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 10, Issue 4, 2016

Content | About  Authors

Abstract and Keywords.

INTERFERENCE ANALYSIS OF THE DEVICE-TO-DEVICE COMMUNICATIONS MODEL WITH REGARD TO A SIGNAL PROPAGATION ENVIRONMENT
  • Yu. V. Gaidamaka Peoples' Friendship University of Russia, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • S. D. Andreev Peoples' Friendship University of Russia, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • E. S. Sopin Peoples' Friendship University of Russia, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • K. E. Samouylov Peoples' Friendship University of Russia, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • S. Ya. Shorgin Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation

Abstract: Modern dense 4G and 5G networks allow placing a wireless access point in every indoor location to optimize wireless coverage. The paper analyzes the signal-to-interference ratio (SIR) for wireless systems operating in adjacent premises, and takes into account the loss of signal power during penetration of a signal through different environments. As analytical formulas for SIR are cumbersome, two approximations of the probability density function of interference are developed: an analytical expression for the total interfering signal in the form of the normal distribution and a three-parameter hyperexponential distribution with numerically assorted parameters for simulation of the interfering signal from each device. As a result, the last one gives the most accurate approximation, which can be used for lower-bound estimation of SIR. The authors analyzed SIR in the case of the uniform distribution of devices in the premises of the rectangular shape for five wall materials (glass, brick, concrete, masonry blocks, and reinforced concrete) and varying wall thickness, and found that the presence of walls between rooms fundamentally changes the structure of SIR density.

Keywords: wireless networks; signal-to-interference ratio; indoor propagation; wall penetration

THE POISSON THEOREM FOR BERNOULLI TRIALS WITH A RANDOM PROBABILITY OF SUCCESS AND A DISCRETE ANALOG OF THE WEIBULL DISTRIBUTION
  • V. Yu. Korolev Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • A. Yu. Korchagin Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • A. I. Zeifman  Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation, Vologda State University, 15 Lenin Str., Vologda 160000, Russian Federation, ISEDT RAS, 56-A Gorky Str., Vologda 16001, Russian Federation

Abstract: A problem related to the Bernoulli trials with a random probability of success is considered. First, as a result of the preliminary experiment, the value of the random variable п € (0,1) is determined that is taken as the probability of success in the Bernoulli trials. Then, the random variable N is determined as the number of successes in k € N Bernoulli trials with the so determined success probability п. The distribution of the random variable N is called п-mixed binomial. Within the framework of these Bernoulli trials with the random probability of success, a "random" analog of the classical Poisson theorem is formulated for the п-mixed binomial distributions, in which the limit distribution turns out to be the mixed Poisson distribution. Special attention is paid to the case where mixing is performed with respect to the Weibull distribution. The corresponding mixed Poisson distribution called Poisson-Weibull law is proposed as a discrete analog of the Weibull distribution. Some properties of the Poisson-Weibull distribution are discussed. In particular, it is shown that this distribution can be represented as the mixed geometric distribution. A two-stage grid algorithm is proposed for estimation of parameters of mixed Poisson distributions and, in particular, of the Poisson-Weibull distribution. Statistical estimators for the upper bound of the grid are constructed. The examples of practical computations performed by the proposed algorithm are presented.

Keywords: Bernoulli trials with a random probability of success; mixed binomial distribution; Poisson theorem; mixed Poisson distribution; Weibull distribution; Poisson-Weibull distribution; mixed geometric distribution; EM-algorithm

ASYMMETRIC LINNIK DISTRIBUTIONS AS LIMIT LAWS FOR RANDOM SUMS OF INDEPENDENT RANDOM VARIABLES WITH FINITE VARIANCES
  • V. Yu. Korolev Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • A. I. Zeifman  Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation, Vologda State University, 15 Lenin Str., Vologda 160000, Russian Federation, ISEDT RAS, 56-A Gorky Str., Vologda 16001, Russian Federation
  • A. Yu. Korchagin Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation

Abstract: Linnik distributions (symmetric geometrically stable distributions) are widely applied in financial mathematics, telecommunication systems modeling, astrophysics, and genetics. These distributions are limiting for geometric sums of independent identically distributed random variables whose distribution belongs to the domain of normal attraction of a symmetric strictly stable distribution. In the paper, three asymmetric generalizations of the Linnik distribution are considered. The traditional (and formal) approach to the asymmetric generalization of the Linnik distribution consists in the consideration of geometric sums of random summands whose distributions are attracted to an asymmetric strictly stable distribution. The variances of such summands are infinite. Since in modeling real phenomena, as a rule, there are no solid reasons to reject the assumption of the finiteness of the variances of elementary summands, in the paper, two alternative asymmetric generalizations are proposed based on the representability of the Linnik distribution as a scale mixture of normal laws or a scale mixture of Laplace laws. Examples are presented of limit theorems for sums of a random number of independent random variables with finite variances in which the proposed asymmetric Linnik distributions appear as limit laws.

