Journal influence
Higher Attestation Commission (VAK) - К1 quartile
Russian Science Citation Index (RSCI)
Bookmark
Next issue
№4
Publication date:
09 September 2024
Journal articles №3 2024
11. Accelerating criterion function calculation in an omnidirectional antenna placement problem [№3 за 2024 год]Authors: Aye Min Thike (ayeminthike52@gmail.com) - National Research University of Electronic Technology, MIET (Doctoral Student), Ph.D; Lupin, S.A. (lupin@miee.ru) - National Research University of Electronic Technology, MIET, Joint Supercomputer Center of RAS (Professor), Ph.D; P.N. Telegin (pnt@jscc.ru) - Joint Supercomputer Center of RAS (Leading Researcher), Ph.D; Shabanov, B.M. (jscc@jscc.ru) - Joint Supercomputer Center of RAS (Corresponding Member of the RAS, Director), Ph.D;
Abstract: The paper presents a study of the tabular effectiveness method for calculating one of the criteria when solving the problem of the optimal solution for placing a set of omnidirectional antennas. The problem of the active network elements placement belongs to multicriteria discrete optimization problems. The coverage area is one of the main criteria of the solution optimality. This parameter determines both the availability of network services for subscribers and operator's costs for creating and maintaining a network infrastructure. In order to solve the antenna placement problem, this paper proposes using a brute-force algorithm (BFA), which can find an exact solution to discrete optimization problems. The computational complexity of BFA depends on the size of the solution search space and on the complexity of calculating criterion functions. The experiments show that the tabular method is 17 times faster than direct calculation of the criterion value that ensures no overlaps of reception areas. The tabular method for the criterion function calculation is also effective for parallel BFA implementation. When running the application on 12 threads, computation time decreased more than sixfold. The substitution of direct computation of a criterion value on pre-calculated tabulated values can be also effective in other discrete optimization problems.
Keywords: brute-force algorithm, computational complexity, parallel computing, maximum coverage problem, omnidirectional antenna
Visitors: 1531
12. Visual quality assessment of the pseudorandom number generator for solving cryptographic tasks [№3 за 2024 год]
Authors: Tatarnikova, T.M. (tm-tatarn@yandex.ru) - St. Petersburg State University of Aerospace Instrumentation (Associate Professor, Professor), Ph.D; Bulgakov, D.A. (dmbulg@gmail.com) - Saint Petersburg State University of Aerospace Instrumentation (Senior Lecturer), ;
Abstract: The paper focuses on studying the methods and tests for evaluating various random number generators (RNG) and developing an application for a rough visual evaluation of RNG. It describes RNG types and key characteristics. It also considers mathematical methods and software test packages designed for evaluating RNG, such as the US National Institute of Standards and Technology (NIST), Dieharder, PractRand, ENT and RaBiGeTe tests. The paper shows an alternative way to perform a fast approximate evaluation of generation quality using visualization. The essence of thу proposed solution is, first, to divide each integer into three parts, 8 bits each. Second, to convert the resulting values into RGB colors, which are also encoded in three bytes. The rendering engine functionality allows generating a bitmap texture - a matrix of pixels from the obtained numbers and overlying this texture on a 2D-plane. Implementing this procedure involves using the Unity game engine. The application also provides a tool for generating sequences of 65,536 positive integers using four different random number generation algorithms: System Random from Microsoft, Unity Random from the developers of the Unity game engine, the standard Rand algorithm of C language, and Lehmer's algorithm. The resulting visualization allows the user to detect the presence of various repetitive structures or artifacts in the initial set of numbers easily. The work of the visualizer tests both on obviously non-random sets of numbers and on truly random numbers obtained from a quantum generator. In the future, the authors propose using visualization for initial quick assessment of random number generation results in a blockchain network.
Keywords: random number generator, entropy, random number generator testing, visualization of random number generator operation, visualization application
Visitors: 1392
13. Components and functionality of a software tool for modelling an import substitution structure [№3 за 2024 год]
Authors: Abu-Abed, F.N. (aafares@mail.ru) - Tver State Technical University (Associate Professor, Dean), Ph.D; Zhironkin, S.A. (zhironkin@tpu.ru) - National Research Tomsk Polytechnic University (Professor), Ph.D;
Abstract: The subject of this research is creating the main components of a tool for modelling an import substitution structure in the Russian economy and determining its functionality. The research is relevant due to the lack of software analogues and developments in the domestic IT market to solve the task. The authors apply the methods of system analysis, simulation modelling, visual design and object-oriented programming as research methods. Modelling the output of products for the domestic market requires finding the optimal combination of the volume values of production factors (capital and labour) and their efficiency. For this purpose, it is necessary to create a software tool that reflects a mathematical model of import substitution based on its target function and links various classes, databases and user interface. This paper summarizes the main classes, each of which plays a certain role in calculating and modelling of import substitution, explores their interaction presented as a diagram.The developed application executes commands and procedures through class methods; there is an algorithm of user interaction with the application. The proposed application database includes tables of variables necessary for determining the target function values (the number of industries and products, arrays of the cost of fixed capital and the number of employees, changes in stock returns and labour productivity, etc.) and results (the desired sectoral volumes of investment, employment and output). The application interface allows the user to calculate the values of the target function of an import substitution structure by using buttons that open dialogue windows for entering values and starting calculations. The scientific novelty consists in obtaining results that allow optimizing the solution of an important problem in the context of national economy regulation – in modelling an import substitution structure under the conditions of increasing external shocks and technological constraints.
