Radio Electronics, Computer Science, Control最新文献

{"title":"MACHINE LEARNING FOR AUTOMATIC EXTRACTION OF WATER BODIES USING SENTINEL-2 IMAGERY","authors":"V. Kashtan, V. Hnatushenko","doi":"10.15588/1607-3274-2024-1-11","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-11","url":null,"abstract":"Context. Given the aggravation of environmental and water problems, there is a need to improve automated methods for extracting and monitoring water bodies in urban ecosystems. The problem of efficient and automated extraction of water bodies is becoming relevant given the large amount of data obtained from satellite systems. The object of study is water bodies that are automatically extracted from Sentinel-2 optical satellite images using machine learning methods. \u0000Objective. The goal of the work is to improve the efficiency of the process of extracting the boundaries of water bodies on digital optical satellite images by using machine learning methods. \u0000Method. The paper proposes an automated information technology for delineating the boundaries of water bodies on Sentinel-2 digital optical satellite images. The process includes eight stages, starting with data download and using topographic maps to obtain basic information about the study area. Then, the process involved data pre-processing, which included calibrating the images, removing atmospheric noise, and enhancing contrast. Next, the EfficientNet-B0 architecture is applied to identify water features, facilitating optimal network width scaling, depth, and image resolution. ResNet blocks compress and expand channels. It allows for optimal connectivity of large-scale and multi-channel links across layers. After that, the Regional Proposal Network defines regions of interest (ROI), and ROI alignment ensures data homogeneity. The Fully connected layer helps in segmenting the regions, and the Fully connected network creates binary masks for accurate identification of water bodies. The final step of the method is to analyze spatial and temporal changes in the images to identify differences, changes, and trends that may indicate specific phenomena or events. This approach allows automating and accurately identifying water features on satellite images using machine learning. \u0000Results. The implementation of the proposed technology is development through Python software development. An assessment of the technology’s accuracy, conducted through a comparative analysis with existing methods, such as water indices and K-means, confirms a high level of accuracy in the period from 2017 to 2023 (up to 98%). The Kappa coefficient, which considers the degree of consistency between the actual and predicted classification, confirms the stability and reliability of our approach, reaching a value of 0.96. \u0000Conclusions. The experiments confirm the effectiveness of the proposed automated information technology and allow us to recommend it for use in studies of changes in coastal areas, decision-making in the field of coastal resource management, and land use. Prospects for further research may include new methods that seasonal changes and provide robustness in the selection and mapping of water surfaces.","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"50 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140755101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"REFINEMENT AND ACCURACY CONTROL OF THE SOLUTION METHOD FOR THE DURABILITY PROBLEM OF A CORRODING STRUCTURE USING NEURAL NETWORK","authors":"O. D. Brychkovskyi","doi":"10.15588/1607-3274-2024-1-9","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-9","url":null,"abstract":"Context. The prediction of the time until failure of corroding hinge-rod structures is a crucial component in risk management across various industrial sectors. An accurate solution to the durability problem of corroding structures allows for the prevention of undesired consequences that may arise in the event of an emergency situation. Alongside this, the question of the effectiveness of existing methods for solving this problem and ways to enhance them arises. \u0000Objective. The objective is to refine the method of solving the durability problem of a corroding structure using an artificial neural network and establish accuracy control. \u0000Method. To refine the original method, alternative sets of input data for the artificial neural network which increase information about the change in axial forces over time are considered. For each set of input data a set of models is trained. Based on target metric values distribution among the obtained sets, a set is selected where the minimum value of the mathematical expectation of the target metric is achieved. For the set of models corresponding to the identified best set, accuracy control of the method is determined by establishing the relationship between the mathematical expectation of the target metric and the parameters of the numerical solution. \u0000Results. The conditions under which a lower value of the mathematical expectation of the target metric is obtained compared to the original method are determined. The results of numerical experiments, depending on the considered case, show, in average, an improvement on 43.54% and 9.67% in the refined method compared to the original. Additionally, the proposed refinement reduces the computational