MULTI-CRITERION METHOD FOR INCREASING THE RELIABILITY OF TRANSMISSION OF DIGITAL INFORMATION SIGNALS

Abstract

A method for increasing the reliability of information signal transmission over digital communication channels with additive interference is proposed. The algorithmic basis of the proposed method is to represent the signal in the form of a decomposition according to one of the orthonormal discrete functional bases (Walsh, Hartley, discrete cosine, wavelet) with the subsequent limitation of the number of decomposition coefficients, optimized according to the criteria of minimizing the information distance between the transmitted and received signals and maximizing the signal/noise ratio.

The implementation of the proposed method includes the following stages: digitization (digitization) of the processed sampled information signal with samples obtained at consecutive points in time; limitation of the number of expansion coefficients (signal compression with partial loss of information content) with subsequent reconstruction of the quantized signal; phase modulation by the digitized signal of the carrier sinusoidal signal with a given frequency and transmission rate.

The determination of the thresholds for limiting the set of coefficients of the expansions of digitized signals is formulated as an optimization problem of achieving predetermined values ​​of the information distance between the transmitted and demodulated signals and the ratio of the signal energies of the demodulated signal and the residual noise.

The Lloyds algorithm was used for digitization, which ensures minimization of the digitization error.

It is established that with an increase in the number of digital levels of digitization, the optimal values ​​of the information distance between the transmitted and demodulated signals increase with a decrease in the signal-to-noise ratio.

The generalization of the method to multidimensional signals is implemented by reducing the dimensionality based on the Peano-Hilbert sweep

Comparison of various discretized functional bases as a basis for implementing the proposed method showed the greatest efficiency according to the specified criteria of the fourth-order Daubechies wavelet basis.

Keywords: discretized signal, orthonormalized discrete functional basis, multispectral image, signal-to-noise ratio.