Keywords: Linnik distribution; Laplace distribution; Mittag-Leffler distribution; normal distribution; scale mixture; normal variance-mean mixture; stable distribution; geometrically stable distribution

CALCULATION OF THE ASYMPTOTIC DEFICIENCY OF SOME STATISTICAL PROCEDURES BASED ON SAMPLES WITH RANDOM SIZES
  • V. E. Bening Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation, Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation

Abstract: Statistical regularities ofinformation flows in contemporary communication, computational and other information systems are characterized by the presence of the so-called "heavy tails." The random character of the intensity of the flow of informative events results in that the available sample size (traditionally, this is the number of observations registered within a certain time interval) is random. The randomness of the sample size cruciall changes the asymptotic properties of the statistical procedures (e.g., estimators). The present paper consists of a number of applications of the deficiency concept, i. e., the number of additional observations required by the less effective procedure and, thereby, provides a basis for deciding whether or not the price is too high. The deficiency was introduced by Hodges and Lehmann in 1970. In the paper, asymptotic deficiencies of statistical procedures based on samples with random sizes are considered. Three examples concerning testing statistical hypotheses, point, and confidence estimation are presented.

Keywords: confidence set; statistical hypothesis; asymptotic deficiency; sample with random size; Poisson distribution; binomial distribution

REGIME SWITCHING DETECTION FOR THE LEVY DRIVEN ORNSTEIN-UHLENBECK PROCESS USING CUSUM METHODS
  • A. V. Chertok Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation, Sberbank of Russia, 19 Vavilov Str., Moscow 117999, Russian Federation
  • A. I. Kadaner Sberbank of Russia, 19 Vavilov Str., Moscow 117999, Russian Federation, Faculty of Mechanics and Mathematics, M.V. Lomonosov Moscow State University, Main Building, Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation
  • G. T. Khazeeva Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation
  • I. A. Sokolov Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation

Abstract: The article considers using a trending Ornstein-Uhlenbeck process, driven by a Levy process, for modeling financial time series. The authors demonstrate that the Levy driven model gives more flexibility to describe financial time series than the simple classical model. In particular, the Levy driven model allows modeling distributions with heavy tails, which is a common property of time series in real applications. The authors describe efficient methods for estimating model parameters using such methods as OLS (ordinary least squares) and RLS (regularized least squares). The article also solves the regime switching problem in a real time data stream. The authors built an algorithm based on CUSUM (CUmulative SUM) methods that is capable of determining regime switches consecutively as they happen online and keep model parameters up to date. Solution of the regime switching problem is important in real applications, since the dynamics of real systems tend to change over time under the influence of external factors.

Keywords: random process; mean-reverting process; Ornstein-Uhlenbeck process driven by Levy process; trending Ornstein-Uhlenbeck process; regime switch; change point detection; CUSUM algorithm

DISPATCHING TO TWO PARALLEL NONOBSERVABLE QUEUES USING ONLY STATIC INFORMATION
  • M. G. Konovalov Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation
  • R. V. Razumchik Institute of Informatics Problems, Federal Research Center “Computer Sciences and Control” of the Russian Academy of Sciences, 44-2 Vavilov Str.,Moscow 119333, Russian Federation, Peoples' Friendship University of Russia, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation

Abstract: Consideration is given to the problem of dispatching independent jobs from one flow to two parallel single server queueing systems each with an infinite capacity queue. There is one dispatcher, which immediately makes decisions where to route newly incoming jobs. In order to make the decision, the dispatcher uses only static information about the system, i. e., servers' speeds, job interarrival time distribution, and job size distribution. The dynamic information (for example, current queue sizes) is unavailable for the dispatcher. For such nonobservable queues, it is known that minimum mean sojourn time is achieved when the dispatcher uses the deterministic (periodic) policy. Even when using this optimal policy, the dispatcher also observes jobs' arrival instants but this information is left unused. The question posed in this paper is the following: Is it possible to reduce the mean sojourn time if one, in addition to the static information, also uses the historical information about jobs' arrival instants? Using simulation techniques, the authors show that the answer to the question is positive. Such policy uses static information and the estimates of the queue sizes based on multiple replications of the system's trajectory. Compared to the optimal policy, the achievable gain varies from 1,5% to 10%, and increases with the decrease of job size variance. When the job size variance is low, the proposed policy is even better than the dynamic join-the-shortest queue policy.

