character in many information transmission systems, it is proposed to use only the code construction b _2j for “bit parity control”. In this case, the leading binary symbol in the code construction b _2j = <16> ₁₀ = <10000> ₂ , equal to "1", is moved to the symbol "bit parity control" newly formed to control the reliability of reception b _2j . As a result, they get a 5-bit code construct b _2j = <16> ₁₀ = <00001> ₂ . Identification of the fact that the remainder b _2j = <16> _{10 has been} transmitted is carried out on the basis of a violation of the logic for generating the “bit parity check” symbol: four binary characters “0” precede the appearance of the “bit parity check” symbol equal to “1”. For other code constructs b _2j , it is believed that the “bit parity” rule is satisfied. Thus, it is possible to control the reliability of the received words or messages using the modified rule for the formation of the symbol "parity bit".

Also, one of the disadvantages of additional coding of information using the SAP algorithm associated with the presentation of the results of additional noise-resistant coding by two residual images (formulas 6 and 7):

<b_1j (mod 2ⁿ-1),b_2j (mod 2ⁿ+1)>_s, s=2,10C _j

<b _1j (mod 2 ⁿ -1), b _2j (mod 2 ⁿ +1)> _s, s = 2.10

consists in the fact that the code structure X _j consisting of only the characters “1” is not encoded, for example, X _j = <11111111> ₂ for the case of byte words (N = 2n = 8). This disadvantage is eliminated due to the fact that this code combination of binary characters is excluded from the additional encoding process. It is added to the encoded data C _j . In this case, the appearance during the “hard” decoding of a message in the form C _j = <11111111> ₂ implies its simple replacement with X _j = <11111111> ₂ .

As a result of such a procedure of low-redundant noise-resistant coding, a minimum code distance between adjacent samples is equal to d _{min =} 2 ⁿ +1.

Moreover, as the algorithm for “hard” decoding of other encoded values of C _j other than C _j = <11111111> ₂ , the algorithm of the constructive residual theorem is used ([5], patent RU No. 2607639 C2, Method for determining the distance to an object with a radiation source signals with different frequencies, published on January 10, 2017, bull. No. 4).

In addition, a known method ([4], patent RU No. 2586833 C2, published on 06/13/2016, bull. No. 17). Its use also leads to an additional economical non-redundant noise-resistant coding of the values of words or messages in a given bit grid N = 2n. In this case, the effect of an additional economical non-redundant noise-resistant coding of the values of words or messages is achieved by using the following algorithm:

C _i ≡ (X _i × m ₂ ) (mod m ₃ ), (9)

where m ₂ is the comparison module equal to the accepted minimum code distance, for example, d _{min =} 2 ⁿ , and m ₃ is the comparison module equal to 2 ^N.

So, for the case of eight-bit words N = 2n = 8 and m ₃ = 2 ⁸ = 256, the results of the additional non-redundant error-correcting coding C _i will also be byte words that differ from each other in at least one bit.

In the method [4], the algorithm of "hard" decoding has the following form:

, (10)X _{j =}

, (10)

where C _j ^* = C _{j +} ε _{j. are} the values of the jth result of telemetry encoded on the transmitting side, containing the error ε _j .

The method of transmitting information ([6]), selected as a prototype, is that on the transmitting side they collect signals from message sources, convert them to binary code, provide synchronization of the generated measurement words represented by N - bit binary code, and form of them, a compressed digital group signal to be transmitted over communication channels, and on the receiving side, the received sequence of transmitted binary code symbols is received, characterized in that on the transmitting side is a code the design of the generated measurement words is divided into components, which are rearranged to form a new message on the results of telemetry, with the same number of digits N as the original measurement words, but with a different value of the minimum code distance between adjacent values obtained as a result of the primary encoding the results of telemetry with a binary code, arrange them in a compressed digital group telemetry signal in a certain sequence with respect to synchronous signals In this case, including the sequence in which the original measurement words should be transmitted, the digitally compressed group telemetry signal thus formed is subjected to subsequent modulation and transmission, and on the receiving side, from the obtained sequence of transmitted binary code symbols, a restored sequence of encoded N - bit messages and carry out their parallel decoding using "hard" and "soft" decoders, when performing operations " “soft” decoding provides the detection and correction of transmission errors of telemetered parameter values based on the group properties of “equanimity” performed in the selected graphic fragments of the transmitted parameter or signal converted as a result of rearrangement of the components of the results of the primary encoding of the telemetry values on the transmitting side, while simultaneously in the mode "Hard" decoding the components are rearranged in the reverse order to obtain the original sequence bit bits of the initially formed measurement words, as a result of which they implement a universal algorithm for restoring the initial results of telemetry without error correction, including in the absence of correlation between the initial neighboring values of the controlled telemetry parameters, smooth and filter the received and restored data and with respect to to the calculated neighboring values of telemeasurements determine their differences, which are used as tolerances when choosing the most suitable values of the values measured by the telemetry sensors generated as a result of soft decoding operations, taking into account the realized values of the minimum code distance between adjacent values and the allowed positions of the non-redundant noise-tolerant code formed on the transmitting side, re-hard code the decoded data of the corrected televisions as a result of soft decoding operations, smoothed or filtered data during the first operation “Hard” decoding is compared with synchronous, coinciding in time, values obtained as a result of the second operation of “hard” decoding, the comparison results are used to evaluate the achieved technical effect in the form of assessments of increasing the reliability indicators of receiving telemetric information, as well as to compare and adjust the results smoothing or filtering the telemetry data obtained during the first "hard" decoding, with the results that are closest to them, but coincide t with the allowed positions of error-correcting code generated as a result of structural and algorithmic transformations or parameter values transmitted signal on the transmission side, whereby the implement advanced features that are used to control the reliability of the results of telemetry information and decision support.

