Structural-linguistic approach to the description of spectrogram of electromagnetic radiation spectrograms generated by computing technology

In this paper, the solution of an urgent scientific and technical problem in the field of information security is considered – the problem of detecting electromagnetic radiation generated by computer technology, within the framework of a structural and linguistic approach to the analysis of experimental data. The implementation of the stage of assigning segmented sections of spectrograms of symbols of some alphabet corresponding to certain types of behavior is considered in detail. As well as the implementation of the analysis stage of the received sequences of symbols. A model of morphological grammar construction and a mechanism for generating texts that differ little from the correct ones are described. The grammar formation stage was implemented in the Python 3.7 programming language. The choice of this programming language is justified by its cross-platform nature, low entry threshold, as well as wide application: from automation of mathematical calculations and machine learning, to the development of web applications. Also, one of the advantages of this language is the availability of many high-quality libraries, including those used for current development. A statistical experiment was conducted to evaluate the effectiveness and accuracy of the developed algorithms. An estimate of the probability of the correct classification of linguistic chains of experimental spectrogram curves into the desired groups and a general assessment of the accuracy of the correct determination of all curves into the desired groups is given. As a result of the application of a structural-linguistic approach to the problem of detecting electromagnetic radiation generated by computer technology, it can be concluded that the obtained linguistic descriptions of the studied spectrograms represent short and reliable rules for their analysis and allow for high accuracy in automated mode to determine deviations in the observed spectra from the specified standards.