“Informatics and Applications” scientific journal

Volume 6, Issue 2, 2012

Content   Abstract and Keywords   About Authors

ALGORITHM FOR COMPUTATION CHARACTERISTICS OF TELECOMMUNICATION NETWORK WITH RESUBMITS AND INCOMPLETE CIRCUIT BUFFER MANAGEMENT MODEL.

• Ya.M. Agalarov  IPI RAN, agglar@ya.ru

Abstract: Telecommunication network with SMQMA (Sharing with Maximum Queue Length and Minimum Allocation) nodes buffer management scheme and the ability to repeat the transmission from the source and transit nodes are discussed. The computation algorithm for average network characteristics such as probability of nodes blocking, total load in lines, average number of packets in nodes and network, average number of packets, waiting for repeat in sources is suggested. Proofs of statements concerning the properties of the algorithm and the results of computational experiments are presented.

Keywords:  telecommunication networks; repeat of transmission; mechanisms of buffers management

ANALYSIS AND OPTIMIZATION PROBLEMS FOR SOME USERS ACTITITY MODEL.
PART 3. EXTERNAL RESOURCES OPTIMIZATION.

• A. V. Bosov  IPI RAN, AVBosov@ipiran.ru

Abstract: This paper completes the mathematical model describing the activity of users and optimization problems for distribution of internal computational resources proposed by the author earlier. The optimization problem for the distribution of external resources is formulated and solved. Suboptimal optimization algorithms are proposed.

Keywords:  information systems; database management system; stochastic observation system; quadratic criterion

ON STABILITY OF NORMAL LOCATION MIXTURES WITH RESPECT TO VARIATIONS IN MIXING DISTRIBUTION.

• A.K. Gorshenin  IPI RAN, agorshenin@ipiran.ru

Abstract: The stability of finite location mixtures of normal distributions with respect to parameter variations of the mixing distribution is studied. The results are stated for the models of addition and splitting of components which are used to test statistical hypotheses about the number of mixture components.

Keywords:  location mixtures of normal distributions; Levy metric

GEOSPATIAL INFORMATION PROCESSING ON THE BASE OF GIS REPOSITORY.

• S.K. Dulin  IPI RAN, s.dulin@ccas.ru
• I.N. Rozenberg  Research & Design Institute for Information Technology, Signalling and Telecommunications on Railway Transport (JSC NIIAS), I.Rozenberg@gismps.ru
• V. I. Umansky   "IntechGeoTrans" Close Corporation, umanvi@yandex.ru

Abstract: Effective integration of the data describing the interacting components is a key to successful management of the entire system of operating objects. Lack of data integration can lead to significant inefficiencies of operational, tactical, and long-term management strategies. The integrated management system can serve to overcome this inefficiency and improve coordination and cost-effectiveness of solutions. Data Integration is no doubt the most responsible action for the implementation of successful management strategies. An approach to the use of the centralized repository of corporate data which allows to combine spatial and nonspatial data object descriptions is presented. Repository is the environment for data sharing and pooling of data and software products in action.

Keywords:  facilities management; data integration; GIS; repository

TECHNIQUE FOR MODELING THE SERVER LOADING IN OPEN CLOUD COMPUTING SYSTEMS.

• D. V. Zhevnerchuk  Tchaikovsky Technological Institute (branch) the Izhevsk State Technical University, drevnigeck@yandex.ru
• A. V. Nikolaev  Tchaikovsky Technological Institute (branch) the Izhevsk State Technical University, elodssa@yandex.ru

Abstract: Planning resource of server system of cloud computing is a challenging task. Such factors as composition and parameters of the hardware platform, parameters of the systemsoftware thatmanages the execution of applications, the properties of traffic generated by users and determining the modes of operation of applications should be taken into account. Technique of estimating the server loading in open cloud computing systems is proposed. The technique is based on the analysis of processes of user interaction with the software. Reliability of the technique is justified. The results of modeling the load on the server side of management system simulation are presented.

Keywords:  cloud computing; imitation modeling; human–computer interaction

RUSSIAN ACADEMY OF SCIENCES LIBRARIES TASKS AND FUNCTIONS IN MODERN CONDITIONS.

• N. E. Kalenov  Library for Natural Sciences, Russian Academy of Sciences, nek@benran.ru

Abstract: The problems of scientific libraries activities changes due to the development of network technology, the burgeoning growth of Internet available electronic publications and databases are being considered. It is shown that academic libraries are an integral part of the scientific infrastructure in the modern situation. The traditional tasks such as science information support remain the prerogative of the libraries, but they must be based on new solutions and performed with an extensive use of modern network technologies. New approaches to the traditional tasks are illustrated on the example of Library for Natural Sciences of the Russian Academy of Sciences (LNS RAS) (http://www.benran.ru). In addition to the traditional tasks, academic libraries are ready for new challenges in the area of bibliometrics research, digitization of printed publication materials, etc.

Keywords:  academic libraries; automation; informatics; user service; informationproviding; computer technologies; digital libraries; automation workplaces; LNS RAS

UNIFICATION OF THE RULE-BASED SYSTEM LANGUAGES TO PROVIDE INTEROPERABILITY OF DECLARATIVE PROGRAMS.