costs required to find a solution by omitting certain steps of the original method. An accuracy control rule of the method is established, which allows to obtain on average a given error value without performing extra computations. \u0000Conclusions. The obtained results indicate the feasibility of applying the proposed refinement. A higher accuracy in predicting the time until failure of corroding hinge-rod structures allows to reduce the risks of an emergency situation. Additionally, accuracy control enables finding a balance between computational costs and the accuracy of solving the problem. KEYWORDS","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"84 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140752538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LAMA-WAVELET: IMAGE IMPAINTING WITH HIGH QUALITY OF FINE DETAILS AND OBJECT EDGES","authors":"D. O. Kolodochka, M. V. Polyakova","doi":"10.15588/1607-3274-2024-1-19","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-19","url":null,"abstract":"Context. The problem of the image impainting in computer graphic and computer vision systems is considered. The subject of the research is deep learning convolutional neural networks for image inpainting. \u0000Objective. The objective of the research is to improve the image inpainting performance in computer vision and computer graphics systems by applying wavelet transform in the LaMa-Fourier network architecture. \u0000Method. The basic LaMa-Fourier network decomposes the image into global and local texture. Then it is proposed to improve the network block, processing the global context of the image, namely, the spectral transform block. To improve the block of spectral transform, instead of Fourier Unit Structure the Simple Wavelet Convolution Block elaborated by the authors is used. In this block, 3D wavelet transform of the image on two levels was initially performed using the Daubechies wavelet db4. The obtained coefficients of 3D wavelet transform are splitted so that each subband represents a separate feature of the image. Convolutional layer, batch normalization and ReLU activation function are sequentially applied to the results of splitting of coefficients on each level of wavelet transform. The obtained subbands of wavelet coefficients are concatenated and the inverse wavelet transform is applied to them, the result of which is the output of the block. Note that the wavelet coefficients at different levels were processed separately. This reduces the computational complexity of calculating the network outputs while preserving the influence of the context of each level on image inpainting. The obtained neural network is named LaMa-Wavelet. The FID, PSNR, SSIM indexes and visual analysis were used to estimate the quality of images inpainted with LaMa-Wavelet network. \u0000Results. The proposed LaMa-Wavelet network has been implemented in software and researched for solving the problem of image inpainting. The PSNR of images inpainted using the LaMa-Wavelet exceeds the results obtained using the LaMa-Fourier network for narrow and medium masks in average by 4.5%, for large masks in average by 6%. The LaMa-Wavelet applying can enhance SSIM by 2–4% depending on a mask size. But it takes 3 times longer to inpaint one image with LaMa-Wavelet than with LaMa-Fourier network. Analysis of specific images demonstrates that both networks show similar results of inpainting of a homogeneous background. On complex backgrounds with repeating elements the LaMa-Wavelet is often more effective in restoring textures. \u0000Conclusions. The obtained LaMa-Wavelet network allows to improve the image inpainting with large masks due to applying wavelet transform in the LaMa network architecture. Namely, the quality of reconstruction of image edges and fine details is increased.","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"96 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140754752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"METHOD OF EVALUATION THE EFFICIENCY OF FIBER-OPTIC CABLES MODELS WITH MULTI-MODULAR DESIGN BASED ON MASS AND DIMENSIONAL INDICATORS","authors":"O. V. Bondarenko, D. M. Stepanov, O. O. Verbytskyi, S. V. Siden","doi":"10.15588/1607-3274-2024-1-1","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-1","url":null,"abstract":"Context. Today, the leading cable production plants in many countries of the world manufacture single- and multi-module designs of fiber-optic cables (FOC) with different protective covers and the number of fibers. This creates a wide range of possible FOC models for different consumer (buyer) requirements. However, the lack of openness of prices for FOC for the consumer, in particular for the project organization, and the desire of the manufacturer to save on production creates a need for the development and research of a method for evaluating the effectiveness of FOC of a multi-module design. In the work, it is proposed to do this by analyzing a number of optical cable models according to parameters-criteria – the compactness coefficient and the economic efficiency coefficient of the FOC by diameter.\u0000Objective. To develop a method of evaluation the efficiency of fiber-optic cables models with multi-modular design based on mass-dimensional indicators, which will allow to quickly choose an appropriate model of the FOC with the given initial data.