Software & Systems

Articles of journal № 1 at 2021 year.

Order result by: Public date | Title | Authors

11. Network anomalies detection by deep learning [№1 за 2021 год]
Author: V.N. Zuev
Visitors: 5447
The paper discusses the machine learning application for detecting anomalies in network traffic. Artifi-cial neural networks of deep learning are used as a tool. In this paper, the NSL-KDD data set is ana-lyzed and used to study the effectiveness of deep learning neural networks in detecting anomalies in network traffic patterns. The most important aspects of this dataset are the imbalanced class distribu-tion. The paper describes the method of effective usage of objective functions backpropagation algo-rithms in order to train the neural network on imbalanced samples. Using the backpropagation algo-rithm is connected with many difficulties. The major problem is the ability to generalize the neural network. The ability to generalize is the most important characteristic of a neural network. It is mean that trained on studying data neural network is capable to produce output value by using unknown da-ta. However, using for training noisy data decreases the ability to generalize the neural network. The proposed method makes it possible to more efficiently calculate the value of the aim function, which is the basis of the error back-propagation algorithm. The method is well fit for the heterogeneous sample and can use priority information about the sample’s significance. The pepper described an algo-rithm of the method. Using this method will improve the accuracy of the neural network for classifica-tion and regression problems. The experimental result shows that it well suits the designed method for network anomaly detec-tions.

12. Ontology processing in attributive access control in cyber-physical systems [№1 за 2021 год]
Author: Poltavtseva M.A.
Visitors: 3450
The paper is devoted to supporting the processing of large-scale ontologies in a relational server and considers a separate problem of representing and processing ontologies when implementing attributive (ontological) access in cyber-physical systems. The relevance of the paper is due to the attack growth on industrial cyber-physical systems and the improvement of access control methods. The most promising direction today is attributive access based on ontologies. On the one hand, distributed large-scale industrial cyber physical systems use a large and increasing number of rules for attributive access control, on the other hand, storage techniques and processing such data using specialized technologies must meet the requirements for information pro-tection. This leads to the necessity of applying advanced (including certified) tools and necessitates the use of a relational server for storing and processing data. Therefore, the searching problem of the most rational representation and processing of access control rules is highly relevant. The paper proposes a method for representing the rules of ontological inference based on the impli-cations of binary trees to support ontologies in the problem of attributive access control of cyber phys-ical systems. There is a data representation, and analysis of methods for displaying information in an industrial relational server. An example of support for access control rules shows experimental testing of the representation of ontological inference rules based on the implications of binary trees. Because of analytical effort and experimental testing, the most rational solution for this problem is to use the storage technique for a forest of trees based on a materialized path.

13. The quality control of training equipment [№1 за 2021 год]
Authors: Ilin V.А., Pakhomov E.S.
Visitors: 3951
Using training tools requires an assessment of their effectiveness to achieve the goals of training and training. The effectiveness of training equipment can be determined only in the process of their intend-ed use, which is not always possible; we can only talk about the effectiveness of training using training equipment. The authors of the paper propose to test the effectiveness of simulators and training complexes through a system of indicators of their qualities. The quality criteria of training equipment can be their ability to implement training and training programs. In accordance with the content and structure of the educational process, the type of autonomous simulators and simulators as part of training complexes, the stages of their life cycle, and operating conditions, the paper proposes four groups of quality indi-cators: didactic, functional, technical, and economic, and establishes their relationships and evaluation methods. Through the indicators of didactic qualities, it is possible to assess the compliance of educational and training tools with the requirements of the educational process, its structure, and content. Indica-tors of functional qualities allow us to evaluate the capabilities of training tools for implementing the didactic requirements of educational programs. Indicators of technical qualities evaluate the character-istics of training tools that ensure their use in the educational process. Economic indicators allow us to estimate the costs at the primary stages of the life cycle of training equipment. Economic indicators allow us to estimate the costs at the primary stages of the life cycle of training equipment. The proposed system of qualities of training equipment, along with a system for evaluating the effectiveness of training with their use, allows us to justify the feasibility of creating training equipment, optimize their structure depending on the requirements and problems of training.

14. The performance evaluation of simulator training by the method of target management [№1 за 2021 год]
Authors: Ilin V.А., Savvateev A.S.
Visitors: 5139
The extensive use of educational processes of educational institutions and in the system of combat training determines the relevance of the development of methods for justifying training equipment, evaluating their effectiveness and the effectiveness of training provides. One of these methods may be the target management method, proposed in this paper and not previously used in this subject area. The method of targeted training management involves determining the goals of training and the re-quirements for its means, the structure, and content of training, and testing the results of training. The paper justifies the choice of different categories of students of different simulators and meth-ods for evaluating the effectiveness of training for different students’ categories. Based on the func-tions of the students' activities, the authors propose and justify the activity classification of the trained operators and their division into three categories, three levels. Under the accepted classification, the authors define the requirements for training equipment and the organization of training equipment. Methods of forming assignments for students under the objectives of training and evaluation of its re-sults, including automation of training assessment, are fundamental in the organization of simulator training based on the method of target management. The paper suggests the following procedure for the development of mathematical software automate the assessment of the preparation: the choice of control parameters and the objective function devel-opment, the parameters and rating scales exercises development, the drafting of the algorithm and pro-vide recommendations. The target management method for evaluating the effectiveness of simulator training is developed because of over ten years of experience in using simulators in the educational process and the authors’ personal participation in their creation and use.

15. Approach to designing software for artificial entity management systems [№1 за 2021 год]
Authors: Vinogradov G.P., Konukhov I.А., Shepelev G.А.
Visitors: 3942
The problems of intelligent control of artificial entities (including robotic complexes (RC)) are closely associated with the problem of decision-making. The formal decision theory was developed by abstracting from subjective fac-tors. This led to the development of a normative theory of decision - making by the "ideal" subject. Analysis of ap-proaches to the construction of RC control systems has shown that they do not have the property of independent de-cision-making. In practice, developers entertain possible behaviors of such systems, and the corresponding algo-rithms are embedded in the RС control system. As a result, such an object doesn’t have a self-sufficient behavior that guarantees the fulfillment of some mission, especially as part of a human-machine system. The demand for be-havior intellectualization forces us to reconsider the logical and mathematical abstractions underlying the construc-tion of their onboard control systems. The work objective is to substantiate the approach to software development of intelligent RC control systems based on the pattern theory. It is necessary to develop an approach that ensures the transfer of effective experience to the RC management system, the compatibility of the theological approach, and the causal approach, which is im-portant when integrating the RC and the personnel of the units. Show that the patterns of the subject's departure from the ideal rational choice to the subjectively rational are associated with the peculiarities of identification and under-standing of the state of the external environment and the properties of their interests. External factors are related to the obligations that the agent assumes. Internal factors reflect the interests of the subject, induced by his needs and the ethical system that he sticks to. The paper uses the methods of the theory of reflexive games and the theory of information management of sys-tems with will and intelligence. It is shown that the choice in conditions of severe time deficit is made based on behavior patterns that reflect ef-fective experience. Patterns form both the information structure of representations and the set of possible variants of representations. Assessments of contentment with the current situation of choice by the subject lead to a change in the structure of interests of the subject, and he can choose it. A formal model of the behavior pattern is developed. An approach to solving the problem of identification and construction of pattern models is proposed. For these pur-poses, four positions of information processing were used, and a method of logical inference on patterns was devel-oped. The results of software solutions for identifying the behavior pattern when using a new generation of training systems are presented.

16. Application of the purposeful behavior principle in the cognitive control system of a radar station [№1 за 2021 год]
Author: A.A. Nepryaev
Visitors: 3320
The application of cognitive technologies in radar is a rapidly developing area with many opportunities for innovation. A significant obstruction in this discipline is the lack of a common understanding of how the architecture of a multi-function radar control system should be designed to include multiple feedback loops that enable the manifestation of cognition. In the radar community, there is still no pre-cise definition of what distinguishes an adaptive system from a cognitive one. This work is intended to expand and substantiate the list of elements and qualitative characteristics that must be present in a radar system in order for it to be classified as cognitive. The author suggests the use of a metacognitive approach to developing a model of purposeful behavior that selects the most profitable strategy and controls the cognitive processes involved in learning. The action selection based on the perception of the environment is the fundamental characteristic of the cognitive system. Finally, the choosing process leads to the optimization problem, when it is de-sirable to choose the action with the maximum reward. This is determined by the degree of similarity of the current internal and external states with the target one. This is based on the principle that radar sys-tems should not be classified as cognitive or non-cognitive, but should be evaluated by the degree of severity of cognitive functions. The author suggests a gradation of cognitive systems based on the prin-ciple of purposeful behavior of control system elements. The article substantiates the need to consider the ability of the system to function in actual time and computing power as a sign that determines the degree of expression of its cognitive abilities.

17. Application of high-level synthesis technology and hardware accelerators on FPGA in protein identifications [№1 за 2021 год]
Authors: G.K. Shmelev, M.A. Likhachev , Arzhaev V.I.
Visitors: 3125
The paper considers the use of high-level synthesis technology using hardware accelerators based on FPGA in the identifying proteins problem. Currently, there are a significant number of hardware solutions with high performance and band-width designed to solve various applications. One such solution is hardware-based computation accel-erators based on field-programmable gate array (FPGA), which have a number of advantages over ac-celerators built on both graphics processing unit (GPU) and application-specific integrated circuit (ASIC). However, there is a certain complexity in the wide application of such devices, which consists in the laboriousness and specificity of the traditional way of developing applications using specialized programming languages for this type of accelerator. Using high-level synthesis technology using one of the popular programming languages opens up new horizons in the wide use of such accelerators. This paper describes one embodiment of a computational hardware and software platform using a hardware accelerator on a FPGA. Special attention is paid to considering the major steps of developing the architecture of applications deployed on hardware and the methodology for developing a high-performance computing core of hardware-accelerated software functions. The results of improving the computational performance of the de novo peptide sequence sequencing software application and the effectiveness of the used hardware platform and the chosen development path in comparison with the original software application are demonstrated.