Keywords: dispatching; static information; parallel service; queueing system; nonobservable

BAYESIAN QUEUING AND RELIABILITY MODELS: DEGENERATE-WEIBULL CASE
  • A. A. Kudryavtsev Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation
  • A. I. Titova Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskiye Gory, GSP-1, Moscow 119991, Russian Federation

Abstract: This paper is devoted to Bayesian queuing and reliability models. In the framework of the Bayesian approach, it is assumed that the key parameters of classical systems are random and only their a priori distributions are known. By randomizing system's parameters such as the input flow intensity and the service intensity, one may randomize system's characteristics, for example, system's loading factor. The Bayesian approach can be used in the case of studying large groups of systems and devices or one system with variable characteristics. The results for probability characteristics of the system's loading factor and the probability that the claim received by the system will not be lost in the case of the system of the M | M1110 type where one of the system's parameters has a degenerate distribution and the other has the Weibull distribution are presented.

Keywords: Bayesian approach; queuing theory; reliability theory; mixed distributions; Weibull distribution; degenerate distribution

ANALYTICAL SOLUTION OF THE OPTIMAL CONTROL TASK OF A SEMI-MARKOV PROCESS WITH FINITE SET OF STATES
  • P. V. Shnurkov National Research University Higher School of Economics, 34 Tallinskaya Str., Moscow, 123458, Russian Federation
  • A. K. Gorshenin Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation
  • V. V. Belousov Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation

Abstract: The theoretical verification of the new method of finding the optimal strategy of control of a semi-Markov process with finite set of states is presented. The paper considers Markov randomized strategies of control, determined by a finite collection of probability measures, corresponding to each state. The quality characteristic is the stationary cost index. This index is a linear-fractional integral functional, depending on collection of probability measures, giving the strategy of control. Explicit analytical forms of integrands of numerator and denominator of this linear-fractional integral functional are known. The basis of consequent results is the new generalized and strengthened form of the theorem about an extremum of a linear-fractional integral functional. It is proved that problems of existence of an optimal control strategy of a semi-Markov process and finding this strategy can be reduced to the task of numerical analysis of global extremum for the given function, depending on finite number of real arguments.

Keywords: optimal control of a semi-Markov process; stationary cost index of quality control; linear-fractional integral functional

GENERALIZED STATISTICAL METHOD OF TEXT ANALYSIS BASED ON CALCULATION OF PROBABILITY DISTRIBUTIONS OF STATISTICAL VALUES
  • A. K. Melnikov STC CLSC "InformInvestGroup;" 125, Bld. 17 Varshavskoye Shosse, Moscow 117587, Russian Federation
  • A. F. Ronzhin S. A. Lebedev Institute of Precision Mechanics and Computer Engineering of the Russian Academy of Sciences, 51 Leninsky Prosp., Moscow 119991, Russian Federation

Abstract: A lot of data streams are a mixture of random and unique data. One of the properties of unique data is the nonuniform distribution of probability of encountering the data on the set of the values. The procedure of two steps is implemented for distinguishing unique data. On the first step of candidate selection, the criterion of consensus with the uniform distribution is implemented. On the second step, resource-intensive calculation in a condition of indeterminacy is performed in order to check other unique attributes of the candidates. The choice of the size of the criterion depends on the amount of resources given for the second step. The accuracy of calculation determines the quantity of overhead of the second term for processing random data and, therefore, a part of unique data loss. The paper analyzes the values of boundary parameters for which at the current level of computer technology, one can calculate the exact distribution. A generalized statistical method of text analysis, which can be used for a wide spectrum of text parameters, is developed.

Keywords: probability; exact distribution; limit distribution; statistics; criterion; frequency; algorithm complexity; performance of multiprocessor computer system; analysis method

ON THE ADVANCED PROCEDURE TO REDUCE CALCULATION OF GALOIS CLOSURES
  • A. A. Grusho Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation
  • M. I. Zabezhailo Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation
  • A. A. Zatsarinny Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation

Abstract: Formalization of similarity by algebraic operation is used as a key element of many modern intelligent data analysis methods. Nevertheless, in some important cases (e. g., in computer network traffic control, network security policy control in cloud computing environment, and some other), direct implementation of this technique is limited by the necessity to process huge amount of data in the real time mode. For example, it is necessary to intersect elements of a large set of Boolean vectors of large length to find fixed points of so-called Galois closure. An advanced algorithm of Galois closure formation is introduced. The algorithm is used to optimize the process of set closeness checking. Some examples of applications of the presented technique in computer network traffic control and deep packet inspection are discussed.

