Eng | Rus

“Systems and Means of Informatics” scientific journal

Volume 33, Issue 2, 2023

Content   Abstract and Keywords   About Authors

MULTIPLEXED SELF-TIMED PIPELINE
  • Yu. A. Stepchenkov
  • Yu. G. Diachenko
  • D. Yu. Stepchenkov
  • D. Yu. Diachenko
  • G. A. Orlov

Abstract: The article considers the problem of correct practical implementation of a self-timed pipeline with multiplexing of actively operating stages taking into account the circuitry limitations imposed by the technology used. Complementing two parallel branches with a self-timed marker FIFO (First- Input-First-Output) reduces the average latency of the pipeline and ensures its operation is self-timed. With sufficient FIFO capacity, successive data portions can be processed simultaneously by different pipeline's branches. The circuitry solutions of the multiplexed pipeline's individual units guarantee pipeline practical feasibility in any complementary metal-oxide-semiconductor process at domestic semiconductor factories.

Keywords: self-timed circuit; pipeline; multiplexing; latency; performance

RECOGNITION OF ANOMALIES ON MULTITIME PANORAMAS USING THE NEURAL NETWORK METHOD OF MODEL AMALGAMATION
  • P. O. Arkhipov
  • S. L. Philippskih

Abstract: A method for classifying anomalies on multitime panoramas using the neural network method of model amalgamation is described. The main idea of this method is to break the deep learning model into its component parts.
Then, existing design patterns are selected and adapted for each part. To solve the problem of classifying anomalies, a composite template was developed and a new neural network model was designed. An improvement of the design pattern of neural networks ConvNet based on the technology of using "dense convolutional blocks" has been made. To classify anomalies, a new template has been developed - SOTA-ConvNet which includes a ready-made neural network model architecture suitable for working with panoramic images. The neural network model built on the basis of the SOTA-ConvNet template showed advantages over the neural network based on ConvNet: the accuracy of image classification increased and the number of trainable parameters decreased by an order of magnitude.

Keywords: panoramic image; model amalgamation; multiclass classification; design pattern; convolutional neural network; data set

DEEPFAKE IMAGE DETECTION USING BISPECTRAL ANALYSIS
  • S. P. Nikitenkova

Abstract: As deep-fake image synthesis tools become more powerful and available, there is a growing need to develop methods for detecting generated content. The main goal of the work is to test the application of bispectral analysis as a tool for detecting images generated by artificial intelligence (AI).
It is shown that higher-order spectral correlations detected by spectral analysis are less present in natural images compared to the images generated using generative-adversarial neural networks GAN (generative adversarial network).
These correlations are probably the result of fundamental properties of the image generation process. The clustering procedure has shown encouraging results: it determines the generated images with an accuracy of 80%.

Keywords: Al-synthesized image; polyspectral analysis; nonlinearity; machine learning algorithms

DIRECTIONS OF DEVELOPMENT OF THE METHODOLOGICAL BASE FOR WORKING WITH GEODATA IN PROMISING GEOINFORMATION SYSTEMS
  • D. A. Nikishin

Abstract: The directions of improving the methodological base for promising geoinformation systems (GIS) are considered. The main subject of consideration are technologies due to the provision of interaction and cooperation in distributed GIS, complex processing of data obtained by various means of remote sensing of the Earth, and functioning of networks of stationary and/or mobile geosensors.
The conclusion is made about the expediency of forming an updated set of conventional geographical knowledge (theories and "practices" fixed in the form of standards) taking into account the modern development of models, methods, and technologies for working with geodata which can become the basis for solving popular types of tasks in perspective GIS.

Keywords: geoinformation systems; geoinformation engineering; modeling of geographical phenomena; geosensory networks

SELECTION OF TECHNOLOGICAL SOLUTIONS TO SUPPORT THE PROCESS OF SYNTHESIS OF GEODATA OF RAILWAY TRANSPORT INFRASTRUCTURE
  • I. N. Rozenberg
  • S. K. Dulin

Abstract: Datafusion is a process in which data combined from multiple sources are used for analysis to produce more consistent, accurate, and useful information. Receiving and merging data from sensors is defined as multisensor data fusion.
The most striking example of data fusion is the person himself who constantly combines data coming from all the senses to perform the necessary actions in the environment. In particular, a person would not be able to drive a car without the fusion of visual, auditory, and motor data. In geoinformatics, instead of geodata synthesis, the term geodata integration is often used. Modern geographic information systems can combine different geodatasets so that the resulting geodatasets contain attributes and metadata not available for the points in the combined datasets. The article analyzes the categories and models of geodata synthesis used to ensure the relevance of the spatial data system in the infrastructure of railway transport.

