This article presents a research study dedicated to the application of the YOLOv8 neural network model for road sign detection. During the study, a model based on YOLOv8 was developed and trained, which successfully detects road signs in real-time. The article also presents the results of experiments in which the YOLOv8 model is compared to other widely used methods for sign detection. The obtained results have practical significance in the field of road traffic safety, offering an innovative approach to automatic road sign detection, which contributes to improving speed control, attentiveness, and reducing accidents on the roads.

Keywords: machine learning, road signs, convolutional neural networks, image recognition

The paper presents results of a study of operational recursive methods for quasiharmonic signals frequency estimating over a short observation interval. Such methods should be used, for example, for adaptive control of directional pattern of hydroacoustic antennas in inhomogeneous medium. We proposed recursive expressions for realizations of four methods of frequency estimation within the limits of sliding observation interval. The expressions require a minimum number of mathematical operations, which does not depend on the observation interval length. We carried out comparative analysis of the speed, accuracy, and stability of the methods. The stability of the methods was analyzed based on results of model signals processing with constant amplitude, frequency, and initial phase in presence of additive white normal noise. The accuracy was investigated by the processing results of signals with slow varying amplitude and phase. The paper also presents the results of processing of long sequences of modulated signals with additive noise within a sliding window with predetermined duration. The obtained results make it possible to select from among the four methods the optimal one according to one or more criteria.

Keywords: communication channel, directional pattern, frequency estimation, recursive method, accuracy of method, stability of method, sliding interval

The paper is devoted by studying of the correctness of the antenna pattern adaptive control problem in a randomly inhomogeneous medium. We use spectral methods for the solving the Helmholtz equation with variable parameters in the frozen medium approximation. We have proved the correctness of the problem of a given antenna pattern forming in a randomly inhomogeneous medium for the measured distribution of the complex wave amplitude in the plane of receivers in the first time. The possibility of using an antenna with a synthesized aperture for adaptive correction of the antenna pattern for maximization of communication channel capacity is substantiated. We use the refractive index model as the sum of the deterministic and random components in numerical modeling. The sampling intervals were chosen in accordance with the sampling theorem. The analytic expressions with given parameters were used as a model of the law of the antenna array operation. The results of numerical simulation confirmed that, in the first approximation, the problem of the antenna pattern controlling of a transmitter with a synthesized aperture based on the distribution of the complex wave amplitude measured in the plane of receivers is correct. It allows to create a system of adaptive antenna pattern correction for maximization of communication channel capacity.

Keywords: communication channel, directional pattern, complex amplitude, synthesized aperture, angular spectrum method, adaptive control

The work contains the approximation of the third boundary value problem of schemes of high order. The approximation of the operator of the convective and diffusive transport difference schem were obtained, which has the fourth-order approximation error. The discrete operator diffusive transport in the absence of the influence of the boundary was considered.The construction of schemes of higher order accuracy approximates the operator diffusive transport in the border nodes in the case of boundary conditions of the third kind with the second order of accuracy was found.

Keywords: Schemes of high order, diffusion, diffusion transfer, discrete operator, approximation, boundary conditions, boundary value problem, finite-difference scheme

The work contains the construction and software implementation of related software systems designed for mathematical modeling of processes occurring in coastal water systems. The space-two-dimensional model of sediment transport in shallow waters allows to forecast the dynamics of bottom topography changes by reason of the movement of water and solid particles, to take into account the porosity of the soil, the critical shear stress at which begins the movement of sediment, turbulent exchange, dynamically changing geometry of the bottom, wind currents and friction on the bottom. The mathematical model of motion of a multicomponent air, takes into account factors such as the transport of pollutants and heat; the effect of vegetation; change in the coefficient of turbulent exchange; water transfer from the liquid to the gaseous state and others.

Keywords: multiprocessor computer system, grid equations, suspended solids, sediment

The method of digital parametric measurement of the phase shift of signals with slowly varying amplitudes of the same slowly varying instantaneous frequencies was proposed. The numerical simulations and experiments on the developed hardware and software system were carried out. The error of measurement the phase difference of harmonic signals in the experiments were not more than  0,00001 radian. The method can be used in applications where the measured value is expressed in terms of the phase difference of quasi-harmonic signals.

Keywords: quasi-harmonic signal, phase shift, real time measurements, hardware and software system

Analog-to-digital converter, with modulation, amplification and demodulation of the input signal is carried out. To compensate the additional error of voltage conversion, concerned with temperature instability, used dynamic cancellation technique, based on the correlation between offset voltage and modulator temperature fluctuation. ADC with response time 0.85 s performs the threshold of sensitivity 1 nV; 8,5 s - 0,31 nV, 85 s - 0.13 nV, 850 s - 0.1 nV, respectively, in the full-scale range 400 mV. Estimation of offset voltage temperature drift is 40 pV / ºC, time drift - 100 pV / h.

Keywords: analog-to-digital converter, ADC, error cancellation.