\u0000Method. A method of evaluating the efficiency of the FOC modular design has been developed and proposed. It is based on the comparison of cable models and the selection of the most appropriate of them at given initial data. The paper proposed and introduced parameters-criteria for this – the compactness coefficient υ and the efficiency coefficient according to the diameter of the cable E0 – which show the connection of the design characteristics of the FOC to a certain parameter of its structure. At the same time, the most effective model (design) of the FOC compared to the basic models in terms of technical conditions, both from the point of view of the manufacturer and the customer, consists in lower material costs for its production while simultaneously ensuring the specified requirements for the cable (first of all, number of fibers and mechanical strength). This will allow, at the stage of cable design, to make an appropriate choice of its model with given initial parameters and to develop such a design of the FOC, which will allow to minimize the dimensions (and therefore the material capacity and cost) of the model without losing its quantitative and qualitative characteristics.\u0000Results. The paper presents the results of the development and research of the method of evaluating the efficiency of the FOC with multi-module structure based on mass and dimensional indicators. For example, using the developed method, it is shown that it is possible to choose a FOC model with a diameter smaller by 10.9% and save 15.5% of the cable cost for each kilometer of the fiber optic communication line while ensuring the initial requirements for the cable.\u0000Conclusions. The scientific novelty of the work results is that, method of evaluation the efficiency of a modular FOC design has been first developed, which allows at the cable design stage to compare the model with the cable design according to the technical con","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"26 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140753672","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SCATTERING OF ELECTROMAGNETIC WAVES ON FLAT GRID TWO-PERIODIC STRUCTURES","authors":"V. A. Vanin, I. I. Pershyna","doi":"10.15588/1607-3274-2024-1-4","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-4","url":null,"abstract":"Context. One of the scientific hypotheses for the creation of nonreciprocal optical metasurfaces is based on the use of a wave channel in which rays of the direct and reverse diffraction scenarios are realized on two-periodic flat structures with nonlinear elements. Such processes in the nanometer wavelength range of electronic devices require precise calculations of the interaction of waves and microstructures of devices. It is also important to describe the behavior of antenna devices in mobile communications. Expanding the wavelength range of stable communication is achieved by using prefractal structures in antenna devices in combination with periodic structuring. Similar modeling problems arise when electromagnetic waves penetrate materials with a crystalline structure (radio transparency). \u0000Objective. To test this hypothesis, it is necessary to carry out mathematical modeling of the process of scattering of electromagnetic waves by metasurfaces under conditions of excitation of several diffraction orders. It is known that among two-periodic flat lattices of different structures there are five types that fill the plane. These are the Bravais grilles. The problem of scattering of an incident monochromatic TE polarized wave on a metal screen with recesses in two-periodic structures filled with silicon was considered. \u0000Method. The paper builds mathematical models for the study of spatial-amplitude spectra of metasurfaces on Brave lattices and gives some results of their numerical study. The condition for determining the diffraction orders propagating over the grating is proposed. Scattered field amplitudes are from the solution of the boundary value problem for the Helmholtz equation in the COMSOL Multiphysics 5.4 package. Similar problem formulations are possible when studying the penetration of an electromagnetic field into a crystalline substance. \u0000Results. Obtained relations for diffraction orders of electromagnetic waves scattered by a diffraction grating. The existence of wavelengths incident on a two-periodic lattice for which there is no reflected wave is shown for different shapes (rectangular, square, hexagonal) of periodic elements in the center of which a depression filled with silicon was made. Distributions of reflection coefficients for different geometric sizes of colored elements and recesses are given. The characteristics of the electric field at resonant modes in the form of modulus isolines show the nature of the interaction of the field over the periodic lattice and the scatterersdepressions. At the resonant wavelengths of the incident waves, standing waves appear in the scatterers. \u0000Conclusions. A mathematical model of the set of diffraction orders propagating from a square and hexagonal lattice into halfspace is proposed z >= 0. It has been shown that flat periodic lattice with square or hexagonal periodicity elements and resonant scatterers in the form of cylindrical recesses filled with silicon