Keywords: import substitution, database, commands and procedures, user interface, classes, structure, software, mathematical model
Visitors: 1535
14. Computer modeling of complex interactions of granular material distinct elements using functionally defined objects [№3 за 2024 год]
Authors: Vyatkin, S.I. (sivser@mail.ru) - Institute of Automation and Electrometry, Siberian Branch, Russian Academy of Sciences, Synthesizing Visualization Systems Laboratory (Senior Researcher), Ph.D; Dolgovesov, B.S. (bsd@iae.nsk.su ) - Institute of Automation and Electrometry, Siberian Branch, Russian Academy of Sciences, Synthesizing Visualization Systems Laboratory (Head of the Laboratory), Ph.D;
Abstract: The paper proposes a method of contact dynamics for granular materials using functionally specified objects. It also considers the problem of modeling complex interactions of a large number of individual elements of granular materials using information about the mass, location and contacts of contacting surfaces to obtain linear complexity based on the total number of contacts detected at each iteration. The proposed method of contact dynamics is based on numerical modeling with access to all variables, on non-smoothness in space, time and force. That is, an area bounded by a set of surfaces is a valid set in the space of configuration variables. In case of colliding, the time function representing velocity will be intermittent. Friction is determined by the ratio between a reaction force and a local relative velocity. The paper proposed a velocity calculation method for modeling large sets of solids. Functionally defined objects, in particular ellipsoids, are used as separate elements of granular materials. The method has no repeated pairwise comparisons between objects. Therefore, there is linear complexity in terms of the total number of contacts detected at each iteration. The devel-oped model of friction in a configuration space of solids is based on the maximum dispersion principle. As a result, the model combines rolling and sliding friction (while providing the expected characteristics) with calculating the permissible speed and reaction to friction for all objects. The efficiency of the method is confirmed by computer simulation. Experiments have shown that a collision can split into several separate episodes of contact on a microscopic time scale. This leads to the fact that a collision outcome depends heavily on a shape of corresponding bodies. When individual elements of granular materials have an ellipsoid shape, the result is more acceptable. The proposed method of contact dynamics might be useful for calculating quasi-static evolutions of granular materials in mechanical engineering, construction, numerical processing using finite elements of dynamic and quasi-static problems associated with deformable bodies, etc.
Keywords: friction, contact, dynamic interaction, computer modeling, functionally defined objects
Visitors: 1361
15. Real-time recognition and tracking of road surface defects based on complex using of standard computational procedures and deep neural networks [№3 за 2024 год]
Authors: Antonov, M.O. (amo.mikhail@gmail.com) - University of Science and Technology MISIS (Assistant), ; I.O. Temkin (igortemkin@yandex.ru) - National University of Scince Research Technological "MISIS" (Professor, Head of Chair), Ph.D;
Abstract: The paper considers several issues related to real-time recognition and tracking of road surface defects. This is an important task for modern monitoring systems, unmanned transportation and road infrastructure management. The proposed system identifies pavement defects using deep neural networks and tracks them by assigning unique identifiers. The YOLOv8 model has demonstrated high accuracy and speed in object detection tasks during defect recognition. Due to training the model on composite datasets, generalization was reliable. A notable feature of this approach is the absence of strict requirements to the technical characteristics of cameras and other system components. This makes the solution flexible and adaptable to various operating conditions. Defect tracking is based on place using the Hungarian algorithm and Kalman filter. This ensures high accuracy and reliability of the observation process. The Hungarian algorithm efficiently establishes correspondences between detections in consecutive frames. The Kalman filter smooths trajectories and predicts future defect positions. This is crucial for accurate assignment of unique identifiers and data analysis. Model training parameters, estimates of computational resource consumption and system performance show that this solution has acceptable performance for real-time applications. These measurements help to estimate implementation costs and determine the potential for integration into existing monitoring systems. Thus, the presented solution provides highly efficient and accurate real-time recognition and tracking of pavement defects. It improves the safety and quality of road infrastructure and simplifies management and maintenance planning.
