«Systems and Means of Informatics»
Scientific journal
Volume 34, Issue 2, 2024
Content | About Authors
Abstract and Keywords
Abstract: Methodological support for synthesis of conditionally: optimal filtering and control for continuous, discrete, and continuous-discrete stochastic systems with unsolved derivatives (StS USD) and random parameters is developed. A survey of conditionally-optimal filtering (COF) and conditionally optimal control (COC) is presented. It is supposed that random parameters are described by multicomponent integral canonical expansions (MC ICE). Synthesis of COF and COC is based on Pugachev's COC concepts. Special attention is paid to the three COC typical problems in the case of local criteria. Applications: nonlinear correlational estimation of potential and instrumental accuracy of COF and COC at nonstationary impact disturbances in StS USD. An illustrative example is given. Future research is connected with effects of disturbances accumulation, drifts, and excurtions based on finite dimension distribution of stochastic processes.
Keywords: conditionally optimal control (COC); conditionally optimal filtering (COF); multicomponent integral canonical expansions (MC ICE); random parameters; stochastic systems with unsolved derivatives (StS USD)
- I. N. Sinitsyn
- V. I. Sinitsyn
- E. R. Korepanov
- T. D. Konashenkova
Abstract: For scalar stochastic processes (StP) at finite time intervals and their canonical expansions (CE), a technology based on wavelet neural networks (WNN) is constructed. For WNN learning, the method of steepest descent was used. The three-layer WNN architecture is presented. The activation functions of the latent layer are based on chosen wavelet basis with general compact carrier.
For StP covariance function, a special WNN algorithm of CE construction is developed. The covariance function CE corresponds to CE StP in the form of linear combination of wavelet basis with zero mathematical expectations and variances defined by the suggested algorithm. A numerical example illustrates CE WNN preference with wavelet CE.
Keywords: canonical expansion; covariance function; modeling; stochastic process; wavelet; wavelet neural network
- A. S. Rumyantseva
- F.A. Moskaleva
- A. K. Samouylov
- S. Ya. Shorgin
- Yu. V. Gaidamaka
Abstract: The increase in the number of users and high network requirements have led to the need to introduce millimeter wave data transmission technology (mmWave) in order to increase data transfer speeds and minimize delays. The transition to the millimeter wave band in the fifth generation of communication networks allows the specifics of the implementation of the provision of services, including the sensitivity of directional antenna beams to radio signal blocking. In addition, when effectively planning a network providing unicast and multicast services, it is necessary to achieve a balance between opposite solutions for these two types of traffic. In this article, using the methods of the queuing theory, a mathematical model has been developed to describe the prioritizationmechanism for servicing a combination of unicast and multicast traffic, for each of which the possibility of line-of-sight (LoS) blockers is regarded. The features of signal propagation are taken into account in the model by dividing the coverage area into two areas. The constructed mathematical model allows one to find analytical expressions for the probabilities of loss of unicast and multicast connections as well as the amount of occupied resource both by individual types of traffic in different areas of the coverage area and in the entire system as a whole. The method of distribution of the radio resource of the network base station is based on the balance between the listed main performance indicators which allows one to monitor compliance with quality requirements for both types of services.
Keywords: mmWave; unicast connections; multicast connections; LoS; LoS blocking; mathematical model
Abstract: The article discusses the problem of statistical test of queuing system stability based on the characteristics of the sojourn time series {Vk}. To solve it, it is proposed to use the tests for unit roots. A relationship is established between the quantitative and temporal characteristics of the system in the style of the Little's formula for the nonstationary case which makes it possible to link the rate-stability with the stationarity {Vk}. Three basic models and the corresponding Dickey{Fuller tests are considered as the stationarity test. The limitations of the used statistical tables are discussed. The conducted simulation study allows one to draw the following conclusions: all three criteria solve the problem of stability analysis with a clear loss of the quality of decisions made in the case of a model with a deterministic linear trend, in the field of system stability, the test behaves worse than in the field of instability; the expected trends in the use of tests are clearly manifested; and some deviations from the expected results indicate the need to expand the range of the models used and to remove the limitations of the statistical tables used.
Keywords: queueing system; time series analysis; unit root tests; Dickey{Fuller tests; simulation
- Yu. A. Stepchenkov
- Yu. G. Diachenko
- N. V. Morozov
- D. Yu. Stepchenkov
- D. Yu. Diachenko
Abstract: Self-timed (ST) circuits have high reliability. They guarantee detection and localization of any persistent faults and demonstrate a high level of fault tolerance. However, designing ST circuits is more labor-intensive compared to synchronous circuits because one should construct an additional indication subcircuit and adhere to the principles of truly ST circuit implementation. Formalized desynchronization provides automatic conversion of the original synchronous circuit description into the self-timed one but when synthesizing sequential ST units, including ST counters, it leads to excessive hardware redundancy and, as a consequence, to their low performance. The article substantiates the approach to the ST counter synthesis based on the heuristic method formalization for their construction and ensuring the guaranteed resulted truly ST implementation that functions in full accordance with the original description and has close to optimal consumer characteristics
Keywords: automated synthesis; self-timed circuit; counter; desynchronization; preset; indication
Abstract: The "software-defined" concept was born in relation to networks. Software-defined networking (SDN) assumed centralized automation of network administration and replacement of expensive routers with cheaper switches. The SDN has undergone significant changes when it was extended from university labs to data centers and provider networks. As a result, in practice, SDN networks represent traditional IP networks with full-featured routers, only supplemented with SDN functionality. Following networks, the "software-defined" concept has penetrated the data storage area where the software-defined storage (SDS) has acquired a new important feature - service orientation. With the further extension of the "software-defined" concept to entire software-defined data centers (SDDC) and software-defined clouds, this feature became dominant. It can be expected that the success of SDDCs would make redundant independent SDNs and SDSs. However, the widespread implementation of SDDCs is hindered by difficulties of creating specific orchestration software. Therefore, for an ordinary data center owner, the only opportunity to acquire an SDDC is the purchase of an integrated turnkey solution with ready orchestration software from a reliable supplier.
Keywords: software-defined data center (SDDC); software-defined networking (SDN); software-defined storage (SDS)
Abstract: A structure and basic design solutions are proposed for constructing a sample geodatabase with support for phase variation of geomodels. This structure is a development of the previously developed multiscale geodatabase with support for representing geoobjects in various application contexts. The essence of phase variation and its fundamental differences from full-fledged temporal modeling are considered. It is concluded that the proposed approach can be widely used in specialized geographic information systems that require taking into account the state of their objects at a specific point in time as an operational geodatabase, since its operation can be separated from the process of continuous refinement of knowledge about the area implemented by maintaining the main temporal geodatabase.
Keywords: geodata; geodata databases; conceptual scheme of the geodata database; temporal modeling of geoobjects; phase variation of geomodels
Abstract: Ideas about an object and a process as a heterogeneous integral system are not completed in a theory and have not become a world picture of highly professional specialists in the heterogeneous problematic environment of diagnostics and prognosis in medicine, technology, etc. Decision-making under conditions of limited resources, ambiguity in assessing the situation, and a large volume and variety of processed information about an object is accompanied by serious errors and risks. Decision-making virtualization technologies are needed. The paper discusses the concept of intelligent decision support systems in medicine, features and problems of the subject area, and approaches to the development.
Keywords: intelligent decision support system; medicine and healthcare; development problems; development approaches
- I. M. Adamovich
- O. I. Volkov
Abstract: The article is devoted to the use of the collective entity resolution method based on a new relational clustering algorithm, which is a modification of the greedy agglomerative clustering algorithm, in concrete historical investigation when processing nominative sources. The article proposes the method for searching for optimal values of parameters of the collective entity resolution algorithm for tasks related to concrete historical investigation. The method is based on the analysis of the specifics of concrete historical data, their comparison with test data for which there are estimates of the effectiveness of the algorithm, and the procedure for finding the optimal process parameters according to the Gauss-Seidel scheme that consists in sequentially searching for the function optimum alternately for each variable. The application of the proposed method makes it possible to use the considered entity resolution algorithm in real concrete historical research in the tasks of automated record linkage in nominative sources.
Keywords: concrete historical investigation; distributed technology; entity resolution; algorithm parameters; relational similarity measure
|
