2687-0517 17 2 2024 1-70

RAR RUS 7-16 Yanchus Victor victorimop@mail.ru 0009-0001-0503-3261 Peter the Great St. Petersburg Polytechnic University Kukulian Victoria kukulyan.vyu@edu.spbstu.ru

St. Petersburg, Russian Federation

The influence of highlighted text on its perception when performing interpreting from a computer screen This paper examines an experimental study of the effect of highlighted text with color underlays on the efficiency of its perception when translating from a screen. The study was conducted using eye-tracking technology and oculography methods, to analyze the perception of stimulus material by interpreters when translating from a screen. Methods of mathematical statistics were used to process the experimental data obtained. During the experiment, participants were asked to interpret from the screen sentences presented in stimuli with different options for highlighting the text with color. Factor analysis of the experimental data obtained from the eye tracker revealed statistically significant differences in text perception effectiveness depending on the combination of the investigated factors. The analysis demonstrated that the use of a serif font, a contrasting black background, and white underlay for highlighting the semantic centers of sentences can significantly improve the quality of an interpreter's work with on-screen text. The results of this study can be applied in the development and optimization of the interface design for specialized simulators and training systems for professional training of interpreters. 10.18721/JCSTCS.17201 004.5 eye tracking experiment highlighted text text perception translation education training simulator https://infocom.spbstu.ru/article/2024.81.1/

RAR RUS 17-23 0000-0002-1477-3181 Peter the Great St. Petersburg Polytechnic University Leontiev Evgeniy johnleon010@gmail.com

St. Petersburg, Russian Federation

Class G power amplifier synthesis based on the probability density function dependence of the transmitted signal The article describes the technique of parametric synthesis of a class G power amplifier with maximum drain efficiency at the average output power with high PAPR (peak to average power ratio). The proposed technique was applied to an advanced OFDM signal of LTE standard. The optimal parameters of the envelope amplifier for LTE signal were received on the basis of proposed theoretical calculation. Proposed technique validity was confirmed experimentally by measuring the 3-level class G power amplifier prototype. The prototype operates in the frequency range of 700–1000 MHz and has a maximum power 44.3 dBm. It also has a drain efficiency of 30% at power output back-off equal to the peak to average power ratio of LTE signal. From the consistency of the computational and experimental results, it is possible to predict increasing a drain efficiency from 30% to 39.5% by using the digital pre-distortion system and the 5-level envelope amplifier. 10.18721/JCSTCS.17202 621.375.026 class G power amplifier PDF drain efficiency OFDM LTE https://infocom.spbstu.ru/article/2024.81.2/

RAR RUS 24-37 Peter the Great St. Petersburg Polytechnic University Varlamov Dmitrii varlamov_dmitry99@mail.ru

St. Petersburg, Russian Federation

Nikiforov Igor igor.nikiforov@gmail.com 6603839750 Peter the Great St.Petersburg Polytechnic University Sergey M. Ustinov usm50@yandex.ru

Polytechnicheskaya, 29, St.Petersburg, 195251, Russia

Algorithm for monitoring and improving the stability of the IT infrastructure based on availability and reliability metrics Most companies have their own IT infrastructure that consists of complex systems and services. The stability of systems and services is important for companies, as problems with them can lead to loss of resources and human time. Thus, it is important to analyze previous IT service outages, which aims to identify and adjust the most critical and vulnerable elements of the infrastructure that are prone to breakage or failure. Research objective is to develop a new algorithm for improving the stability of IT infrastructure of a company by analyzing and taking into account the statistics of previous services outages. As a result, a new algorithm is proposed to identify and fix problems in IT services before they lead to serious consequences and reduce the time to find the source of problem. The algorithm is based on two new metrics: availability and reliability, which distinctive feature is the consideration of statistics of previous failures and outages in the system. The architecture of a high-performance software tool that allows real-time monitoring and evaluation of IT services stability metrics is presented. The effectiveness of the proposed algorithm is demonstrated by implementing it in a software tool and observing the growth of stability indicators – availability and reliability – after the detection and elimination of a weak link in IT services. The use of the developed algorithm allowed to reduce the time during which the material and human resources of the company were idle by 25%. The practical significance of the presented algorithm was tested in one of the large industrial information technology companies with more than 10000 employees. Based on the information obtained with created software, it was possible to obtain recommendations for improving the stability of company’s IT services. 10.18721/JCSTCS.17203 004.052.3 metrics availability reliability stability IT infrastructure outage monitoring https://infocom.spbstu.ru/article/2024.81.3/

