Software & Systems

Articles of journal № 4 at 2014 year.

Order result by: Public date | Title | Authors

21. Methodology of training recurrent artificial neural network with dynamic stack memory [№4 за 2014 год]
Authors: Lila V.B., Puchkov E.V.
Visitors: 14264
The work proposes to generalize recurrent artificial neural networks by adding the delay feedback signals of the hidden layer to a few bars in the form of a dynamic stack memory. It allows providing an adaptive memorization of the past time events and creating a flexible tool to create nonlinear models. The proposed universal architecture of a recurrent artificial neural network with dynamic stack memory generalizes such networks as the multilayer perceptron, Jordan and Elman neural networks, as well as neural networks with feedback neurons. The paper presents a training methodology of standard neural network architecture for solving the problem of time series forecasting. It is based on the learning sample transformation. The feedbacks from a hidden layer or outputs of the neural network are eliminated by adding feedback signals to the training set. The opportunities of NeuroNADS neural network emulator have been expanded to implement the proposed methodology of recurrent neural network with dynamic stack memory training. The article presents a new object-oriented model and main software classes of neural network emulator. Based on the data for 2010–2011 there has been predicted the average monthly density of solar activity at a 10,7 cm wavelength for the first six months of 2012. After analyzing it was concluded that the recurrent artificial neural network with dynamic stack memory can be trained using the proposed methodology and the constructed models of artificial neural networks can be used for time series forecasting.

22. Verification and validation methods for complex software systems [№4 за 2014 год]
Authors: Demin A.A., Karpunin A.A., Ganev Yu.M.
Visitors: 9751
The paper discusses the problems of the large-scale software systems development. It shows the necessi-ty of using specialized quality control systems for the final product. It also suggests an algorithm for discipline review, as well as user requirements review in terms of joint design and software products d evelopment. There are the main advantages of special software means for validation and verification comparing to other software systems estimation and compliance testing methods. The article is primarily oriented at software developers and production managers wanting to automate the collabo-rative code review and verification process. It is also useful for quality assurance engineers eager to reduce labor costs at the testing stage, to detect errors in the program code at the earliest stages of system’s lifecycle. Relevance of the ar-ticle is determined by the usage of contemporary and comprehensive methods to ensure software code development’s flexibility and transparency. They allow reducing development deadlines, improving the quality and reliability of the process, reducing costs, bringing all the code to a common styling and refining the mechanisms of interaction within the team, contributing to the evolution of mutual trust between the members of development and testing teams.

23. Enterprise information resources modeling for process organization of enterprise management system [№4 за 2014 год]
Authors: Shvedenko V.N., Veselova N.S.
Visitors: 11866
The article describes a method for enterprise resource modeling for process organization of enterprise man-agement system. The authors suggest a new object-oriented data model that includes business processes as mandatory com-ponents with a pre-set schedule delivery of initial data to the information system. The initial data supply is the result of a business process in the centers where the information about input data, output data, resources and decision-making results is formed. There is a detail review on the subject area information and reference system based on the object-oriented framework "COBRA++". The paper also includes an objects classification. The authors form five groups of enterprise information sys-tem objects: simple directories, complex directories, control objects, management objects, indicators. The paper shows the examples of varying complexity system objects created with "COBRA++" software package. Model creating is performed using the basic concepts of algebra-logic; the description is represented using predicates. The authors define the set of system objects, describe their construction rules, formalize the rules that specify object relations. In each case, the description of a predicate is accompanied by a truth set that defines clear criteria for the suitability of predicate arguments. Special attention is paid to objects characteristics and properties, such as a name, ID and a type. According to the objects formalization method, the authors specify the object system fixation criteria, as well as the unique object identifier generating rules. The paper pro-poses to consider a business process as a set of fixed control objects and indicators. Using "COBRA++" object-process tech-nology, a mathematical model of the information management system is implemented. It includes descriptions of business processes and corresponding domain objects. It gives a mathematical representation of the processes of collecting, using and presenting initial data and includes necessary classified sets of objects and processes, as well as their description rules con-sidering existing relationships.