Keywords: data synthesis; categories of data synthesis; data synthesis model

A VISUAL LANGUAGE OF THE REPRESENTATION OF THE CONFLICT MANAGEMENT PROCESS IN HYBRID INTELLIGENT MULTIAGENT SYSTEMS
  • S. B. Rumovskaya
  • I. A. Kirikov

Abstract: Hybrid intelligent multiagent systems (HylMAS) with problem- and process-oriented conflicts and a visual representation of the conflictological aspect are relevant to the group dynamics of a team of experts in various fields solving the problem. Conflict management in HylMAS will allow suppressing destructive aspects of conflicts and stimulating the constructive ones. Visualization of a conflict situation makes the arising contradictions contrast visible. It provides the user with a quick understanding of which members of the team and at what stages the conflict occurs, what type and tension it is, which method of collective reasoning of specialist agents has made the greatest contribution to the process of stimulating a constructive conflict, and, also, at what stage the conflict is minimized. Also, it shows all used strategies for resolving conflicts and subgroups in the team of agents. A visual language of the representation is proposed. The syntax of the language is described using extended Backus-Naur forms augmented with positional syntax operators.

Keywords: team of specialists; agent conflict; visual language; extended Backus-Naur forms; conflict management representation; hybrid intelligent multiagent systems

CAUSE-AND-EFFECT RELATIONSHIPS IN ANALYSIS OF UNOBSERVABLE PROCESS PROPERTIES
  • A. A. Grusho
  • M. I. Zabezhailo
  • V. V. Kulchenkov
  • D. V. Smirnov
  • E. E. Timonina
  • S. Ya. Shorgin

Abstract: When implementing information technology, there exist sequences of objects that can be identified but it is difficult or impossible to observe their properties. In some objects of such sequences, an unobservable property A may appear, the detection of which is of interest. To identify objects with property A, one has to use the search for the effects of property A as the cause of such effects. If one considers a sequence of objects in which objects with property A may appear, then there are problems of assessing the probability of the existence of false objects with the identified property A when it actually does not exist (false alarms) and estimating the probability of skipping the object with property A. The work uses the determinism of the occurrence of effects when the cause appears, if there is an interaction of the object containing the cause with the object containing the observed effect. In this case, the existence of noise is allowed which makes it difficult to observe the effect when it is, and can create a false effect when the cause is absent. The methodology for assessing the probability of identifying objects with property A constructed in the work is quite easy to use but is important in assessing the solvability of a search problem.

Keywords: cause-and-effect relationships; probabilistic error analysis; searching for the property in unobservable data

METHODOLOGICAL APPROACH TO THE SELECTION OF KEY PERFORMANCE INDICATORS FOR EVALUATING THE INFORMATION AND TECHNOLOGY ORGANIZATIONS
  • A. A. Zatsarinny
  • Yu. S. Ionenkov

Abstract: The article presents a methodological approach to the selection of key performance indicators (KPI) for evaluating information and technology (IT) organizations. The value of KPIs in assessing the effectiveness of organizations is shown. The main approaches to the selection of the nomenclature of KPI, including the types of such indicators and requirements for them, are presented.
The proposals have been developed on the KPI nomenclature in relation to IT organizations. It is suggested to use three types of indicators: target, functional, and financial. The proposals on methods for calculating individual indicators are formulated.

Keywords: information system; key performance indicators; targets; functional indicators; financial indicators

SITUATIONAL MANAGEMENT SYSTEM AS A MULTISERVICE TECHNOLOGY IN A CLOUD ENVIRONMENT: ANALYTIC SERVICES
  • A. P. Suchkov

Abstract: This article is the fourth in a series of articles devoted to the methodology of development and implementation of automated situational management systems as a public multiservice technology in a cloud environment. It is proposed to use existing IaaS (Infrastructure as a Service) and DaaS (Desktop as a Service) service models that allow implementing the required infrastructure (servers, workstations, storage, and databases) and specialized workstations with application software that implement the necessary functionality. The main approaches to the creation of a multiservice standard situational management system as a service CSaaS (Control System as a Service) are proposed. The article describes approaches to the construction and detailing of one of the main groups of CSaaS services which ensure the creation, maintenance, and use of a system of analytical services that support decision-making processes.