RAR RUS 38-51 Baikal State University Timofeev Sergey timofeevsv12@gmail.com

Irkutsk, Russian Federation

Baikal State University Baenkhaeva Ayuna ayunab2000@mail.ru

Irkutsk, Russian Federation

A mathematical model of information confrontation: discrete adaptive control of the system This article presents initial results of controlling a mathematical model of information confrontation, proposed by the authors in earlier works. The model is a system of ordinary differential equations with quadratic nonlinearity on the right side. The introduction defines the novelty of the approach and outlines the differences from previously used ones. Additionally, the substantive meaning of the variables and parameters of the system is described, and the stages of research on this model are briefly presented. In the parameter space, a component is established, by controlling which we obtain relations that determine the predictable behavior of the system trajectory from any initial point corresponding to the substantive meaning. In the main part of the work, an algorithm for constructing discrete adaptive control is proposed, which allows reducing the information confrontation to a scenario advantageous for one of the parties. The example shows the procedure for using the model and the algorithm for building the control step-by-step. The numerical solution of the system of differential equations was performed using the solve_ivp module for solving ordinary differential equations of the SciPy library of the Python programming language. 10.18721/JCSTCS.17204 517.938:070 mathematical model information confrontation information promotion differential equation adaptive control numerical solution of differential equation system https://infocom.spbstu.ru/article/2024.81.4/

RAR RUS 52-61 Gebel Elena gebel_es@mail.ru Approaches and principles for advanced control of multi-flow tube furnace To improve the efficiency of multi-flow tube furnaces, a multi-parameter advanced cohesive control system is proposed. Stabilization of the main technological parameter, i.e. the resulting temperature of the output material flow, is carried out by supervisory control of flow rates on the inlet coils, taking into account the existing limitation on the loading of the apparatus, while meeting the requirement of a uniform temperature profile. To create an optimal combustion mode, an additional circuit for regulating the discharge in the radiant chamber of the furnace on the line of atmospheric air supply to the burner registers according to the residual oxygen content in the flue gases is introduced. In the developed simulation model of the control system the possibility of shockless transition between the basic, as in production, and advanced variants of the control system is realized. Transfer functions of the multilink control object, as well as the tuning parameters of the regulators on individual channels are obtained by the built-in tools of Matlab Simulink. Computational experiments on the model showed high speed and control accuracy. 10.18721/JCSTCS.17205 681.513.1 multi-flow furnace cohesive control system simulation model temperature profile efficiency factor https://infocom.spbstu.ru/article/2024.81.5/

RAR RUS 62-70 0000-0002-7060-8826 Shariaty Faridoddin shariaty3@gmail.com 0000-0003-0726-6613 Peter the Great St. Petersburg Polytechnic University Pavlov Vitalii pavlov_va@spbstu.ru

St. Petersburg, Russian Federation

Peter the Great St. Petersburg Polytechnic University Surakov Murat murat.surakov@gmail.com

St. Petersburg, Russian Federation

Comparison and selection of radiomic and deep convolutional features for improving the accuracy of CT-image texture classification The article explores the problem of comparing and selecting radiomic and deep convolutional features extracted from CT images to enhance the accuracy of texture classification in CT diagnostics. By using the mRMR method, the study assesses the significance of these features in predicting genetic mutations in patients with lung cancer, highlighting their importance for refining diagnostic procedures. The developed predictive model demonstrates high classification accuracy of 92%, which indicates its high efficiency. Analysis of the results reveals that deep learning features effectively capture complex, high-level abstract textures that indicate the presence of pathologies. At the same time, radiomic features provide key information about the phenotypic characteristics of tumors, such as shape, texture, and intensity. This comprehensive approach not only improves the accuracy of non-invasive diagnostics, but also contributes significantly to personalized medicine by facilitating the development of more precise treatment strategies based on genetic profiles. 10.18721/JCSTCS.17206 004.932.72 radiomics deep convolutional features computed tomography machine learning feature selection https://infocom.spbstu.ru/article/2024.81.6/