24. Simulation of potential distribution in the operation area of the [№4 за 2014 год]
Author: Masalsky N.V.
Visitors: 10115
The article discusses the simulation questions of transistor structures characteristics that are created according to a new generation technology "germanium on an insulator". It also considers one of the possible approaches to the analytical solution of a 2D Poisson equation for a potential in the operation area of a field double gate nanotransistor with "germanium on an insulator" structure and architecture "without overlapping of gate regions and a drain/source". The authors have received an analytical solution of a Poisson equation for a potential using superposition principle. The paper considers physical restrictions of technological parameters of researched transistor structures for their electro-physical characteristics optimization. In particular, for effective suppression of short-channel effects. These effects are shown more in devices under study than in similar "silicon on an insulator" structures. Based on numerical solutions of a Poisson equation the paper analyzes dependences of the main electro-physical characteristics on technological parameters. These characteristics include potential distribution in operation area, threshold voltage and a subthreshold characteristic slope. Non-linear nature of the received dependences is caused by the exponential growth of a volume charge in the transistor operation area. For selected topological norms an optimization of parameters defining drain and source regions provides an additional degree of freedom control of key characteristics along with operation area and frontal gate oxide. This is important for an applicability analysis of transistor structures "germanium on an insulator".

25. Modeling of stability of a plate stiffened by thin-walled beams [№4 за 2014 год]
Author: Chernov S.A.
Visitors: 9253
The paper describes a torsional-flexural buckling of an open profile thin-walled beam with one line of symmetry. It is assumed that there are constant longitudinal forces on the length of the beam, i.e. the beam is consid-ered weightless. The author obtained a matrix of capacity load of thin-walled beam finite element that works for bend-ing and torsion. It is based on the potential deformation energy expression with torsional-flexural buckling of such a thin-walled beam and nodal displacement energy relations. There are matrixes of thin-walled beam finite element ca-pacity load and a square box-section section beam finite element with nodes along sections outlines corresponding to the connection nodes of a plate and a beam with symmetrical mounting and in chess order. Modeling a stiffened plate is implemented in the PC program. The paper describes the characteristics of the program, its functionality and numer-ical algorithm features. There are comparative calculations under uniaxial tension of a square plate modeled by rectan-gular finite elements working to bend. The paper gives calculations results of critical load of a stiffened plate and un-stiffened plate considering beams sections and without them.

26. Fuzzy sliding mode control system of unstable dynamic system cart-pole based on soft computing [№4 за 2014 год]
Authors: Nefedov N.Yu., Ulyanov, S.V.
Visitors: 11351
The paper studies sliding mode control for cart-pole unstable dynamic system using soft computing optimizer. The authors consider three major approaches to eliminate oscillations in the sliding mode and describe the advantages and disadvantages of these approaches. The paper shows the knowledge base designing algorithm for a sliding mode controller in a new set of tools to create robust control systems – Optimizer Knowledge Bases on soft computing. It also compares knowledge bases quality created by an expert and intelligent control systems: soft-computing optimizer and ANFIS. The arti-cle gives necessary formal definitions and the corresponding illustrations. This control system showed better robustness, comparing to systems created with other modern tools.

27. On the approach to the development of the radar system constructing methodology [№4 за 2014 год]
Authors: Semenov S.A., Coban A.Y.
Visitors: 8764
The problem of constructing a long-range radar system is a complex semistructured task due to the high complexity of its structures. Therefore, the existing methodological apparatus of its construction is a set of techniques that are not full y intercon-nected. Each of those techniques is a tool to structure the system construction at some point using a variety of indicators and criteria. For example, on the stages of the functional image of information processing that is created on the basis of standard policies and procedures when designing and constructing elements of the group. For example, on the stages of the functional image of infor-mation processing that is created on the basis of standard policies and procedures in the designing and constructing group elements. Moreover, the said steps don’t formalize a common goal, an integrated measure of the achievement and its single method of con-structing. As a result, there is no possibility for comparative evaluation of the potential and real system efficiency. Automated infor-mation processing procedures that implement functionality are not enough. For targeted construction, the article suggests an ap-proach to the development of a common methodology for all phases of long-range radar system construction in the formalization of the productions and abstract problem solving with emphasis on the construction phase of ref inement options in shaping the image of the data collection and processing functional subsystem.