can produce a non-mirrore","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"25 17","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140753795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DECISION-MAKING MODELS AND THEIR APPLICATION IN TRANSPORT DELIVERY OF BUILDING MATERIALS","authors":"A. M. Bashkatov, O. A. Yuldashova","doi":"10.15588/1607-3274-2024-1-5","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-5","url":null,"abstract":"Context. The task of determining a generalized parameter characterizing a comprehensive assessment of the action of criteria affecting the sequence of execution of orders for the manufacture and delivery of products to the customer. \u0000Objective. The purpose of the work is to develop an algorithm for calculating priorities when solving the problem of transport services in conditions of uncertainty of choice. \u0000Method. When considering the problem of the efficiency of order fulfillment, the reasons are given that affect the efficiency of the tasks being solved for the delivery of paving slabs to the customer in the shortest possible time. In order to select a scheme that reflects the main stages of decision-making, a justification was carried out and a comparative analysis of existing models was carried out. The criteria for the requirements for describing such models have been determined. It is indicated that the objective function depends on a group of reasons, i.e. represents a composite indicator. The stochastic nature of such factors led to the use of statistical analysis methods for their assessment. The limits of variation of the parameters used in the calculations are established. The solution to the multicriteria problem consists in bringing the role of the acting factors to one unconditional indicator, grouping and subsequent ranking of their values. The decision-making and the choice of the indicator will depend on the set threshold and the priority level of the factor. The indices that form the priority of the factor are determined analytically or expertly. The sequence of actions performed is presented in the form of an algorithm, which allows automating the selection of a model and the calculation of indicators. To assess the adequacy of the proposed solutions, tables of comparative results for the selection of the priority of the executed orders are given. \u0000Results. The method allows a comprehensive approach to taking into account the heterogeneous factors that determine the order in which the order is selected when making managerial decisions, ensuring the achievement of a useful effect (streamlining the schedule for the delivery of paving slabs to the customer) by ranking the values of priority indices. \u0000Conclusions. The proposed scheme for the transition to a complex unconditional indicator (priority index) makes it possible to quantitatively substantiate the procedure for choosing the next order when performing work. A special feature is that the list of operating factors can be changed (reduced or supplemented with new criteria). The values of these parameters will improve and have a higher reliability with the expansion of the experimental design, depending on the retrospective of their receipt, the accuracy of the data. As a prospect of the proposed method, the optimization of the process of selecting applications using queuing methods (for the type of the corresponding flow – homogeneous, without consequences, stationary, gamma","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"26 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140752738","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DESIGN MODELS OF BIT-STREAM ONLINE-COMPUTERS FOR SENSOR COMPONENTS","authors":"L. Larchenko, A. V. Parkhomenko, B. D. Larchenko, V. R. Korniienko","doi":"10.15588/1607-3274-2024-1-6","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-6","url":null,"abstract":"Context. Currently, distributed real-time control systems need the creation of devices that perform online computing operations close to the sensor. The proposed online-computers of elementary mathematical functions can be used as components for the functional conversion of signals in the form of pulse streams received from measuring sensors with frequency output. \u0000Objective. The objective of the study is the development of mathematical, architectural and automata models for the design of bit-stream online-computers of elementary mathematical functions in order to create a unified approach to their design, due to which the accuracy of calculating functions can be increased, functional capabilities expanded, hardware costs reduced, and design efficiency increased. \u0000Method. Mathematical models of devices were developed using the method of forming increments of ascending step functions based on inverse functions with minimization of calculation error. Automata models of online-computers based on Moore’s Finite State Machine have been developed, the graph diagrams of which made it possible to ensure the clarity of function implementation algorithms, to increase visibility and invariance of implementation in formal languages of programming and hardware description. \u0000Results. The paper presents the results of research, development and practical approbation of design models of bit-stream onlinecomputers of power functions and root