RU2517717C1 - Method of determining violations and correcting violated codes of bit positions in "1 from 4" coding - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: in the encoding device of the transmitting side of a communication channel, information coming from a source in a binary code is converted to a "1 from 4" code with an active zero; the converted information is generated in form of an array of words which can be presented in the form of a table comprising P₁ rows (words), each comprising P₂ quaternary bits in a "1 from 4" code, wherein like quaternary bits in rows form P₂ vertical columns of the array; a check code, K_r and K_v, is formed for each word and vertical column, respectively, via successive summation of bits of the word and the vertical column without carry over; in the control device, the received information is checked for conformity with the "1 from 4" code and check codes K_r and K_v; upon detecting a fault, the information is corrected and transmitted to the end device.

EFFECT: high reliability of transmitted information and low power consumption.

3 dwg

Description

The invention relates to the field of computer engineering, communication technology and can be used to build computing and communication tools in systems with increased reliability of the processed data, low power consumption, which significantly affects the reduction of the overall dimensions of these systems.

A fault-tolerant device is known (RF patent No. 2297036, IPC G06F 11/08, filed May 20, 2004, published November 10, 2005), which transfers information encoded by the method of a two-dimensional iterative code, which allows correcting errors in one byte of information and detecting them in other bytes .

The disadvantages of the device (coding method) are the lack of reliability of the processed and transmitted information and the unevenness of electricity consumption during operation.

The closest in technical essence to the prototype of the claimed method is a method of transmitting discrete information (RF patent No. 2239941, IPC ⁷ H03M 13/00, claimed 19.12.2002, published 10.11.2004).

At the transmitting end of the communication line, the method provides for the availability of a source of information, an encoding device and a transmitter, and at the receiving end, a receiver, a control device (decoding device) and a terminal device.

Data from the information source enters the encoder, undergoes conversion in it - information bits are framed by control bits and the transmitter is transmitted to the receiving part of the channel.

The receiver receives encoded information and transmits it to a control device (decoding device).

The control device checks the received information for errors, if available, it corrects the received information (eliminates errors) and transfers it to the terminal device for further processing. Since the control device does not always provide error correction, in the absence of such a possibility, it fixes their presence.

It should be noted that the composition of the transmitted information (the number of units in a binary number) is not constant and the amount of electricity consumed, which directly depends on the composition of the processed information, should be calculated on its maximum value.

The prototype method has inherent disadvantages of the fail-safe device indicated above.

The task to which the proposed technical solution is directed is to increase the reliability of the transmitted information and reduce the level of electricity consumption.

This is achieved by the fact that in the known method of transmitting discrete information in the encoder of the transmitting side, the information received from the source in binary code is converted to code “1 of 4” with an active zero. An important feature of the “1 of 4” coding system is transparent communication with the binary system, which is implemented by simple schemes. For clarity, we present a table of correspondence of two bits of the binary system and the system “1 of 4” with an active zero:

No. p / p Binary system System “1 of 4” one 00 0001 2 01 0010 3 10 0100 four eleven 1000

The converted information is formed in the form of an array of words, which can be represented in the form of a table containing P ₁ rows (words), each of which contains P ₂ quaternary digits in the “1 of 4” code, while the same quadruple digits of rows form P ₂ vertical columns array.

For each word of the array, a control code (K _r ) is generated by sequentially summing the word digits without taking into account the transfer, and each vertical column in the same way generates a control code (K _in ) of the same vertical digits.

The array prepared (encoded) for transmission has the form (see FIG. 1) and in this format it arrives at the receiving part of the information transmission channel.

The control apparatus (the decoder) received from the receiver information is checked against the code "1 of 4" and a control code K _{ri Bi} and K, and when an error is detected in the received information is determined by the nature of the error and its location in the array.

The final operation in the control device is to eliminate the error - to restore the received information.

The proposed method for correcting violations of the codes of digits of a number when encoding “1 out of 4” is as follows.

The control scheme determines the four-digit discharge in which the “1 of 4” code violation occurred. The contents of this category are replaced by a zero code (0001) and the control code (K _ro and K _in ) obtained by sequential summation of the digits of the number without transfer is determined by the received auxiliary number.

Next, the difference between the control code of the source number and the control code of the auxiliary number is determined:

K _r -K _ro = ΔK _r or K _in -K _in = ΔK _in

To restore the violated quadruple code, the previously set zero code in this category is replaced by ΔK _r (ΔK _in ).

The above method will be called the “zero code method”.

After restoring the broken discharge code, the number is checked for the coincidence of the original control code with the corrected control code. When ΔK _r (ΔK _in ) = 0, the recovery operation is performed.