The prototype method also differs according to claim 1 in that when performing soft decoding operations designed to detect and correct telemetry information transmission errors, gaps are found that define the boundaries of the graphic fragments of the telemetered parameters converted on the transmitting side using structural-algorithmic transformation algorithms telemetry data based on the rearrangement of the components of the original N - bit code constructions of telemetry data obtained as a result primary coding and representing the values of the samples of the controlled process at the time of the survey of the values of the telemetry parameters, determined in accordance with the sampling theorem of V.A. Kotelnikov, then using the signs of identification of gaps in the form of first-order differences between the subsequent and previous values of the converted transmitted parameter or signal are determined their absolute values, which fall within the interval (0.8 - 1) m _1, where m ₁ - a certain way the second selected mo ul comparisons equal to 2 ^N, where N - number of digits of a binary code used to represent words measurements taken with errors telemetry data converted on the transmitting side using algorithms permutation parts source words measurements, which belong to the dedicated graphics fragments controlled the transmitted parameter or signal is confirmed upon receipt, is divided into the first comparison module m ₁ equal to 2 ^k , where k is the number of bits of the code structure of the component of a different measurement word, called the younger one in accordance with its positional structure during the initial primary coding, as a result of which integer residues from division are found, a histogram of the distribution of their values is constructed and, as an invariant, manifested in the form of a group value of "equal adequacy", is selected in the generated statistical the sample consisting of residues, the most common value, while all other values of the residues that do not coincide with the value of the found invariant are used to detect the errors of transmission of the results of television measurements, which are corrected by substituting data instead, the reliability of which is confirmed by the fact that when divided by the second comparison module m _{2 they} give a residual value equal to the invariant found for the selected graphic fragment, those values are selected from the selected converted television measurements that belong to the closest in absolute value allowed positions, spaced from each other by an amount equal to lm _2, m _{2 =} d _min , l = 1,2,3, .., d _min is the minimum code distance e, under the condition that the differences between the received data and their nominal values determined by the allowed positions do not go beyond the tolerances that are determined based on the results of “hard” decoding of the received signals and their subsequent smoothing or filtering based on various smoothing or filtering methods .

The concept of “soft decoding” used in the patent [6] for inverse structural-algorithmic transformations (OSAP) is common to patents ([2 ... 4]), which do not have fundamental differences. However, it differs from a similar concept used, for example, in [7], Clark J., Jr., Kane J. Coding with error correction in digital communication systems: Trans. from English - M .: Radio and communications, 1987. - 392 s, given on p. In [7], as well as in other sources of information, “soft decoding” is understood in the narrow sense - in the form of two rules. One of them consists in choosing a codeword that minimizes the distance from the codeword to the received code sequence of the same bit depth. Such a decoder is called a maximum likelihood decoder. Another rule is to decode each symbol of the codeword with minimizing the average probability of symbol error. These two rules may precede the proposed approach to the problem of “soft decoding”. Moreover, the results obtained are used as preliminary. The proposed subsequent operation of “soft decoding” uses other mathematical methods for detecting and correcting errors in the transmission of information based on the use of constructive mathematical theory of finite fields ([8], Kukushkin S. S. Finite field theory and computer science: vol. 1 “Methods and algorithms, classical and non-traditional, based on the use of the constructive theorem on residues ", - M .: MO RF, 2003 - 284 p.)

The prototype method [6] also involves the use of additional non-redundant noise-resistant coding of message values and measurement words that are received at the output of digital sensors or analog-to-digital converters (ADCs). At the same time, the basis of the direct structural-algorithmic transformation (PSAP) algorithm is the division of measurement words and messages presented at the output of digital sensors or ADCs by an N-bit binary code into components (segments), which are then rearranged while maintaining the same word depth - measurements and messages. In this case, the number of the bit grid N allocated for the representation of messages and measurement words can be either even, for example, N = 2n (examples of such a representation were considered earlier), or odd, N = 2n + 1.

In the first case, when N = 2n is an even number, by the example of a basic PSAP, maximizing the number of corrected transmission errors of TMI with constant d _min ^(Per) = 2 ⁿ (if N = 2n = 8, then d _min ^(Per) = 2 ^{4 =} 16), there can be a division of the source measurement words and messages represented by an N-bit binary code into a high (a _2i ) and a low (a _1i ) halfword with preservation of the traditional positional system of their presentation with a binary code with half the bit capacity (n).