Keywords: intelligent data analysis; object similarity formalized as algebraic operation; combinatorial search optimization; header analysis; traffic control in computer networks; information security in cloud computing environment

SYSTEMS ENGINEERING APPROACHES TO THE ESTABLISHMENT OF A SYSTEM FOR DECISION SUPPORT BASED ON SITUATIONAL ANALYSIS
  • A. A. Zatsarinny Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation
  • A. P. Suchkov Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilova Str., Moscow 119333, Russian Federation

Abstract: The article discusses the issues of decision-making support systems (DMSS) creation based on the situational analysis of the current and projected situation in the controlled space. Typically, such control systems in real time are based on situational centers, which are sets of information, software, hardware, and staff implementing information technology to monitor the situation and its situational analysis to develop solutions and algorithms application of control actions. The paper considers characteristics of the DMSS components, implementing the full management cycle from goal setting to execution control decisions. It is noted that the implementation of the decision support system depends on the level of management - strategic, operational, tactical, basic, and functional features and methods of analysis of the situation at different levels of the control system.

Keywords: situational analysis; system of decision-making process support; management system; situational center

INFORMAL AXIOMATIC THEORY OF THE ROLE VISUAL MODELS
  • A. V. Kolesnikov Immanuel Kant Baltic Federal University, 14 A. Nevskogo Str., Kaliningrad 236041, Russian Federation, Kaliningrad Branch of the Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 5 Gostinaya Str, Kaliningrad 236000, Russian Federation
  • S. V. Listopad Kaliningrad Branch of the Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 5 Gostinaya Str, Kaliningrad 236000, Russian Federation
  • S. B. Rumovskaya Kaliningrad Branch of the Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 5 Gostinaya Str, Kaliningrad 236000, Russian Federation
  • V. I. Danishevsky Immanuel Kant Baltic Federal University, 14 A. Nevskogo Str., Kaliningrad 236041, Russian Federation

Abstract: The relevance of creation of the informal axiomatic theory of the role visual models is caused by modeling visual-imaginative reasoning in hybrid and synergistic intelligent systems. Most of the research in the field of visual-imaginative reasoning is focused on developing special visual languages to represent certain kinds of data, information, and knowledge. The lack of formal models of the visual languages is the cause of high research and development intensity of special media for handling and processing of visual models. Creation of the informal axiomatic theory of the role visual models is a step to a new class of intelligent systems that are relevant to the real decision-making teams, i. e., hybrid intelligent systems with heterogeneous visual field, imitating cooperation, complementarity, and relativity of collective intelligence, reasoning using the symbolic and visual languages.

Keywords: hybrid intelligent system; heterogeneous visual field; visual language; semiotic system

FEATURE-BASED TIME-SERIES CLASSIFICATION
  • M. E. Karasikov Moscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny, Moscow Region 141700, Russian Federation, Skolkovo Institute of Science and Technology, Skolkovo Innovation Center, Building 3, Moscow 143016, Russian Federation
  • V. V. Strijov A. A. Dorodnicyn Computing Center, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: The paper is devoted to the multiclass time series classification problem. The feature-based approach that uses meaningful and concise representations for feature space construction is applied. A time series is considered as a sequence of segments approximated by parametric models, and their parameters are used as time series features.
This feature construction method inherits from the approximation model such unique properties as shift invariance.
The authors propose an approach to solve the time series classification problem using distributions of parameters of the approximation model. The proposed approach is applied to the human activity classification problem. The computational experiments on real data demonstrate superiority of the proposed algorithm over baseline solutions.

Keywords: time series; multiclass classification; time series segmentation; hyperparameters of approximation model; autoregressive model; discrete Fourier transform

DATABASE OF RUSSIAN IMPERSONAL VERBAL CONSTRUCTIONS.
  • Anna A. Zalizniak Institute of Linguistics, Russian Academy of Sciences, 1-1 Bolshoy Kislovskiy pereulok, Moscow 125009, Russian Federation, Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
  • M. G. Kruzhkov Institute of Informatics Problems, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: This article presents the Database of Russian Impersonal Verbal Constructions that has been developed to support a linguistic research of Russian impersonal verbal construction as mirrored in translations into other languages. This information resource was developed based on the concept of Supracorpora Databases (SCDBs). Translation correspondences in the Database ofRussian Impersonal Verbal Constructions are presented as ordered pairs that combine formal descriptions of corresponding lexical-grammatical forms found in source and target texts of a parallel corpus. The paper also provides description of the methodology for creation of translation correspondences in the database. Some of the problems related to the task of finding Russian impersonal verbal constructions in corpora are considered and approaches to solving those problems are proposed. Thanks to integrated search and statistical functions, the Database of Russian Impersonal Verbal Constructions and other SCDBs significantly extend capabilities of linguistic experts using corpus-based methods to analyze specific linguistic items, both independently and in contrast with other languages.

Keywords: computer linguistics; contrastive linguistics; information technologies; electronic corpora; supracorpora databases; Russian language; impersonal constructions