18. Software environments for studying the basics of neural networks [№1 за 2021 год]
Authors: P.Yu. Bogdanov , E.V. Kraeva , S.A. Verevkin, E.D. Poymanova, Tatarnikova, T.M.
Visitors: 3288
The paper describes the ways and methods of studying and constructing neural networks. It is shown that the study of the functioning guidelines of neural networks, their application for solving certain problems is possible only through practice. There is the analysis of various software environments that can be used in the laboratory and prac-tical classes for the study and application of neural networks in the paper. Highlighted the modern cloud service Google Colaboratory, which is recommended for teaching the basics of neural networks due to the presence of a pre-installation of the Tensorflow library and a library for working in Python, free access to graphics processors, the ability to write and execute program code in a browser, and no need for special configuration of the service. Examples of designing neural networks in the Colaboratory are considered. In particular, solving recognition problems and image classification, predictive modeling. The authors show that a convolu-tional neural network can be used for image recognition and classification, a feature of which is obtain-ing the image features a map with subsequent convolution. There are chunks of code for the connecting phases the necessary libraries, loading data sets, normalizing images, assembling a neural network, and its training, in the paper. The solving of the forecasting problem is considered on the example of a feed-forward neural net-work with an algorithm for backpropagation of errors in the learning process, the essence of which is to obtain the expected value at the output layer when the corresponding data is fed to the input layer. Backpropagation of errors consists of adjusting the weights that give the greatest correlation between the input dataset and its corresponding result.

19. Software complex for detection and classification of natural objects based on topological analysis [№1 за 2021 год]
Authors: S.V. Eremeev , A.V. Abakumov
Visitors: 2480
The algorithm of natural objects search on satellite images requires a certain balance. Due to natural character, there are no two completely identical objects, so this problem requires some stability from the algorithm. For such purposes, topological data analysis methods can be used. These methods allow us to obtain a unique characteristic of the image as barcodes, which can be used as training by most modern classifiers. Software complex, based on topological analysis methods, has been developed. It allows us to search for a necessary natural object on a raster image for its further classification and processing. The software complex structure includes several subsystems. They are a subsystem of interest areas selec-tion on the image, a subsystem of barcode building, a subsystem of similar objects search, and a sub-system of found objects output. The principles of selecting objects of interest in images, building barcodes, and comparing them are described in detail. Topological characteristics in the form of Betty numbers are calculated for each spatial object selected on the geo-image. These characteristics are the basis for building the barcode. The process of image decomposition into a sequence of binary images to obtain stable topological characteristics is shown. The principle of barcode comparison for determining the similarity of selected areas of interest with reference objects is demonstrated. There are examples of using software complex for the search problem for ice on the raster image in the paper. The results of found objects with different degree of similarity regarding templates depend-ing on the specified parameters are shown. The software complex can be used for a wide range of problems of natural objects analysis on satellite images including data processing for a different time and on different scales.

20. Developing ontology schemas based on spreadsheet transformation [№1 за 2021 год]
Authors: Dorodnykh N.O., Yurin A.Yu., A.V. Vidiya
Visitors: 3482
Using ontologies is a widespread practice in the in creating intelligent systems and knowledge bases, in particular, for the conceptualization and formalization of knowledge. However, most modern ap-proaches and tools provide only manual manipulation of concepts and relationships, which is not al-ways effective. In this regard, using various information sources, including spreadsheets, is relevant for the automated creation of ontologies. This paper describes a method for the automated creation of ontological schemes in the OWL2 DL format based on the analysis and transformation of data extracted from spreadsheets. A feature of the method is the use of the original canonical relational form for the intermediate representation of spreadsheets, which provides the unification of input data. The method is based on the principles of model transformation and comprises four primary stages: converting the original spreadsheets with an arbitrary layout into a canonical (relational) form; obtaining fragments of the ontological scheme; ag-gregation of separate fragments of the ontological scheme; generation of the code of the ontological scheme in the OWL2 DL format. The method is implemented in the form of two software tools integrat-ed by the data: TabbyXL as the console Java application for table conversion and the PKBD.Onto plugin as the extension module for Personal Knowledge Base Designer (software for expert systems prototyping). The transformation of a spreadsheet with information about minerals is considered as an illustrative example, and the transformation result is presented in the form of a fragment of an ontolog-ical scheme. The method and tools are used in the educational process at the Institute of Information Technologies and Data Analysis of the Irkutsk National Research Technical University (INRTU).

◄ ← Preview | 1 | 2 | 3 | Next → ►