An encoding algorithm for the source measurement words and messages, focused on the above example,

X _j = << a _2j >₂;_{<a1j> 2> 2} , (11)

involves a rearrangement of the older (a _2j ) and younger (a _1j ) half-words:

C _j = << a _1j >₂;_{<a2j> 2> 2} , (12)

where C _j is the result of an additional redundant noise-resistant coding of measurement words and messages X _j .

However, in the general case, the results of dividing the original code structure of a measurement word or message can be not only a halfword, but also its other components (segments). For example, with odd N (N = 2n + 1) the following two basic divisions are possible:

1) the leading part (a _2j ), consisting of (n + 1) binary bits, and the younger part (a _1j ), represented by an n-bit traditional positional binary code;

2) the leading part (a _2j ), consisting of n binary digits, and the younger part (a _1j ), represented by (n + 1) - a bit traditional positional binary code.

Other options for dividing the original binary code structure into components (segments) are also possible.

The difference between the above examples of additional coding based on different bit depths of the segments will be manifested in the values of the minimum code distance d _min . So, for example, in the first case d _min ⁽¹⁾ = 2 ^{n +} 1, and in the second d _min ^{(2) =} 2 ⁿ .

From the point of view of the mathematical description of additional coding of information using the theory of finite fields, the junior half-word or junior segment (a _1j ) represents the remainder (b _3j ) modulo m ₃ = 2 ⁿ (a _{1j =} b _3j ):

X _j ≡ b _3j (mod m ₃ ), (13)

where the indices b₃and m₃equal to i = 3 are selected taking into account the original formula (2), in which the index i = 1 corresponds to the comparison module m_one= 2ⁿ - 1, and the index i = 2 - to the comparison module m₂= 2ⁿ+1

Rearrangement of half-words or other segments of the original code construction of measurement words or messages under conditions when the minimum code distance d _{min is} increased compared to traditional information transfer, and the bit grid for presenting the results of additional coding remains the same N-bit, leads to the following continuation of the mathematical formulation of the problem inventions - re-comparing the results of coding C _j modulo N.

C _j = << a _1j >₂;_{<a2j> 2> 2} (mod N). (fourteen)

Thus, the proposed method also used two comparison operations: the first, represented by formula (13), modulo m ₃ , and the second, represented by formula (14), modulo N.

The disadvantage of the prototype method is that it, like the other methods represented by the patents [2 ... 4], is not focused on the data of primary coding of words and signals, which can take both positive and negative values.

1 Analysis of promising areas for improving existing data transfer systems shows that rapidly growing streams of transmitted data are in conflict with the ability to provide the required reliability indicators for the information received. So, for example, when transmitting information from Earth remote sensing spacecraft (RS) in the Russian Federation, the required speed was 100 Mbit / s in 2010, now it has been increased to 600 Mbit / s, and by 2020 it should be at least 2 Gbit / s . Moreover, real reliability indicators in the form of bit distortion probability were determined by the following values: P _b ≤ 10 ^-5 (2010) and P _b ≤ 10 ^-3 (2016). At the same time, for existing redundant methods of error-correcting coding (Reed-Solomon, BCH, etc.), there is a limit to their effectiveness in the form of P _b ≤ 10 ^-2 . As the calculations show, the required speed of at least 2 Gbit / s can be achieved, however, the probability of bit distortion, according to the most optimistic estimates, cannot be less than P _b = 10 ^-2 - 10 ^-1 . In this situation, existing redundant methods of error-correcting coding may turn out to be ineffective and not justify the hopes placed on them ([9], Emelyanov GA, Shvartsman VO Transfer of discrete information. M: Radio and communication, 1982. 240 p. )

So, for example, in the existing practice of telemetry during flight tests of promising products today reliability indicators are characterized by the values of P _b = 10 ^-1 . At the same time, address code constructions are distorted in batch telemetry, as a result of which it is not possible to unambiguously determine the sources of information messages. The reliability of the information received was increased and continues to increase through the use of highly efficient antenna systems (BAC). However, they are expensive and therefore their fleet in the Russian Federation is rapidly declining. In addition, they have restrictions on the speed of movement and therefore they cannot be used when receiving information from some control objects, for example, flying at low altitudes with high speeds.

Therefore, new reserves are needed to improve the efficiency of information transfer systems. One of such reserves is the bit grid N of the representation of message values and measurement words. It represents the resource that can be used to increase the performance indicators of data transmission systems and telemetry results. The most common telemetry practice today uses a scale of data represented only by positive values. Moreover, the meanings of words or messages, which by their nature can take both positive and negative meanings, for example, vibration measurements during telemetry, speech and acoustic signals in data transmission systems, are displayed only by positive numbers when the position “0” is shifted half the scale of telemeasurements Ш (0,5Ш), the values of which are given by the selected discharge grid N. However, there is a revolution in the field of system engineering, as a result of which modern models of on-board equipment (BA) are built on ove FPGA, microcontrollers and microprocessors greatly expanded capabilities onboard data processing performed for various purposes. One of them is associated with the provision of the possibility of operational control of processes that allow avoiding the negative development of contingencies detected as a result of telemetry. The second involves the use of the most effective procedure for compressing the obtained data by replacing it with the results of the on-board processing performed. Such a need arises in the development of high-speed radio channels for transmitting information to reduce the amount of data transmitted.

