Институт проблем информатики Российской Академии наук
Институт проблем информатики Российской Академии наук
Российская Академия наук

Институт проблем информатики Российской Академии наук




«INFORMATICS AND APPLICATIONS»
Scientific journal
Volume 3, Issue 4, 2009

Content | Bibliography | About  Authors

Abstract and Keywords.

PROBABILISTIC METHODS OF INFORMATION MODEL BUILDING FOR THE EARTH ROTATION IRREGULARITY

  • I.N. Sinitsyn    IPI RAN. sinitsin@dol.ru.

Abstract:    New linear and nonlinear probabilistic methods of model building for fluctuation of the Earth irregularity are considered. Methods are the basis of a priori data processing for the problem “Statistical Dynamics of the Earth Rotation.” Test examples are provided.

Keywords:    information model; information resources; linear and nonlinear probabilistic methods; hereditary; fluctuations; one- and multidimensional distributions; asymmetry; excess; Gaussian and Poisson white noises; distribution parameterization; moments methods

ELASTIC DEFORMATIONS IMPACT ON FINGERPRINT RECOGNITION PERFORMANCE.

  • O. S. Ushmaev    IPI RAN. oushmaev@ipiran.ru
  • A.R. Arutyunyan    Nuclear Safety Institute RAN, artem@ac.ibrae.ru.

Abstract:    Elastic deformations are the strong negative factor in fingerprint recognition. The mechanical approach to fingerprint deformation modeling was employed to determine elastic deformation impact on fingerprint recognition. The statistic analysis of structure of elastic deformations was carried out. It revealed that an arbitrary deformation is the combination of elementary principal deformations: rotation, torsion, and traction. Finally, numerical measure for the impact of deformations on fingerprint recognition was found.

Keywords:    fingerprint recognition; nonlinear distortions of fingerprints

CALCULATION ALGORITHM OF WORKLOAD OF TELECOMMUNICATION NETWORK WITH REPETITIVE TRANSMISSIONS.

  • Ya.M. Agalarov    IPI RAN. agglar@yandex.ru.

Abstract:   The models of packet switching network with repetitive transmissions for two schemes of buffer memory distributions — complete sharing and complete partitioning — are considered. The iterative method of calculation of stream intensity in network and probabilities of node blocking where node model is the queueing system of type is proposed.
The necessary condition for existence of solution of stream balance conservation equations in steady-state regime was established. The monotone convergence of stream intensities sequence and probabilities of blocking derived in the proposed method to the solution of these combined equations was proved.

Keywords:    network of packets switching; buffer memory; repetitive transmissions; probabilities of blocking; iteration method

BAYESIAN QUEUEING AND RELIABILITY MODELS: GENERAL ERLANG CASE.

  • A.A. Kudriavtsev    Department o fMathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, nubigena@hotmail.com
  • V. S. Shorgin    IPI RAN. vshorgin@ipiran.ru
  • S. Ya. Shorgin    IPI RAN. sshorgin@ipiran.ru

Abstract:    The paper is a next step in bayesian queueing and reliability models investigation. The method provides the randomization of system characteristics with regard to a priori distributions of input parameters. The new results are presented for two cases when both a priori distributions are Erlang distributions and when a pair of a priori distributions is the pair “Erlang distribution –degenerate distribution.”

Keywords:    bayesian approach; queueing systems; reliability; mixed distributions; modeling; Erlang distribution

ASYMPTOTIC ANALYSIS OF THE Er(t)|G|1 QUEUE.

  • O. V. Petrova    Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, o.petrova@inbox.ru
  • V.G. Ushakov    Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University; IPI RAN, ushakov@akado.ru

Abstract:    The single server queue with time-dependent Erlangian input is considered. The service times have general distribution. The asymptotic behavior of the queue length with traffic intensity less than 1 and with arrival rate weakly depending upon time is investigated.