• L.A. Kalinichenko  IPI RAN, leonidk@synth.ipi.ac.ru
• S. A. Stupnikov   IPI RAN, ssa@ipi.ac.ru

Abstract: The W3C standard RIF (Rule Interchange Format) that is provided for the interoperability of various rule based systems by the introduction of the extensible family of unified languages (dialects) oriented on creation of semantic preserving mappings of rule based languages of various systems into the dialects is analyzed. To characterize the motivation for the RIF project, a short survey of development and application of rule based languages and systems in the areas of knowledge representation, deductive databases, and logical reasoning is made. Various semantics of logical rule based languages that influenced the RIF decisions are also analyzed. Main classes of application cases of the interoperable rule based programs used for development of the requirements for RIF are considered. Finally, the main decisions of the RIF project are overviewed.

Keywords:  language unification; language extensibility; logic programming systems; active rule systems; production systems; knowledge representation; deductive databases; logical models of reasoning; stratified semantics; stable model of a logic program; well-founded semantics; RIF dialects; RIF Framework

COGNITIVE STUDY OF AN ASSISTIVE MULTIMODAL USER INTERFACE FOR HANDS-FREE HUMAN–COMPUTER INTERACTION.

• A.A. Karpov  St. Petersburg Institute for Informatics and Automation, Russian Academy of Sciences (SPIIRAS), karpov@iias.spb.su

Abstract: The process of research and development of a multimodal user interfaces aimed at hands-free interaction with a personal computer by means of the natural speech input and pointing gestures/movements by user’s head is described. The proposed interface uses a low-cost audio-visual equipment for information input and provides an universal access to computer systems both for regular human-operators for contactless (without using hands) personal computer control, and for handicapped users having difficulties with hand motion or even without arms. The methods and results of quantitative evaluation of speech and performance and contactless human–computer interaction are described and comparison with contact-based means of information input is made.

Keywords:  multimodal user interface; automatic speech recognition; computer vision; cognitive research

A LOGIC OF BIOGRAPHICAL FACTS.

• N.A.Markova  IPI RAN, nMarkova@ipiran.ru

Abstract: A method for formalizing biographical facts in the form of logical formulas is proposed. The method enables to integrate and analyze data obtained from heterogeneous sources and allows to improve the efficiency of the reference-informational service of biographical resources.

Keywords:  biographical research; information retrieval; formalization; biographical fact

CALCULATION AND OPTIMIZATION OF SOME CHARACTERISTICS OF THE MODEL FOR COMPUTATIONAL COMPLEX.

• I. V. Pavlov  BaumanMoscow State Technical University, ipavlov@mbstu.ru

Abstract: The problem of optimal packet sizing while processing large volume information is considered. In the model of a computer system, possible failures or faults of elements during tasks execution are taken into account. An asymptotic solution is obtained for the case of highly reliable components and small packets transfer time.

Keywords: optimal packet size; reliability; failure rate; transfer time

FUZZY VARIABLES AS A TOOL FOR EXPRESSING ERROR CHARACTERISTICS IN DATA PROCESSING.

• K.K. Semenov  St. Petersburg State Polytechnic University, semenov.k.k@gmail.com

Abstract: The main application of fuzzy variables theory is to take into account badly formalized information about uncertainty sources. In this paper, the results of another application are presented: to take into account wellformalized information about errors that is classical for metrology. The principal possibility of error characteristics representation as fuzzy variables that is compatible with metrological norms and standards has been shown.

Keywords:  fuzzy variables; error characteristics; measurement results

SEMANTIC SEARCH OF NATURAL LANGUAGE INFORMATION ON THE BASIS OF KNOWLEDGE BASE TECHNOLOGY.

• M.M. Sharnin  IPI RAN, keywen1@mail.com
• I. P. Kuznetsov   IPI RAN, igor-kuz@mtu-net.ru

Abstract: A system for semantic search in the natural language texts is considered. The search is based on the use of a linguistic processor which analyzes the text in natural language and extracts from it the information objects (named entities), their signs, links, and participation in actions. As a result, the processor forms the semantic structures in the Knowledge Base. The structure of quiries is formed by analogy. The semantic search consists in comparison of such structures and finding the unknown objects. In the process, the connections between objects and their participation in the actions are taken into account.

Keywords:  semantic search; semantics oriented linguistic processor; knowledge extraction from texts; Knowledge Base

ON THE RATE OF CONVERGENCE TO THE NORMAL LAW OF RISK ESTIMATE FOR WAVELET COEFFICIENTS THRESHOLDING WHEN USING ROBUST VARIANCE ESTIMATES.

• O. V. Shestakov  Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. LomonosovMoscow State University; IPI RAN, oshestakov@cs.msu.su

Abstract: The asymptotic properties of risk estimate for thresholding wavelet coefficients of signal function are analyzed. Some estimates for the rate of convergence to the normal law are obtained.

Keywords:  wavelets; thresholding; risk estimate; normal distribution; rate of convergence