Thus, one of the distinguishing features of a number of generated words or messages is that they are bipolar, represented by both positive and negative meanings. However, this representation involves the use of an additional binary character to represent positive (character "0") and negative (character "1") values.

One of the most common options for writing negative numbers in a binary data representation system can be obtained by subtracting a larger absolute number from a smaller one, for example, the number “1” from the number “0” in a binary number system.

As a result of this, it is obtained that the binary entry “... 111” represents the number “-1” in the decimal number system. In this case, the points facing the binary characters "... 111" mean the repetition of the leading character, standing on the left and occupying a senior position to the value N, where N is the bit depth of the measurement words. If we continue the count in the negative area, then the code combination “... 110” will be interpreted as representing the decimal number “-2”. The next value "... 101" is "-3", etc.

If, at the same time, the loan extending to the left is cut off if the number of binary characters is equal to the value of the bit grid N, then the negative number represented by the binary code cannot be distinguished from the positive. Therefore, an additional binary symbol “0” is introduced, which is called signed and written in the generated code structure on the left, in the case when the number is positive, and the symbol “1” if it is negative.

A distinctive feature of the invention lies in the fact that it does not violate the traditional system of presenting data only positive integers (used in the existing telemetric complexes) (representation scale Ш _Tr = 0 - (2 ²ⁿ -1)), and transmit negative and positive values of bipolar words or messages in binary code without a signed character. In this case, the separation of positive and negative values is carried out by forming in the positive area of the presentation of the telemetry data of two subscales, conventionally designated as Δ ⁺ and Δ ^- and shifted relative to each other by the set value ΔН, the reference point of which is considered to be lower from the value "0", and on ΔB, which is counted from above from the value of "+ W = 2 ^N ".

In the first case, the transmission of negative values of bipolar TMP (figure 1) is carried out in the subscale:

Δ ^- = Y ^- ₀ = Y ^- _max - Y ^- _min , (15)

and the transmission of positive values of bipolar TMP in the subscale:

Δ ⁺ = Y ⁺ ₀ = Y ⁺ _max - Y ⁺ _min . (16)

When receiving data of the positive subscale Δ⁺distinguish from the data of the negative subscale Δ^-based on the differences of the levels (+ ΔН), counted in the direct count from 0 to (+ ΔН) and the countdown from W = 2^N -1 by value (+ ΔН): (Ш - ΔН). As a result of this, the value (+ ΔН) when transmitting positive numbers and the value (Ш - ΔН) when transmitting negative numbers are taken as a value equal to 0. Graphic illustrations explaining the distinguishing features of the proposed presentation are shown in figure 1. At the same time, in Fig. 1 (A), in which, for clarity, the bipolar telemetry parameter (TMP) is presented in the form of a sinusoid taking values in the range from A_min= -400 to A_Max= + 400. For representation, we used the bit grid N of the data representation equal to 10 bits of the binary code N = 10. In this case, to increase the accuracy of television measurements, the limiting values of A_min and A_Max could be equal: A_min= -511, and A_Max= + 511. However, at the same time, there was a high probability of the real TMP values going beyond A_min and ... Ah_Max.Therefore, in real practice, this range is considered as potentially achievable, and real values are chosen smaller in absolute value, for example, equal to A_min= -400 and A_Max= + 400, while the potential capabilities of the discharge grid allocated for telemetry (N = 2n = 10) allow us to expand this range to values A_min= -511, and A_Max= + 511.

The essential characteristics of the invention are that the positive and negative values of the bipolar digital signal or telemetry parameter (TMP) represent the positive values of the data presentation scale Ш, the values of which are determined by the used bit grid N: Ш = 0 ... (2 ^N -1). Figure 1 (B) is an illustration of such a representation. In this case, the algorithm of direct structural-algorithmic transformations (PSAP) of the first stage is as follows.

Negative values of the bipolar signal or TMP, in the General case, summarize with the stand, which is a constant positive value And_Max (X_j ^-+ A_Max) with the formation of a subscale Δ^-corresponding to the negative values of the transmitted data (15). Its distinctive feature is, therefore, that the reference point of negative values is shifted relative to the upper limit value of the data presentation scale Ш = 2^N.-one by the amount

(-ΔН), the absolute value of which is found from the ratio:

ΔH = W - A _max . (17)

The positive values of the bipolar signal or TMP are summed with the stand, which is a constant positive value ΔН (X _{j +} + ΔН) with the formation of a subscale of positive numbers Δ ⁺ (16).

