Audio Steganography Method Using Determined Chaos

Abstract

At present, the discovery of deterministic chaos is one of the significant subjects of research in various fields of science. In this work, deterministic chaos considered as a sequence with specific properties, which is used to transmit information in a stegosystem. The efficiency of the algorithm for steganographic embedding and extraction of digital information into audio containers by the method of spreading the spectrum using chaotic signals is analyzed. The spread spectrum method is characterized by the presence of a large number of errors on the receiving side when used in the classical version, without the original audio, i.e. there was no possibility of blind extraction. Errors arising when transmitting an audio signal over a communication channel made it impossible to extract the embedded information correctly. It proposed to use an algorithm that increases the efficiency of information signal extraction on the receiving side. The efficiency of the method is estimated by the amount of distortion in the original audio and by the reliability of extracted information, which is characterized by the value of the error probability. It showed that to extract embedded data with high reliability from stegoaudio, one should use an adaptive filtering algorithm - thresholding using discrete wavelet transform (DWT). The best filtering method, using various orthogonal mother wavelets (Haar, Daubechies, and Symlet) at different levels of signal decomposition, was determined. In the course of this study, recommendations were made on the use of mother wavelets, and was given preference to Daubechies and Symlet wavelets. Requirements for the numerical values of the embedding parameters and the parameters of the generators of chaotic processes used to expand the spectrum formulated.

Features of Fractal Conformity and Bioconsolidation in the Early Myogenesis Gene Expression and Their Relationship to the Genetic Diversity of Chicken Breeds

“…The analysis of fractals can ultimately be linked to the derivation of fractal dimension D. If the data has fractal properties, we can approximate it with an exponent. Herewith, a curve will be obtained, and when taking a logarithm (in ordinary coordinates), it will be a straight line described by Equation (2) while its slope angle will conform to the fractal dimension D. The fractal dimension D is determined as the tangent of the slope of a plotted curve [58]. Based on this, we evaluated the possibility of calculating the fractal dimension D by simplifying Formula (2) and removing the variable c from it to obtain the following equation found also in other literature sources:…”