Keywords: surface defects, Kalman filter, Hungarian algorithm, YOLOv8, object tracking, deep neural networks, real-time recognition
Visitors: 1346
16. Dynamic generation of a user interface for a medical mobile application based on an ontological approach [№3 за 2024 год]
Authors: Gribova, V.V. (gribova@iacp.dvo.ru) - Institute of Automation and Control Processes Far Eastern Branch of RAS, Ph.D; Lifshits, A.Ya. (mmvb@iacp.dvo.ru) - Institute of Automation and Control Processes of the Far Eastern Branch of the Russian Academy of Sciences (Leading Engineer-Programmer), Ph.D; F.M. Moskalenko (philipmm@iacp.dvo.ru) - Institute of Automation and Control Processes Far Eastern Branch of RAS (Senior Researcher), Ph.D; E.A. Shalfeeva (shalf@iacp.dvo.ru) - Institute of Automation and Control Processes Far Eastern Branch of RAS (Associate Professor, Senior Researcher), Ph.D; Shevchenko, N.E. (shev.nikita@mail.ru) - Vladivostok State University, ;
Abstract: Medical decision support systems are designed to help doctors make timely and correct decisions. This is particularly important to use such systems remotely from medical centers and clinics, at the patient’s bedside. The structure of an electronic medical record has many hierarchically presented sections, and input information for such records is assessed by tens of thousands of possible terms and their meanings. Taking this into account, it is relevant to develop new methods for generating mobile user interface for filling out electronic medical records. This paper presents a method for generating an adaptive interface for filling out patient’s electronic record in mobile applications based on an ontological approach. The approach assumes visualization by vertical layers selected by node levels in the hierarchical semantic structure of an electronic medical record document. The described generation technology is tested on a diagnostic mobile shell. The result is also a new technology of generating user interface and a software shell for combining diagnostic services for arbitrary groups of diseases for mobile devices. The practical significance of the technology consists in flexibility of the presented interface generation method: it can be used to create adaptive user interfaces in other subject areas and tasks, which are characterized by large volumes of input complex information.
Keywords: mobile application, knowledge base, medicine, user interface, dynamic interface generation, ontology, intelligent systems
Visitors: 1561
17. Automation of semantic analysis for textual responses of students in a digital educational platform [№3 за 2024 год]
Authors: Leonov, A.G. (dr.l@math.msu.su) - M.V. Lomonosov Moscow State University, Federal State Institution SRISA RAS, Moscow State Pedagogical University, State University of Management (Associate Professor, Professor, Head of Chair, Leading Researcher), Ph.D; Martynov, N.S. (nikolai.martynov@math.msu.ru) - Federal State Institution SRISA RAS (Engineer); Mashchenko, K.A. (kirill010399@vip.niisi.ru) - M.V. Lomonosov Moscow State University, Federal State Institution SRISA RAS, State University of Management (Junior Researcher); Kholkina, A.A. (kholkina.a2021@gmail.com) - National University of Scince Research Technological "MISIS" (Student); Shlyakhov, A.V. (shlyakhov@vip.niisi.ru) - M.V. Lomonosov Moscow State University (Junior Researcher);
Abstract: The purpose of this study is to develop and apply a system of automated checking of students' answers to open-ended text questions within the framework of a digital educational platform. The developed system enables significant reducing of teacher workload and minimizing both type I and type II errors during evaluating students' answers. The research methodology involves modern methods of information extraction, text analysis and concept comparison. Con-structing a sentence-parsing graph forms syntactic trees. After that, taking into account the peculiarities of the Russian language, the relations between their nodes are analyzed. For example, there might be double negation elimination, named entity extraction and lexical relationship recognition. To compare trees with each other, the authors use the search for the optimal edit path implemented using dynamic programming. The result of the research is a system that is able to evaluate students' answers taking into account the semantic depth of texts’ content, which can ensure objectivity and transparency of knowledge evaluation. In addition, it reduces teacher workload by automating the process of checking answers. The possibility of parametric setting of the system operation scenario helps to adjust it to specific requirements to the answers set by a teacher. The system is able to ignore grammatical errors, as it focuses on tracking differences or similarities in the semantic component of texts. The practical significance of the work is in the possibility of effective and fair control of students' knowledge taking into account a text semantic component, in reducing teachers’ routine and accelerating the process of evaluating students' answers. Also, it adapts to teachers’ specific requirements and provides prompt feedback. This study is relevant and significant for educational practice, as it corresponds to modern educational trends and techno- logies.
Keywords: automated verification, educational technology, response evaluation system, semantic analysis, semantic analysis, Mirera digital learning platform
Visitors: 1304
◄ ← Preview | 1 | 2