In this case, there is no need to introduce an additional signed binary symbol “0” to identify positive values and “1” to determine the negative values of the monitored bipolar parameter or signal. As a result of this, the resource for increasing the efficiency of information transmission systems in the form of a bit grid, limited by the number of bits of the binary code N used to represent the transmitted values, is used more economically. In addition, the estimated range of possible changes in the data values A _min = -400 and A _max = + 400 can be expanded at least twice, ideally, to the values: A _max ^(p) = Ш - ΔН = 1023 -100 = + 923 and A _min = -923.

The disadvantage of this representation, the illustration of which is shown in Fig. 1 (B), is the insufficiently high frequency of occurrence of sign 0, separating positive and negative values in the form of identifiable when undistorted in the form of horizontal lines related to the values 0 (ΔН for the positive subscale Δ ⁺ and (W - ΔН) for the negative subscale Δ ^- ) connecting the break points of the received controlled parameter or signal (Fig. 1 (B)).

The essential characteristics of the invention also lie in the fact that the monitored bipolar parameter or signal converted at the first stage of the SAP and displayed in the most commonly used positive data presentation scale without introducing an additional signed binary symbol is additionally encoded using noise-resistant non-redundant or low-red SAP, the algorithms of which are given in patents [2 ... 4]. As a result of the second stage of the SAP, providing increased noise immunity of information transmission, the initial parameter or signal (Fig. 1 (A)) takes the form shown in the illustration (Fig. 3), and the monitored parameter or signal converted in the first stage of the SAP (Fig. 1) (B)) is displayed in a graphical representation in the form shown in Fig.4. The fundamental difference between the parameter or signal shown in Fig. 4 from the initial display (Fig. 1 (B)) is that the signs of the value 0 separating the positive and negative values are more clearly expressed due to their repeated display during breaks arising from -for increasing the minimum code distance d _min .

If in Fig. 1 (B) at the same time interval of the TMP representation, only two break points are visible, which are perceived on the receiving side as 0, separating the positive and negative values, then in the final form after the second stage of the SAP, providing increased noise immunity (Fig. 4), there will be 78 such values. As a result of this, the value 0 when presenting and transmitting bipolar data is isolated with significantly greater stability under distortion by noise.

As a result of the considered two stages of SAP, the first, which preserves the previous accuracy of the digital TMP or signal, while reducing by one (sign) bit of the bit grid allocated for measuring and transmitting bipolar data, and the second, which protects information, including interference , ensure the achievement of a comprehensive positive effect, consisting in the fact that with limited resources at the same time not one but several indicators of the effectiveness of the data transmission system (SPD) increase.

A method for economical presentation and transmission of bipolar data and signals is proposed, in which for the transmission of bipolar data it is assumed to use an additional signed bit of the binary code, characterized in that the additional signed bit of the binary code is not entered, but when presenting monitored parameters and signals having an internal redundancy that manifests itself in that their neighboring values are connected by a correlation dependence, positive and negative values are displayed in the region of positive a series of numbers representing samples obtained in accordance with V.A. Kotelnikov on the sampling frequency of analog parameters and signals, while for identification on the receiving side, positive values of the monitored parameters and signals having a selected internal range of the expected change (0 - (+ A_Max)), where + A_Max- the estimated maximum positive value of the transmitted parameters and signals Y⁺represented by a digital code is formed by summing with the selected value of the stand ΔН, while the resulting sum of the stand ΔН and the specified range of the proposed change (+ A_Max) must not exceed the maximum value of the data presentation scale Ш for N - bit data presentation grid, limited by the number of bits of the binary code equal to N: (ΔН + А_Max) ≤ W = 2^N, and the negative values of the transmitted parameters and signals having a selected internal range of the proposed change (0 - (- A_Max)), where -A_Max- the minimum negative value of the transmitted parameters and signals Y^-represented by a digital code is formed by summing it with the selected value (+ A_Max), resulting in the transmission of negative data for a value equal to 0, take a number equal to (+ A_Max), spaced in a series of numbers from the value of W = 2^N-1 by the value of (-ΔН): (Ш - ΔН), or by the amount of (+ ΔН) in the case of a return count, and the value of "-A_Max"Represent N - bit binary code a structure made up of the same symbol "0" of a binary word or message to be transmitted, as a result of which, when transmitting positive data of the transmitted parameter or signal, the number (+ ΔН), most often repeated when the received parameter or signal is broken, is taken having the properties of internal redundancy, manifested in the form of a correlation between neighboring values, and when transmitting negative data of a controlled parameter or signal as 0, present the positive and negative subscale (Δ⁺and Δ^-) data presentation, take the value (W - ΔН).

The proposed method according to claim 1, is also characterized in that the obtained results of the structural-algorithmic transformations of the first stage are subjected to the second stage of the structural-algorithmic transformations based on additional noise-resistant non-redundant or low-redundant coding, which are used to increase the reliability of identification of levels 0 that separate the positive and negative subscales (Δ⁺and Δ^-) representations of bipolar data and signals.

