Abstract: With the development of wireless network technology and the advent of the cognitive radio networks, there is a need for mechanisms for effective and fair resource allocation. In this paper an overview of the possibilities of the application of cognitive approaches and game theory for solv-ing problems related to resource management in new generation telecommunication networks and some important research aspects in this field are considered. It is stated that an interesting and rel-evant scientific task to be solved is related to the development of new, more efficient methods for resource management in next generation telecommunication networks based on cognitive game approaches.
Key words: Cognitive Approaches, Game Approaches, Next Generation Telecommunication Net-works, Resource Management
Ventsislav Trifonov, Ivaylo Atanasov, Evelina Pencheva
Abstract: Security functions in Machine-to-Machine communications are studied and abstraction is identified using application programming interfaces. The interfaces can be used for mutual authentication between mobile entities and a central unit, as well as for protection of unique identities. The trust and security management interfaces are described by the use of class diagrams, state diagrams and sequence diagrams illustrating the interface application.
Key words: Machine-to-Machine communication, mobile monitoring, security
Yavor Tomov, Georgi Iliev
Abstract: The goal of this article is to evaluate the TL_MCP algorithm. The algorithm with Two Linear functions for Multi Constrained Path (TL_MCP) proposed by us is able to solve MCP when there are two constraints. As we know the MCP is NP hard. In literature there are exact and heuristic algorithms which have a good performance to find a feasible path .Few of them use linear, while others use non-linear functions of path length. We choose one exact and two heuristic algorithms with high success rate to compare them with TL_MCP algorithms.
Key words: algorithms for optimization, algorithms for routing, heuristic algorithms, multi constraint path problems
Valentin Topchiev, Lidia Jordanova
Abstract: The paper deals with the possibilities for expanding the operating wavelength range and improvement of the noise characteristics of hybrid fiber optic amplifier composed of erbium-doped fiber amplifier (EDFA) and distributed Raman amplifier (DRA). Mathematical models of EDFA and DRA, which allow research of their gain and noise figure spectrum characteristics, are presented. On the base of the results attained by conducted simulations, the parameters and operating modes of both co-amplifiers which provide maximum of gain flatness, minimum of noise figure and maximum of operation waveband are determined.
Key words: EDFA, DRA, WDM system, gain flatness, noise figure
Abstract: High performance routers are fundamental building blocks of the system wide interconnection networks for high performance computing systems. Through collective interaction they provide reliable communication between the computing nodes and manage the communicational dataflow. The development process of specialized router architecture has high complexity and it requires many additional factors to be considered. The architecture of the high-performance routers is highly dependent on the flow control mechanism, as it dictates the way in which the packets are transferred through the network. In this paper novel high-performance „Step-Back-On-Blocking” router architecture has been proposed.
Keywords: High Performance Router Architecture, System Area Networks, Flow control mechanism.
Vladimir Daskalov, