The article examines the prerequisites for developing a comprehensive system for protecting the population and critical infrastructure equipment from exposure to radiofrequency electromagnetic fields emitted by 5G mobile communications equipment. The need for this system stems from the anticipated widespread use of TDD (time-doubled duplex) mode in 5G/6G mobile communications systems, with a pulsed equivalent isotropically radiated power one to two orders of magnitude higher than that of mobile base stations operating in traditional (frequency-doubled duplex) mode with quasi-continuous radiofrequency electromagnetic fields. The current practice of defining protective zones and development restrictions based on average values of the energy flux density of radiofrequency electromagnetic fields does not provide protection for either critical infrastructure equipment and facilities (whose susceptibility to these fields is determined by their instantaneous field strengths) or the population, due to their significantly higher bioaggressiveness compared to quasi-continuous FDD (frequency-doubled duplex) radiofrequency electromagnetic fields.

The effect of nitridation of silicon oxide layers obtained by thermal oxidation by pulsed photon treatment in a nitrogen atmosphere on the electrical characteristics of transistors and capacitors with a metal-oxide-semiconductor structure was studied. Silicon oxide layers with thicknesses of 7.7 and 35.0 nm were obtained by oxidizing single-crystal silicon substrates in dry oxygen. Nitridization of the layers was achieved by heating the substrates in a nitrogen atmosphere to a temperature of 1150 °C for 7 s using a pulse of incoherent radiation of constant power from quartz halogen lamps directed at the non-working side of the silicon substrate. In a study of the current- voltage and capacitance-voltage characteristics of test elements with a metal-oxide-semiconductor structure, it was found that nitridation by pulsed photon treatment leads, as a result of the restructuring of the oxide layer, its compaction, and a decrease in the density of charge states from 1.5 to 4 times, to a decrease in the gate leakage current from 3.3 to 7 times, an increase in the gate breakdown voltage by 1.05 times, a decrease in the threshold voltage shift during thermal field tests from 1.3 to 1.6 times and an increase in the breakdown charge by 9–13 %. The obtained results can be used in enterprises of the microelectronics industry in the manufacture of electronic products with metal-oxide-semiconductor structures.

A modification of the method of parametric optimization of non-stationary thermal conduction processes is proposed in relation to the problems of laser heat treatment of a wide class of materials used in electronic engineering, including semiconductors, dielectrics and superhard materials. Using the Matlab computer mathematics package, a computational algorithm has been developed for the automatic construction of the optimal heating mode according to the specified quality criterion, which makes it possible to significantly reduce the time for the experimental selection of processing parameters. The results of verification of the method are presented on the example of calculating the optimal modes of laser marking of electronic products, confirmed by comparison with known experimental data, which demonstrates the high accuracy of the proposed approach. Particular attention is paid to laser-induced graphitization of diamond, which makes it possible to form conductive structures in the volume of superhard materials, which opens up new prospects for the creation of functional elements of electronic technology based on diamond substrates.

A millimeter-wave antenna design with an integrated phase-controlled metasurface for advanced B5G adaptive wireless communication systems is presented. The antenna system combines a planar antenna array of 32 patch antennas with an active metasurface based on microresonators with embedded varicaps, which provides a highly focused beam and its electronic deflection in various directions. The power supply network ensures in-phase excitation of the emitters with a phase spread of no more than 16° and an amplitude imbalance of less than 3.8 dB at a frequency of 25.7 GHz. The metasurface unit cell implements a Huygens resonance, resulting in a transmission coefficient no worse than –3.5 dB and a reflection coefficient of less than –15 dB. By varying the varicap bias voltage from 0 to 20 V, an electronic phase tuning range of approximately 304° is achieved, enabling dynamic beam direction control. Numerical simulation confirmed the feasibility of forming a narrow beam with electronic deflection over a wide range of angles (up to 40°). The proposed design has a low profile, is easy to manufacture, and meets the requirements of adaptive antenna systems for advanced communication standards to improve energy efficiency and spatial multiplexing. The results demonstrate the high potential for using the developed antenna in reconfigurable millimeter-wave antenna systems.

An evolutionary synthesis algorithm for neural network architectures is proposed. It is designed to address real-time process monitoring tasks with limited computing resources and training data. The algorithm is based on a modified neuroevolution approach, in which the individual genome comprises a graph of large functional blocks and the history of innovations is tracked. The evaluation on population individuals was performed by the multi-criteria algorithm NSGA-II based on classification accuracy and the forward pass time network on the target device. The algorithm’s performance is demonstrated on an image classification task using the CIFAR-10 dataset, as well as on the applied task of classifying operating modes during the single-beam laser polishing of quartz glass.

This article presents a study of the physical processes at the interface in thin-film structures based on two- and three-component chalcogenides, which determine the mechanisms of phase transitions and memory. An alternative concept for resistive threshold switching at the interfaces of two- and three-component chalcogenide semiconductor materials is proposed, resulting in the formation of metastable dichalcogenide compounds. Methods for the structural stabilization of such compounds are discussed.

DVB-T2 systems deployed in urban areas are significantly susceptible to multipath propagation and intersymbol interference. This article examines the possibility of optimizing guard interval settings for DVB-T2 systems supporting orthogonal frequency division multiplexing to minimize the bit error rate and improve signal transmission efficiency in urban environments. Using MATLAB simulations, four different guard intervals (1/4, 1/8, 1/16, and 1/32) were analyzed, taking into account channel variations. The results showed that the optimal guard interval setting was 1/8, which reduced the bit error rate from 10−3 to 10−5 at an SNR of 18 dB, and increased signal coverage efficiency by 22 %.

An adaptive modulation control algorithm for visible-light communication systems is proposed. It utilizes a combined analysis of the Hurst exponent and a local estimate of the signal-to-noise ratio derivative. The algorithm utilizes asymmetric switching logic: modulation order upshifts are permitted only when the conditions of a positive local derivative estimate and high time series persistence are simultaneously met, while downshifts are performed immediately upon detection of a directional degradation. The algorithm’s threshold parameters are substantiated by analyzing speed-gain heat maps constructed for experimental scenarios. The proposed algorithm reduces the number of modulation order switches relative to the classical threshold method by 7.1 to 81.2%, depending on the channel dynamics.

The article presents analytical models for calculating and a methodology for studying the noise immunity, energy, and information efficiency of systems with sequential composite coding based on long-length convolutional codes in combination with binary block Bose – Chaudhuri – Hocquenghem codes, and convolutional and nonbinary Reed-Solomon codes. The calculation models allow for detailed studies of the properties of composite code structures with multi-position modulation types and decoding using the Viterbi algorithm with a soft decision for convolutional and Bose – Chaudhuri – Hocquenghem codes and a hard decision for Reed – Solomon codes. Unlike known approaches, this methodology does not require knowledge of the spectral weighting coefficients of the codes used, does not limit the choice of code locations within the code structure, and does not involve the use of complex computer modeling procedures in the calculations. The studies are performed in a closed-loop manner based on the specified parameters of the codes in the composite structure and the requirements for ensuring reliable reception. It is shown that the noise immunity of systems with two-stage sequential coding based on convolutional codes with a design length of K = 13 in combination with simple convolutional codes, such as Bose – Chaudhuri – Hocquenghem or Reed – Solomon codes, is comparable to the noise immunity of the best turbo code models with iterative parallel MAP decoding. Furthermore, the disadvantages of turbo codes caused by the use of high-dimensional interleavers are eliminated.

Neural network processing of high-dimensional data requires significant computing resources. An incorrect choice of neural network architecture or data transformation can lead to significant losses of time and resources in network retraining. This article proposes an algorithm for selecting the architecture of compressive layers for a deep belief network and data transformation based on the analysis of a multi-label confusion matrix. The introduced generalized skewness coefficient describes the uniformity of the distribution of objects between neural network classes and allows for a numerical evaluation of the effectiveness of the network’s compressive layer architecture and data transformation. Using an applied problem as an example, it is demonstrated that the proposed algorithm improves the quality of its solution by more than 10 % by selecting the most effective neural network architecture.

To ensure safe transitions between automated and manual driving in automated vehicles, it is necessary to monitor the driver’s state. Existing research in this area primarily assesses the driver’s current state without predicting their potential behavior in complex driving situations, which could negatively impact the safety of automated vehicles when switching between driving modes. To address this shortcoming, this article proposes a method for assessing relatively stable individual driver psychophysiological characteristics that could serve as predictors of their ability to mobilize internal reserves when making decisions in complex driving situations. The method allows one to determine the numerical values of indicators of such individual psychophysiological characteristics as reaction time, ability to take emergency actions, vigilance, concentration and distribution of attention, perception of speed and distance, as well as risk propensity during manual driving.