In addition, the method according to claim 2, characterized in that as an additional noise-resistant non-redundant coding of data and signals obtained by economical presentation, they use the separation of the code structures of the generated bipolar measurement words or messages into components, which are rearranged with the formation of a new messages about transmitted data and signals with the same number of digits N as the original measurement words and messages, but with a different value of the minimum code distance between adjacent The data obtained as a result of the primary encoding of the transmitted data and signals with a binary code are arranged in a compressed digital group signal in a certain sequence with respect to synchronization signals, including the sequence in which the original measurement words or messages should be transmitted thus formed digital compressed group signal is subjected to subsequent modulation and transmission, and on the receiving side from the received sequence of transmitted symbols in the binary code, a reconstructed sequence of encoded N-bit measurement words or messages is generated and decoded in parallel using hard and soft decoders; when performing soft decoding operations, they detect and correct transmission errors of transmitted data values based on group the properties of "equanimity" performed in the selected graphic fragments of the transmitted parameter or signal converted as a result of rearrangement parts of the results of the primary encoding of data and signals on the transmitting side, while simultaneously in the “hard” decoding mode, the components are rearranged in the reverse order to obtain the original sequence of bits of the originally generated measurement words or messages, as a result of which a universal algorithm for restoring the original data and signals without error correction, including in the absence of correlation between the initial neighboring values of data and signals, they smooth or filter the received and reconstructed data and signals and, in relation to the calculated neighboring values of the data and signals, determine their differences, which are used as tolerances when choosing the most appropriate values of the data and signals generated as a result of soft decoding operations, taking into account the implemented values of the minimum code distance between adjacent values and the resulting allowed positions formed on the transmitting side of the loss-free noise-immunity Of the current code, repeated “hard” decoding of the data and signals corrected as a result of the operations of “soft” decoding is performed, the smoothed or filtered data during the first operation of “hard” decoding is compared with synchronous, coinciding in time values obtained as a result of the second operation of “hard” "Decoding, the comparison results are used to evaluate the achieved technical effect in the form of estimates of increasing the reliability of reception of transmitted information, as well as to compare and adjusting the values of the results of smoothing or filtering the data and signals obtained during the first "hard" decoding, with the results that are closest to them, but coincide with the allowed positions of the error-correcting code generated as a result of structural-algorithmic transformations of the values of the transmitted data or signals to the transmitting side, as a result of which the advanced features used to control the reliability of the received data and signals and information support are implemented Iya decisions.

The proposed method according to claim 3 is also characterized in that when performing soft decoding operations designed to detect and correct data and signal transmission errors, gaps are found that define the boundaries of the graphic fragments of the transmitted data and signals converted on the transmitting side using structural algorithms algorithmic transformation of data and signals based on the permutation of their original components of N - bit code structures obtained as a result of primary encoded I and representing the values of the samples of the controlled parameters and signals at the time points of the interrogation of the values of the transmitted data and signals, determined in accordance with the sampling theorem of V.A. Kotelnikov, then, using the signs of identifying gaps in the form of first-order differences between subsequent and previous values of the converted transmitted parameter or signal, determine their absolute values are those that fall within the interval (0.8 - 1) m _1, where m ₁ - a certain way the selected second module cf. Nia, equal to 2 ^N, where N - number of digits of a binary code used to represent words measurements and messages received with errors data and signals converted on the transmitting side using algorithms permutation parts source words measurements and messages whose belonging to selected graphic fragments of the transmitted parameter or signal is confirmed upon receipt, subjected to division into the first comparison module m ₁ equal to 2 ^k , where k is the number of bits of the code structure of the component of the original layer a va-measurement or message called the youngest in accordance with its positional structure during the initial primary coding, as a result of which integer residues from division are found, a histogram of the distribution of their values is constructed and, as an invariant, manifested in the form of a group value of “equal adequacy”, is selected in the generated statistical sample consisting of residues, the most common value, while all other values of residues that do not match the value of the found invariant are used for bnaruzheniya data errors or signal, which is corrected by substituting these data the reliability of which is confirmed by the fact that they, when divided by the second comparison unit m ₂ give a residue value equal invariant found for the selected graphic fragment selected among the selected transformed data the values that belong to the closest in absolute value allowed positions, spaced from each other by an amount equal to lm _2, m _{2 =} d _min , l = 1,2,3, .., d _min is the minimum code distance when and conditions that the differences between the received data and their nominal values determined by the allowed positions do not go beyond the tolerances that are determined based on the results of "hard" decoding of the received data and signals and their subsequent smoothing or filtering based on various smoothing or filtering methods.

When digital data is represented by their residual images [3], as well as other parts or code segments [4] resulting from the division of the initial data, words and messages, they are considered when they are taken as independent information elements of low resolution, which are unique bricks ”, from which new forms of representation of the received and processed information can be obtained.

Another component of the effectiveness of data transmission systems (SPD) is the accuracy and resolution of a digital parameter or signal, specified by the operating range (A _min ... A _max ) of the representation of numbers, which is limited by the length N of the bit grid. Existing practice indicates that if nonzero digits of A during measurements or operations during on-board data processing go beyond (A _min ... A _max ), then they are lost, which can lead to distortion of the results.