extraction function. A generalized architecture of an online-computer was proposed. \u0000Conclusions. The considered functional online-computers are effective from the point of view of calculation accuracy, simplicity of technical implementation, and universality of the architecture","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"16 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140753974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TWO-FRAGMENT NON-LINEAR-FREQUENCY MODULATED SIGNALS WITH ROOTS OF QUADRATIC AND LINEAR LAWS FREQUENCY CHANGES","authors":"O. O. Kostyria, A. A. Нryzo, H. V. Khudov, O. M. Dodukh, B. А. Lisohorskyi","doi":"10.15588/1607-3274-2024-1-2","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-2","url":null,"abstract":"Context. The rapid development of the technology of digital synthesis and processing of radar signals, which has been observed in recent decades, has practically removed restrictions on the possibility of implementing arbitrary laws of frequency modulation of radio oscillations. Along with the traditional use of linearly-frequency-modulated signals, modern radar means use probing signals with non-linear frequency modulation, which provide a lower level of maximum side lobes and a higher rate of their descent. These factors, in turn, contribute to improving the detection characteristics of targets under conditions of passive interference, as well as increasing the probability of detecting small targets against the background of targets with larger effective scattering surfaces. In this regard, a large number of studies are conducted in the direction of further improvement of existing and synthesis of radar signals with new laws of frequency modulation. The use of multifragment nonlinear-frequency-modulated signals, which include fragments with both linear and nonlinear modulation, provides an increase in the number of possible versions of the laws of frequency modulation and synthesis of signals with predicted characteristics. Synthesis of new multifragment signals with a reduced level of side lobes of autocorrelation functions and a higher rate of their descent is an important scientific and technical task, the solution of which is devoted to this article. \u0000Objective. The purpose of the work is to develop mathematical models of the current and shifted time of two-fragment nonlinear-frequency modulated signals for the case when the first fragment has a root-quadratic, and the second linear frequency modulation and determine the feasibility of using such a signal in radar applications. \u0000Method. The article theoretically confirms that for the mathematical model of the current time, when moving from the first fragment to the second at the junction of fragments, jumps of instantaneous frequency and phase (or only phases for the mathematical model of shifted time) occur, which can significantly distort the resulting signal. Determination of value of frequency-phase jumps for their further elimination is performed by finding difference between value of initial phase of second fragment and final value of phase of first fragment. A distinctive feature of the developed mathematical models is the use of the first fragment of the signal with root-quadratic, and the second – linear frequency modulation. \u0000Results. Comparison of the signal, the first fragment of which has root-square frequency modulation, and the signal with two linearly-frequency-modulated fragments, provided that the total duration and frequency deviation are equal, shows that for the new synthesized signal the maximum level of side lobes decreased by 1.5 dB, and their rate of decay increased by 6.5 dB/dec. \u0000Conclusions. A new two-fragment signal was synthesized, the first fragment of which h","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"47 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140755273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"UA-LLM: ADVANCING CONTEXT-BASED QUESTION ANSWERING IN UKRAINIAN THROUGH LARGE LANGUAGE MODELS","authors":"M. V. Syromiatnikov, V. M. Ruvinskaya","doi":"10.15588/1607-3274-2024-1-14","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-14","url":null,"abstract":"Context. Context-based question answering, a fundamental task in natural language processing, demands a deep understanding of the language’s nuances. While being a sophisticated task, it’s an essential part of modern search systems, intelligent assistants, chatbots, and the whole Conversational AI field. While English, Chinese, and other widely spoken languages have gathered an extensive number of datasets, algorithms, and benchmarks, the Ukrainian language, with its rich linguistic heritage and intricate syntax, has remained among low-resource languages in the NLP community, making the Question Answering problem even harder. \u0000Objective. The purpose of this work is to establish and benchmark a set of techniques, leveraging Large Language Models, combined in a single framework for solving the low-resource problem for Context-based question-answering task in Ukrainian. \u0000Method. A simple yet flexible framework for leveraging Large Language Models, developed as a part of this research work, enlights two key methods proposed and evaluated in this paper for dealing with a small amount of training data for context-based question-answering