Home – Home – Jornals – Avtometriya 2023 number 4

Given an ensemble of datasets, we study the object classification accuracy in terms of the error probability depending on the amount of processed information using various fusion schemes. Schemes of combining weak discriminant functions in each dataset as well as in an ensemble of different modality datasets are suggested. For the proposed fusion schemes, the redundancy of the error probability relative to the information-theoretic lower bound defined by the modified rate-distortion function with the Hamming distortion measure is evaluated. The experimental evaluations in datasets of signature and face images show a decrease in the error probability and its redundancy with the amount of the processed information being increased by combining weak discriminant functions.

The study deals with a conflict situation associated with the movement of clusters, including hundreds of small control objects in the field of responsibility of the radar station of an external observer. The task of the external observer is to monitor and trace the cluster movement and control it if a threatening situation occurs. The task of the cluster is to create the maximum possible uncertainty in decision making by the external observer. The cluster motion model is developed as K. Reynolds’ swarm behavior, supplemented by special coefficients. Three methods are developed for the formation of various cluster shapes, close to geometric and changing in the process of movement. A hypothesis is formulated and confirmed by mathematical modeling that, due to variations in the geometry, shape, and number of objects in a swarm cluster, there exist a possibility of creating situations where a swarm will be perceived by the external observer as a single large object different from the original one. It is shown that it is possible to create clusters simulating dangerous objects if the criterion for making a decision is the value of the effective scattering surface.

This article describes features of automated implementation encryption of both biometric data and user information resources as a whole. The created mathematical model is based on an artificial neural network designed for encryption of biometric images in the account administrator folder using mathematical methods. The object of the study is the information systems of the PC. The subject of research is the tools of protection of user information using biometric multifactor authentication.

The article is devoted to the application of machine learning methods for sentiment analysis of texts and to the study of the effectiveness of various architectures of neural networks. This direction is relevant in connection with the development of social networks and online recommendation services, where many users express their opinions about goods and services. The article presents the results of forecasting and comparing the structures of neural networks on real data from social networks. This allows determination of the most efficient architecture for the sentiment analysis of texts. The study may be useful for developers of social networks and recommendation services, as well as for researchers involved in natural language processing. The results can help improve the quality of the user opinion analysis and improve user satisfaction with services and products. Thus, the article contributes to the development of the field of machine learning and text data analysis.

A non-asymptotic procedure for constructing the confidence region of the parameter of the Gaussian autoregression process of the p-th order with an unknown variance of the process noise is proposed. The confidence estimation procedure is based on the martingale property of the numerator of the deviation of the least squares estimator. The paper presents the results of numerical simulation.

A method is proposed for synthesis of energy-efficient controllers in both open and closed domains of definition of spatiotemporal control actions for linear multidimensional objects of the parabolic type with distributed parameters under the conditions of a specified accuracy of uniform approximation of the system's final state to the required spatial distribution of the controlled variable. The suggested approach is based on the previously developed alternance method of constructing parameterized algorithms for optimal programmed control, which utilizes fundamental regularities of the subject area. It is shown that the sought controller equations are reduced to linear laws with constraints for feedbacks based on incomplete observation of the state of an object with non-stationary transfer coefficients, which are determined by a preliminary calculation of the programmed control action.

The problem of functional observer design for dynamic systems described by nonlinear differential equations under external disturbances is considered. Relations that allow design of the observer of a reduced-order-dimensional invariant with respect to the disturbances and estimating the prescribed function of the state vector are derived. Theoretical results are illustrated by an example.

The MTSS simulation system implements the possibility of distributed simulation. To implement this feature, the HLA standard is used. As an example of the use of distributed simulation, a federation consisting of two federates is presented: a simulation model of a coal mine working face and a simulation model of a coal mine conveyor network.

The paper formulates a numerical algorithm for searching for optimal initial concentrations of substances in a catalytic reaction, which combines the principles of kinetic modeling and evolutionary calculations. The problem of optimal control of the catalytic reaction is formulated, where the control parameter is the vector of the initial concentrations of the initial substances. A step-by-step algorithm for solving the problem is described, which is based on the method of differential evolution. A computational experiment is carried out for the reaction of aminomethylation of thiols with the help of tetramethylmethanediamine. With the help of the above-described algorithm, the optimal concentrations of the initial substances are calculated, at which the highest value of the concentration of the reaction product is reached. It is shown that the calculated values of the initial concentrations of substances are consistent with the results of a chemical experiment.

Classical opto-acoustic receivers (OAR) are successfully used to solve a number of scientific and applied tasks, such as measuring weak optical radiation fluxes in the IR and THz ranges, quantitative and qualitative analysis of gas mixtures, and studies of absorption spectra of gases and vapors. The purpose, design, and evolution of the main structural elements of classical OAR are considered. The prospects of using single-layer graphene as the most promising material for creating membranes are analyzed. A significant increase in sensitivity is due to the exceptionally high elasticity of graphene combined with its high mechanical strength. It is shown that the use of a flexible membrane made of single-layer graphene reduces the OAR susceptibility to acoustic and vibration noise by more than three orders of magnitude without the use of any vibration protection devices. Due to the preservation of high elasticity by the graphene membrane at helium temperatures, the OAR operating range can be extended from 320 K to helium temperatures, which ensures reaching the theoretically maximum threshold sensitivity by reducing noise and opens up the possibility of wide application of the OAR for solving scientific problems of astrophysics and cosmology. Capacitive, optical, tunnel, and cantilever detectors of microdeformations of a flexible membrane are considered. It is shown that the maximum metrological characteristics of the OAR can be reached due to combining the unique capabilities of two developments: a membrane made of single-layer graphene and a tunnel microdeformation detector.

This article is a logical continuation of the previous analysis of the dynamic characteristics of adaptive optics systems operating in a turbulent atmosphere, in particular, the influence of such an important factor of an active corrector mirror as its hysteresis is analyzed. An analytical approach is used to calculate the requirements to the dynamic parameters of the feedback loop of an adaptive system for focusing laser radiation through a turbulent atmosphere. The possibilities of using control algorithms in adaptive optics systems based on the description of their operation as a constant delay control system are considered. A corrector mirror model is used, which takes into account the time delay between the time of the signal and the time of its processing. To estimate the frequency of operation of the adaptive optics system, an analytical expression is used that relates the maximum achievable level of correction by the Strehl parameter with the most important system parameters: the accuracy and frequency of the wavefront sensor and the aperture size of the optical system, as well as with atmospheric parameters: the Fried parameter and the wind speed. The differences between two types of tracking loops are analyzed: open and closed loops in adaptive optics systems.

The results of numerical simulation of the irradiation by parasitic radiation of a cooled matrix photodetector in a cryostat with different shapes and designs of the cold diaphragm are presented. The optimal shape of the cold diaphragm is found to be a cylinder with two internal blends (additional diaphragms). The level of irradiation of the matrix photodetector with parasitic radiation is about 9.82×10^-05 W/cm² with an uneven distribution over the area of about 3%.