ERROR CORRECTION WHEN TRANSMITTING INFORMATION BY A PARITY CHECK OF BINARY MATRIX COORDINATES

Abstract

A method is offered for detecting and correcting the errors of information transmitted over communication channels based on the use of well-known in the theory of communication "parity coordinate" of a binary matrix which is the content of the message being transmitted. The parity is commonly used in two coordinates: rows and columns of the matrix. In contrast to the well-established practice of considering only two coordinates, it is proposed to do the error correction on the basis of an extended set of coordinates of matrix elements, among which are rows, columns, as well as main and auxiliary diagonals. Troubleshooting is made by generating a set of possible addresses of erroneous elements and a subsequent analysis of this set to avoid false addresses (non-existent) errors. This takes into account the four coordinates of binary matrix elements, which allows a low cost and fast detection of single, double, and group errors. The effectiveness of the method becomes higher with increase in the ratio "number of columns / rows number" of a binary matrix.