tasks. The first one utilizes Zero-shot and Few-shot learning – the two major subfields of N-shot learning, where N corresponds to the number of training samples, to build a bilingual instruction-based prompt strategy for language models inferencing in an extractive manner (find an answer span in context) instead of their natural generative behavior (summarize the context according to question). The second proposed method is based on the first one, but instead of just answering the question, the language model annotates the input context through the generation of question-answer pairs for the given paragraph. This synthetic data is used for extractive model training. This paper explores both augmentation-based training, when there is some annotated data already, and completely synthetic training, when no data is available. The key benefit of these two methods is the ability to obtain comparable prediction quality even without an expensive and long-term human annotation process. \u0000Results. Two proposed methods for solving the low-to-zero amount of training data problem for context-based questionanswering tasks in Ukrainian were implemented and combined into the flexible LLM experimentation framework. \u0000Conclusions. This research comprehensively studied OpenAI GPT-3.5, OpenAI GPT-4, Cohere Command, and Meta LLaMa-2 language understanding capabilities applied to context-based question answering in low-resource Ukrainian. The thorough evaluation of proposed methods on a diverse set of metrics proves their efficiency, unveiling the possibility of building components of search engines, chatbot applications, and standalone general-domain CBQA systems with Ukrainian language support while having almost zero annotated data. The prospect for further research is to extend the scope from the CBQA task evaluated in this paper to all ","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"39 17","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140752372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"RESEARCH OF THE FEATURES OF DIGITAL SIGNAL FORMATION IN SATELLITE COMMUNICATION LINES","authors":"V. І. Мagro, O. G. Panfilov","doi":"10.15588/1607-3274-2024-1-3","DOIUrl":"https://doi.org/10.15588/1607-3274-2024-1-3","url":null,"abstract":"Context. Remote sensing of the Earth is now widely used in various fields. One of the challenges of remote sensing is the creation of inexpensive satellite systems operating in polar circular orbits. These systems require the development of a receptiontransmission system that allows tens of gigabits of video information to be transmitted to an earth receiving station within ten minutes. That is, there is a need to create a communication system that provides high speed data transmission from small satellites weighing up to 50 kg. \u0000Objective. The aim of the work is to study the features of digital signal formation in modern satellite communication lines and to develop a communication system with a high data transfer rate (usually 300 Mbit/s), which can be applied to small Earth Observation satellites. \u0000Method. Proposed concept for building a high-speed data transmitter from a remote sensing earth satellite using commercially available off-the-shelf technology. Calculations of the power flow density created at the input of the receiving earth station were performed to find out the possible power of the on-board transmitter. \u0000Results. A diagram of a communication system based on the DVB-S standard using the technology of commercially available off-the-shelf products has been developed. The high-speed data transmitter is implemented on a Xilinx® Zynq Ultrascale+ ™ MPSoC FPGA microchip, which is located on an Enclustra Mercury XU8 module with a high-performance dual 16-bit AD9174 DAC. The on-board transmitter with a power of up to 2 W meets the requirements of the ITU Radio Regulations for the power flux density on the surface of the Earth, which is created by the radiation of the space station EESS in the range 8025–8400 MHz. It is shown that the energy reserve of the communication line of 3 dB is achieved for various commands for coding and modulation changes with an increase in the elevation angle, which allows to increase the speed of information transmission. \u0000Conclusions. An original receiving-transmitting system was developed for use in small satellites for remote sensing of the Earth. It is shown that the function of adaptive modeling of ACM of the DVB-S standard allows you to automatically change the transmission parameters in real time depending on the changing conditions of the channel, providing opportunities for more flexible and effective data transmission in various conditions, which will allow to increase the volumes of information transmitted by communication session. The proposed system operates in the X-band and is built using commercially available off-the-shelf products. An antenna with double circular polarization is used as the emitter. Two physical channels represent two polarization modes: right circular polarization and left circular polarization, each of which has three frequency channels.","PeriodicalId":518330,"journal":{"name":"Radio Electronics, Computer Science, Control","volume":"28 18","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140753162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}