Keywords: automated situational management system; multiservice technology; cloud environment; analytics services

SOME ISSUES OF DISAGGREGATION AND COMPOSABILITY OF THE DATA CENTER INFRASTRUCTURE
  • V. B. Egorov

Abstract: The composable disaggregated infrastructure (CDI) promises to become the next step in improving organization and functioning principles of data centers. The essence of CDI is in transition to a "fine-grained" infrastructure and automatic layout of local executive infrastructures that should be optimal for each specific workload. With obvious advantages in flexibility of using data center hardware resources, the CDI has, nevertheless, some limitations in feasibility and applicability. Although there are offerings of data center infrastructures announced as disaggregated or composable, those appear in many cases nothing more than advertising tricks to attract users with fashionable terms.
In reality, other infrastructural solutions are often proposed under the disguise of the CDI. This is largely provoked, in addition to purely marketing aspects, due to the vagueness of the CDI term. In particular, the concept of composability is applicable at various infrastructure "granularity" levels including not necessarily disaggregated ones. As a general rule, the CDI concept implementation is feasible only within a software-defined data center.

Keywords: disaggregation; disaggregated infrastructure; composable infrastructure; CDI

DEVELOPMENT OF A MODULE FOR WORKING WITH THE plasticStressedFoam SOLVER BASED ON THE OpenFOAM PACKAGE
  • D. I. Chitalov

Abstract: The paper describes a study aimed at creating a software module for managing numerical experiments using the plasticStressedFoam solver and connecting it to the basic version of the graphical shell proposed by the author.
The work is aimed at optimizing the process of filling in the configuration files of the design case using the graphical interface and automating the launch of the plasticStressedFoam solver which is responsible for generating the final numerical model. The purpose of the development is determined and a list of tasks necessary to achieve it is formulated. The stages of module development are described including the construction of diagrams of its structure and logic of application, the choice and justification of the stack of development technologies as well as the results of the study, its scientific novelty, and practical significance. The results of the product performance check are presented by the example of a standard problem solved in OpenFOAM using the plasticStressedFoam solver.

Keywords: numerical simulation; continuum mechanics; graphical user interface; OpenFOAM; Python programming language; open source software; plasticStressedFoam solver; PyQt5 library; SQLite

THEORY OF S-SYMBOLS: FORMALIZATION OF KNOWLEDGE ABOUT S-PROBLEMS
  • V. D. Ilyin

Abstract: The theory of S-symbols is an extended generalization of the theory of S-modeling. It is considered as a part of the methodological support for the development of artificial intelligence systems in the S-environment (including knowledge systems, systems of S-modeling of problems and program design, etc.). The S-environment based on interconnected systems of S-(symbols, codes, signals) serves as the infrastructural basis for the implementation of information technologies for various purposes. The article presents the second part (out of four) of the description of the theory. Definitions of concepts related to S-problem objects (S-problem, S-problem graph, etc.) are given. The basics of designing S-problem objects (including the construction of resolving structures on S-problem graphs) are described.

Keywords: theory of S-symbols; S-problem; S-problem graph; resolving structure; programming automation

APPLICATION OF THE CHAID ALGORITHM IN THE TECHNOLOGY OF CONCRETE HISTORICAL INVESTIGATION SUPPORT
  • I. M. Adamovich
  • O. I. Volkov

Abstract: The article continues the series of works devoted to the technology of concrete historical investigation support. The technology is based on the principles of co-creation and crowdsourcing and is designed for a wide range of users which are not professional historians and biographers. The article is devoted to the application of the decision tree method based on the CHAID algorithm to automatically fill information gaps in the set of historical facts in order to determine potentially promising areas of research. The algorithm is described and the reliability of its results with a high proportion of missing values in the data is evaluated. The proportion of lacunas in the main sources of multiple facts is estimated and the conclusion of the applicability in principle and the effectiveness of the algorithm is made taking into account the specifics of the technology. It is also shown that the CHAID algorithm develops and supplements the means of anomalies in concrete historical data detecting existing in the technology.

Keywords: concrete historical investigation; distributed technology; CHAID algorithm; missing data; anomalies




 

RUS