So, if the number A> A _max appears during the operation, then the grid format overflows (overflow error). This effect in telemetry is called “roll off” the values of the transmitted parameter or signal (TMP).

When the number A <A _min appears, then anti-overflow is observed, accompanied by the appearance of an error of loss of code value, in which, in particular, a nonzero number can turn into machine zero. In telemetry, this effect is manifested in the form of “zero” TMP values, while in reality they differ from zero. In this case, they speak of insufficient resolution of the obtained TMP due to the fact that its small changes are not visible even when there is no noise or distortion caused by interference.

Errors of representation and errors of operations on numbers of limited accuracy are determined not only by overflow and anti-overflow of the machine bit grid, but also by the methods of rounding the numbers that are realized during the operations and presenting their results with symmetric or asymmetric rounding. In this case, the additional low-redundant coding of information implemented on the basis of algorithm (2) is also useful because it implements data representation in a system of residual classes (RNS), the advantage of which is that there is no need to round off the results of calculations ([10], Akushsky I. Ya., Yuditsky DI Machine arithmetic in residual classes. - M.: Sov.radio, 1968 - 140 s .; [11] Torgashev VA System of residual classes and the reliability of a digital computer - M.: Sov.radio , 1973 - 120 p.).

Therefore, it is important not only to choose the right bit depth N of the machine grid to represent digital measurement results, perform operations on them, and round off intermediate and final results. It is also needed to provide the required indicators of accuracy and reliability of the measurement results.

Thus, the use of the alleged invention will provide the opportunity to achieve a comprehensive positive effect, which consists not only in providing cost-effective noise-resistant coding of transmitted data and signals, but also in the possibility of significantly simplifying and increasing the reliability of the processing of the received information.

Information sources

1. Semenyuk V.V. Economical coding of discrete information. - SPb .: SPbGITMO (TU), 2001 .-- 115 p.

2. Patent RU No. 2434301 of 11/20/2011, bull. Number 32.

3. Patent RU No. 2586605 C2, published 05.17.2016, bull. No. 16.

4. Patent RU No. 2586833 C2, published on 06/13/2016, bull. Number 17.

5. Patent RU No. 2607639 C2, Method for determining the distance to an object with a radiation source of signals with different frequencies, published on January 10, 2017, bull. Number 4.

6. Patent RU No. 2609747 C2, published 02.02.2017, bull. Number 4.

7. Clark, J., Jr., Kane, J. Coding with error correction in digital communication systems: Per. from English - M.: Radio and Communications, 1987. - 392 p.

8. Kukushkin SS. Finite field theory and computer science: v.1 “Methods and algorithms, classical and non-traditional, based on the use of the constructive remainder theorem”. - M .: MO RF, 2003 - 284 p.

9. Emelyanov G.A., Schwartzman V.O. Discrete information transmission. - M .: Radio and communications, 1982. 240 p.

10. Akushsky I.Ya., Yuditsky D.I. Machine arithmetic in residual classes. - M .: Owls. Radio, 1968 - 140 p.

11. Torgashev V.A. The system of residual classes and the reliability of the computer. - M .: Owls. Radio, 1973 - 120 s.

Claims (4)

1. A method for economical presentation and transmission of bipolar data and signals, in which for the transmission of bipolar data it is assumed to use an additional signed bit of the binary code, characterized in that the additional signed bit of the binary code is not entered, but when presenting controlled parameters and signals having internal redundancy, which manifests itself in the fact that their neighboring values are related by a correlation dependence, positive and negative values are displayed in the region of the positive series of numbers ate, which are samples obtained in accordance with the theorem of V.A. Kotelnikov on the sampling frequency of analog parameters and signals, while for identification on the receiving side, positive values of the controlled parameters and signals having a selected internal range of the expected change (0 - (+ A_Max)), where + A_Max- the estimated maximum positive value of the transmitted parameters and signals Y⁺represented by a digital code is formed by summing with the selected value of the stand ΔН, while the resulting sum of the stand ΔН and the specified range of the proposed change (+ A_Max) must not exceed the maximum value of the data presentation scale Ш for N - bit data presentation grid limited by the number of bits of the binary code equal to N: (∆Н + А_Max) ≤ W = 2^N, and the negative values of the transmitted parameters and signals having a selected internal range of the proposed change (0 - (- A_Max)), where -A_Max- the minimum negative value of the transmitted parameters and signals Y^-represented by a digital code is formed by summing it with the selected value (+ A_Max), resulting in the transmission of negative data for a value equal to 0, take a number equal to (+ A_Max), spaced in a series of numbers from the value of W = 2^N-1 by the value of (-∆Н): (Ш - ∆Н), or by the amount of (+ ∆Н) in the case of a countdown, and the value of “-A_Max"Represent N - bit binary code a structure made up of the same symbol “0” of a binary word or message to be transmitted; as a result, when transmitting positive data of the transmitted parameter or signal, the number (+ ∆Н), which is most often repeated when the received parameter is broken, is taken as 0 a signal with the properties of internal redundancy, manifested in the form of a correlation between neighboring values, and when transmitting negative data of a controlled parameter or signal as 0, positive and negative subscales (Δ⁺and ∆^-) data presentation, take the value (W - ∆Н). 2. The method according to p. 1, characterized in that the obtained results of the structural-algorithmic transformations of the first stage are subjected to the second stage of the structural-algorithmic transformations based on additional noise-resistant non-redundant or low-redundant coding, which are used to increase the reliability of identification of levels 0 that separate the positive and negative subscales (Δ⁺and ∆^-) representations of bipolar data and signals. 3. The method according to p. 2, characterized in that as an additional noise-resistant non-redundant coding of data obtained by economical presentation, they use the separation of the code structures of the generated bipolar measurement words or messages into components, which are rearranged with the formation of a new message about transmitted data and signals with the same number of digits N as the original measurement words and messages, but with a different value of the minimum code distance between adjacent values, were obtained as a result of the primary encoding of the transmitted data and signals with a binary code, they are arranged in a compressed digital group signal in a certain sequence with respect to synchronization signals, including in the sequence in which the original measurement words or messages would have to be transmitted Thus, the digital compressed group signal is subjected to subsequent modulation and transmission, and on the receiving side from the received sequence of transmitted binary code symbols f They form a reconstructed sequence of encoded N-bit measurement words or messages and decode them in parallel using “hard” and “soft” decoders. When performing “soft” decoding operations, they detect and correct transmission errors of transmitted data values based on the group properties of “equal-adequacy” "Performed in the selected graphic fragments of the transmitted parameter or signal converted as a result of the rearrangement of the components of the result the primary encoding of data and signals on the transmitting side, while simultaneously in the “hard” decoding mode, the components are rearranged in the reverse order to obtain the original sequence of bits of the originally generated measurement words or messages, as a result of which a universal algorithm for recovering the original data and signals without error correction, including in the absence of correlation between the initial neighboring values of data and signals, carry out smoothing To obtain or filter received and reconstructed data and signals, and in relation to the calculated neighboring values of data and signals, determine their differences, which are used as tolerances when choosing the most suitable values of data and signals generated as a result of soft decoding operations taking into account realized values minimum code distance between adjacent values and allowed as a result of this position formed on the transmitting side of the redundant noise-resistant code, producing repeated hard decoding of the data and signals corrected as a result of soft decoding operations is performed; smoothed or filtered data during the first hard decoding operation is compared with synchronous, coinciding in time values obtained as a result of the second hard decoding operation, the comparison results are used to evaluate the achieved technical effect in the form of estimates of increasing the reliability of reception of transmitted information, as well as for comparison and adjustments and the values of the results of smoothing or filtering the data and signals obtained during the first “hard” decoding, with the results that are closest to them, but coincide with the allowed positions of the error-correcting code generated as a result of structural-algorithmic transformations of the values of the transmitted data or signals on the transmitting side As a result, they implement the advanced features used to control the reliability of the received data and signals and information support for decision making. 4. The method according to claim 3, characterized in that when performing soft decoding operations for detecting and correcting data and signal transmission errors, gaps are found defining the boundaries of the graphic fragments of the transmitted data and signals converted on the transmitting side using structural algorithms -algorithmic transformation of data and signals based on the rearrangement of their original components of N - bit code structures obtained as a result of primary coding and representing which are the values of the samples of the controlled parameters and signals at the time points of the interrogation of the values of the transmitted data and signals, determined in accordance with the sampling theorem of V.A. Kotelnikov, then, using the signs of identification of gaps in the form of first-order differences between the subsequent and previous values of the converted transmitted parameter or signal, determine their absolute values that fall in the interval (0.8 - 1) m ₁ , where m ₁ is a certain way the selected second comparison module, equal to 2 ^N , where N is the number of bits of the binary code used to represent the measurement words and messages, data and signals received with errors converted on the transmitting side using algorithms for rearranging the components of the original measurement words and messages, whose belonging to the selected graphic fragments of the transmitted parameter or signal reception is confirmed when subjected to division by the first comparison module m _1, equal to 2 ^k, where k - number of bits of the code structure part of the original words and measuring and a message called the younger one in accordance with its positional structure during the initial primary coding, as a result of which integer residues from division are found, a histogram of the distribution of their values is constructed and, as an invariant, manifested as a group value of "equal adequacy", is selected in the generated statistical sample, consisting of residues, the most common value, while all other values of the residues that do not match the value of the found invariant are used to detect the error Data transmission side or signals which are corrected by substituting in their place the data, the reliability of which is confirmed by the fact that they, when divided by the second comparison unit m ₂ give a residue value equal invariant found for the selected graphic fragment selected among the selected transformed data the values that belong to the closest in absolute value permitted positions spaced apart by a value equal to lm _2, m _{2 = d min, l = 1,2,3,} .., d min - minimum distance, when the condition, Thu value difference of the received data from their nominal values, determined by the allowed positions, are within the tolerances that determine on the basis of a "hard" decoding of received data and signals and their subsequent smoothing or filtering based on various methods of smoothing or filtering.

Images