28. On one method of blood cell classification and its software implementation [№4 за 2014 год]
Authors: Belyakov V.K., Sukhenko E.P., Zakharov A.V., Koltsov P.P., Kotovich N.V., Kravchenko A.A., Kutsaev A.S., Osipov A.S., Kuznetsov A.B.
Visitors: 15820
A method is offered to classify the leukocytes, erythrocytes and thrombocytes. This method is based upon a comprehensive study of various segmentation methods of microscopic images and algorithms for calculation of blood cell feature sets. Our approach assumes the application of an improved combined segmentation method for microscopic images, the use of an optimized feature vector of an object and a neural network classifier. The important role in the design of our segmentation method belongs to the EDEM method for a comparative study of image processing algorithms developed in SRISA RAS. The segmentation method includes such steps as edge detection, contour closing and over-segmentation elimi-nation (based upon a set of features calculated for each initial segment). For the edge detection we use a combination of the classical Canny detector and the Ritter-Cooper method designed for blood cell segmentation. This combination comprises the advantages of both algorithms. For the boundary enhancement and contour closing steps we use an approach based upon the graph theory which develops the adaptive contour closure algorithm proposed by Jiang. The over-segmentation elimination is an iterative procedure. Our segmentation method is suitable for both red and white blood cell segmentation. To solve the blood cell classification task by a feature set we use a neural network of a multilayer perceptron type (three-layer feedforward neural network with a sigmoid function in the hidden layer). The neural network classifier allows one to effectively separate the cells into different types used in practical hematology. The program library, where the proposed classification method was implemented, is created. Our tests with various blood smear images have shown a high potential of our method for prac-tical application.

29. Spectral measure playback accuracy using permutative method of modeling stochastic processes [№4 за 2014 год]
Authors: Kuznetsov B.F., Shishkina S.V., Borodkin D.K.
Visitors: 7393
The paper considers one of the standard permutation methods of modeling stochastic processes with given spectral and proba-bilistic measures. An advantage of this method is the lack of restrictions on the combination of the probability and the spectral measure. There is a hypothesis of the reasons of appearing and structure of the general playback accuracy of autocorrelated function of modelled pro-cess using this method. The authors suggest that the modelling error consists of two parts: the variation and methodical. The error analysis method is developed. It allows making a constant value of the variation component of an error. In other words, making it independent on the autocorrelation function parameters. In turn, it allows carrying out the dependence analysis of the methodological component of the error modeling from the autocorrelation function parameters. To test this hypothesis four series of numerical experiments on modeling of stochas-tic processes were carried out using specially selected combinations of distribution laws. Distribution laws were chose n for reasons of ab-sence or presence of systematic error conditions. Based on the characteristics of the analyzed method (on the presence of lea ding and driven processes) the authors have developed a procedure for the statistical analysis of modeling results. Implementing the statistical analysis proce-dures developed by results of modeling gave no reason to reject the hypothesis of the existence of two error components: variation and me-thodical. The paper shows that using of this analysis method can significantly reduce the effect of variation in the results of systematic error analysis. It proves the method of stochastic processes modeling for solving various problems.

30. Creating a telecommunications market multiagent model [№4 за 2014 год]
Authors: Dli M.I., Gimarov V.V., Ivanova I.V.
Visitors: 9832
Economic development of the modern state is defined by the presence of advanced telecommunications system. It is one of the indicators of the investment climate in the country. The telecommunications industry is i m-portant for the information society development in Russia. It involves using information resources by citizens, public authorities and enterprises. Currently one of the most important factors increasing telecommunications companies competitiveness is their ability to respond quickly to the dynamic fluctuations of the environment. A large number of different non-interacting operations and technological infrastructure complexity obstructs taking efficient decisions. In these circumstances, it becomes appropriate to use computer decision support systems (DSS). One way of creating such systems is applying of simulation models of telecommunication companies. It helps to o b-tain a data set for choosing useful management decisions depending on operational and financial capacity of the mod-eling object. The paper presents the general structure of the multi-agent model of a telecommunications market de-signed to support decision-making in the marketing field. The process of creating multi-agent systems involves follow-ing steps: analysis, design and implementation, debugging and testing. Software implementation of a multi -agent simulation model assumes a modular structure. It is performed in Java object-oriented language using the Eclipse de-velopment tools and Microsoft Structured Query Language Server. The main emphasis is on the software implementa-tion of the model. The paper also shows the general architecture of the system, UML class diagram and code snippets in Java.

◄ ← Preview | 1 | 2 | 3 | 4 | 5 | Next → ►