The considered method for correcting codes in one quadruple category of a word or column is also applicable for repeated errors in the quadruple category of words and columns. Suppose that the array of information received by the control device contains eight words and ten vertical columns and each word and column is accompanied by control codes (K _ri and K _bi ), as shown in the figure (see figure 2).

Each quaternary digit of the array (P _ij ) has its own number P, where i is the number of words in the array, aj is the column number. Let the shaded digits (for example, P ₂₂ , P ₂₄ , etc.) do not meet the requirements of the code “1 of 4”.

Correction of codes is carried out in the following sequence of steps:

1 - words with distortion in one quaternary digit are corrected (P ₄₇ , P ₇₂₎ ;

2 - corrected columns having a distortion in one quaternary discharge (P _25, P _59);

3 - the word is corrected, which, after stage 2, has a distortion in one quadruple discharge (P ₅₄ );

4 - the column is corrected, which, after stage 1, has a distortion in one quaternary digit (P ₂₂ );

5 - the word is corrected, which, after stage 4, has a distortion in one quaternary digit (P ₂₄ ).

Information recovery operation completed.

The considered method for correcting distortions in quadruple digits is also applicable in cases where the distortions do not violate the “1 of 4” encoding principle, that is, a double error occurs in the quadruple digit. For example, instead of code 010 or instead of code 0001, code 0100. A method of comparing the received control code with the generated one at reception will record their mismatch, however, a failed quadruple digit will not be known.

The use of the vertical columns of the array and their control codes defines this category. Indeed, each quadruple discharge of an array at the same time belongs to only one word and only to one vertical column of this array. And since the number of the vertical column is determined by the number of the four-digit category of the word, the number of the vertical column, in which the control codes did not match, will also determine the number of the failed four-digit category of the word.

The presented method for detecting violations and correcting violated codes of digits of a number when encoding “1 out of 4” significantly increases the reliability of the transmitted information.

It should be noted that the method detects all errors, but not all possible combinations of errors are subject to correction.

The proportion of uncorrectable errors is negligible. For example, for an array of ten words, each of which contains one hundred digits, the proportion of uncorrectable errors is approximately 6 × 10 ^-6 .

The option of impossibility to correct the error can be considered by example. If we assume the existence of an error in the discharge P ₂₉ , then it is impossible to recover information in the second and fifth words.

When creating computing and communication tools, the use of the “1 out of 4” coding method compared to the binary coding system with equal reliability of their functioning reduces the energy consumption level by 5–5.5 times.

The invention is illustrated in figures 1, 2, 3.

Figure 1 shows the structure of an array of words formed for transmission over a communication line.

Figure 2 presents, as an example, the structure of the array received in the receiving part of the communication line.

Figure 3 marked:

1 - source of information;

2 - encoding device;

3 - transmitter;

4 - receiver;

5 - control device;

6 - terminal device.

The proposed method for identifying and correcting violations during the exchange of information, the existing element base (microprocessors and controllers implemented on basic LSIs) make it possible to create a variety of computing and communication tools for building real-time control systems, one of the main characteristics of which is a high level of reliability of the processed information (correction of errors in the transmission of information).

Claims (1)

A method for detecting violations and correcting violated codes of digits of a number when encoding “1 of 4”, which provides information transfer via a communication channel, which on the transmitting side contains a source of information, an encoding device and a transmitter, and on the receiving side, a receiver, a control device, and a terminal device, this:
- on the transmitting side of the channel, data from the information source in the form of words in binary code is sent to the encoding device, where they are formatted and then sent by the transmitter over the communication line to the receiving part of the channel;
- on the receiving side of the channel, the data received by the receiver are sent to the monitoring device, where the received information is checked for errors, and if any are detected, they are eliminated (information is corrected) and then transmitted to the terminal device, characterized in that the information in the encoding device received from the source in binary code is converted to code “1 of 4” with an active zero, is formed as an array of words, which can be represented as an array of words, which can be represented in the form of blits containing P ₁ rows (words), each of which contains P ₂ four-digit bits in the “1 of 4” code, with the same four-digit bits of lines forming P ₂ vertical columns of the array, a control code is generated for each row and vertical column (K _r and K respectively _in) by sequentially summing quaternary discharges rows and vertical columns excluding transfer information received in the control device is checked against the code "1 of 4" and a control code K _{ri Bi} and K, and when it detects discrepancy ( Errors) is determined by its character and location in the array, then there is a correction (error elimination) information in a manner zero code, which is as follows: the contents of the faulty quaternary discharge is replaced with a zero value (0001) generated the verification code (K _ro and K _in) taking into account the replacement made, the difference is determined
K _r -K _ro = ΔK _r or K _in -K _in = ΔK _in
and the zero code set earlier in the faulty discharge is replaced by ΔK _r (ΔK _in ).

Images