Keywords:    Er|G|1 queue; time-dependent Erlangian input; asymptotic analysis

ASYMPTOTIC ESTIMATES OF THE ABSOLUTE CONSTANT IN THE BERRY–ESSEEN INEQUALITY FOR DISTRIBUTION WITH UNBOUNDED THIRD MOMENT.

  • M.O. Gaponova    Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, margarita.gaponova@gmail.com
  • I.G. Shevtsova    Faculty of Computational Mathematics and Cybernetics, M.V. Lomonosov Moscow State University, ishevtsova@cs.msu.su

Abstract:    The Prawitz’ asymptotic estimates for the absolute constant in the Berry–Esseen inequality are sharpened for the case of independent identically distributed random variables with finite third moments. Similar estimates are constructed for the case of unbounded third absolute moment. Also, upper estimates of the asymptotically exact constants in the central limit theorem are presented.

Keywords:    central limit theorem; normal approximation; convergence rate estimate; sum of independent random variables; Berry–Esseen inequality; Lyapounov fraction; asymptotically exact constant

LIMIT DISTRIBUTION OF RISK ESTIMATE OF WAVELET COEFFICIENT THRESHOLDING.

  • A. V.Markin    Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. LomonosovMoscow State University, artem.v.markin@mail.ru

Abstract:    Asymptotic properties of risk estimate of wavelet coefficient thresholding are studied. Under certain conditions, there is a convergence of the difference of risk estimate and risk itself to normal distribution.

Keywords:    wavelets; thresholding; risk estimate; limit distribution

TECHNOLOGY OF POORLY   FORMALIZED DOCUMENTS STORAGE ON THE BASIS OF LEXICOLOGICAL SYNTHESIS.

  • B. V. Chernikov    Limited Liability Company “ANT-Inform,” bor-cher@yandex.ru

Abstract:    The technology of storage of poorly formalized documents that are created using lexicological synthesis is considered. The technology provides formation of the kept index sequences containing indexes of document forms and their substantial components. Additionally, thanks to simultaneous preparation of documents and creation of kept index sequences, the economy of time is provided. The experiments have shown the efficiency of the approach for the documents created for management of different kinds of activity.

Keywords:    poorly formalized document; lexicological synthesis; index; index sequence; compression

SELF-ORGANIZATION OF INTELLIGENT AGENTS GROUPS SIMULATION DEPENDING ON DEGREE OF INTERACTION.

  • I.A. Kirikov    Kaliningrad branch of the IPI RAN, kfipiran@yandex.ru
  • A. V. Kolesnikov    Kaliningrad branch of the IPI RAN, avkolesnikov@yandex.ru
  • S. V. Listopad    Kaliningrad branch of the IPI RAN, ser-list-post@yandex.ru

Abstract:   One of the approaches to the creation of self-organizing intellectual computer system for decision-making support based on an analysis of the experts’ goals is considered. An algorithmfor determining the type of multiagent system architecture based on the extent of interaction between agents is considered, that is relevant to determining the effectiveness of expert groups and to improving the quality of decision-making.

Keywords:    decision support computer system; self-organizing multiagent system; similarity measure of agents’ fuzzy goals; algorithm for determining the type of multiagent system architecture based on the extent of interaction between agents

TIME-DEPENDENT SEMIOTIC MODEL FOR COMPUTER CODING OF CONCEPTS, INFORMATION OBJECTS, AND DENOTATA.

  • I.M. Zatsman    IPI RAN. iz_ipi@a170.ipi.ac.ru

Abstract:   The time-dependent semiotic model, which has been developed during research of problems of generation and evolution of goal-oriented knowledge systems in digital libraries and other kinds of information systems, is considered. These problems concern to the new direction of the researches, which have been named “Cognitive Informatics.” This model is positioned as theoretical foundations for computer coding of concepts, information objects, and denotata in view of their evolution in time.

Keywords:    semiotic model; goal-oriented knowledge systems; denotata; concepts; information objects; computer codes; three-component coding; concept evolution trajectory