Abstract. By modifying the algorithm for generating non-binary Gordon — Mills — Welch sequences (GMWS) in finite fields GF(p^S), where S = mn (m, n > 1), an algorithm for building GMWS for p ≥ 5 with an arbitrary degree of expansion S of the field GF(p) is developed. It is shown that for simple GF(p) fields in which the Euler function φ(p – 1) > 1, it is possible to form non-binary GMWS in the extended GF(p^S) fields, where the parameter S can be either a composite or a prime number. The vectors of decimation indices for forming GMWS in the GF(5²), GF(5³), GF(7²), GF(7³), GF(11²), GF(11³) fields are given, as well as the values of the equivalent linear complexity l^S of these sequences.

The problem of detecting and localizing defects in linear nonstationary dynamic systems in the presence of external disturbances and measurement noise is considered. To solve the problem, interval observers are used, built by converting the initial system to its reduced model, which is sensitive to defects and insensitive to disturbances. The special feature of the model is its nonlinear stationary dynamics, which, unlike known approaches, significantly simplifies the procedure for solving the problem.

The problem of precise terminal control of nonlinear systems with State-Dependent Coefficients (SDC) over a finite time interval is solved. The purpose of the article is to develop a method for synthesizing suboptimal control, which ensures high accuracy of bringing the system output to a set value by a fixed point in time. Further development of the method of inverse integration of the Riccati equation with state-dependent coefficients is discussed by using a terminal controller synthesized for the auxiliary control problem in reverse time to determine the state of the system during reverse integration instead of the standard stabilizing regulator (SDRE regulator). The effectiveness of the method is been tested on the reference task of controlling the Van der Pol oscillator. The simulation results are presented, demonstrating a higher accuracy of the proposed terminal method of inverse integration compared with the classical approach using the SDRE controller, and comparability with the iterative method of approximating the sequence of the Riccati equation. The method is recommended for the tasks of precise control of complex objects at finite time intervals.

Dynamic shooting conditions create the effect of blurring images, which leads to distortion of the recorded scene. To ensure the necessary efficiency of the star sensor, additional image processing methods aimed at compensating for distorting factors are required. An algorithm for estimating the centroids of stars in blurred images based on the methods of noise reduction and image deconvolution is proposed. At the noise reduction stage, a bilateral filter is applied to the original image, which makes it possible to increase the signal-to-noise ratio of the stars several times. Next, background noise is removed to detect suspected stars. The image cleared of the background is sent to the input of the Lucy—Richardson algorithm for solving the deconvolution problem. A single-pass clustering algorithm is applied to the image recovered during deconvolution to estimate the centroids of the detected stars. According to the experimental results of testing the developed algorithm on three test sets of images with different blur parameters, the described approach provides higher accuracy and reliability with fixed blur parameters than two other previously known analogues. However, to increase the efficiency of the proposed algorithm, an accurate assessment of the blur parameters is required, otherwise large errors in the estimation of coordinates are possible.

The process of calibration of a spectroscopic analytical instrument (SAI) designed to analyze the spectral composition of optical radiation from remote objects is considered. The aim of the study is to develop a SAI calibration method that eliminates the need to account for atmospheric transmission effects. The proposed approach is based on the use of narrow-spectrum radiators acting on the test SAI directly in the plane of its entrance porthole. To present the method, the concept of a two-dimensional spectral characteristic of SAI sensitivity is introduced, and the expression for the spectral sensitivity of SAI is derived using the SAI apparatus function. The presented theoretical calculations and conclusions are confirmed by experimental results. The use of the developed method eliminates the need to take into account the spectral transmission of the working air path (atmosphere) and expands the scope of its application for remote sensing equipment, medicine, agricultural technologies and environmental monitoring.

The possibility of creating a measure of the spectral coefficient of diffuse reflection for the wavelength range of 2–20 µm for use in laboratory spectral equipment is discussed. The results of the production of measurement samples and the study of their characteristics are presented in relation to the tasks of measuring the spectral coefficient of diffuse reflection of rough surfaces with low reflection on devices with diffuse reflection prefixes. The proposed approach to the selection of the surface structure, its coating and measure is the basis for measuring the characteristics of a wide range of materials by comparison in the infrared range of 2–20 µm. One of the variants of the comparison measure has been adapted to work on diffuse reflection consoles of domestic fsm series spectrometers. The application of the proposed method makes it possible to significantly increase the efficiency and accuracy (reliability) of measuring the spectral dependences of diffuse reflection coefficients of materials and coatings with rough surfaces and low reflection coefficients.

Method of the diffusion coefficient non-destructive testing in thin products made of anisotropic porous materials is the subject of this work. The method makes it possible to obtain the required results in the absence of the applied diffusant converter real static characteristic in the solid phase. The effective flexibility application conditions in terms of the possibility for using different sections of the response curve to pulse diffusant dose exposure is analyzed. The aim of the research is to optimize the method in terms of accuracy and speed. Methods of mathematical metrology and error theory were used to solve this problem. The influence on the accuracy and speed of the non-destructive testing method of the ratios of the maxima of the output characteristic of the Emax concentration converter observed in experiments and the values of the output characteristic E_cal, at which the time points included in the calculated ratios are fixed, is analyzed. A natural decrease in the errors in fixing the time points included in the calculated ratios and a decrease in the performance of the non-destructive testing method with an increase in the difference (E_max – E_cal) are proved. The problem of multiparametric Pareto optimization has been solved in the presence of two irreducible criteria: minimum error and minimum duration of the diffusion coefficient measurement process. The results of the efficiency of the nondestructive testing method for determining the diffusion coefficient are presented, taking into account the optimal values of the parameters found in comparison with the initial data.

Results of the development and manufacture of three types of neonatal electrode systems for electrical impedance tomography in pediatrics, adapted to the anatomical and physiological characteristics of newborns, are presented. The proposed systems differ in the materials of the electrodes and their design. The results of experimental studies of normal respiration (eupnea), using a simplified physical model of the mediastinum of newborns, carried out on an experimental stand for electrical impedance tomography with an EI tomograph are presented. The results of the assessment of the operability of manufactured products confirm the prospects of their use for monitoring global and regional air filling parameters in a given section, including the monitoring of the respiratory function of newborns.

A new approach to text classification is proposed that does not employ machine learning methods or require a training set. The method is based on the Damerau-Levenshtein distance, which is the minimum number of editing operations required to transform one string into another and takes into account the semantic similarity of words, weighting of editing operations, and the order of importance of words. The main metrics for assessing the quality of a text classifier and the results of testing the proposed method against these metrics are presented.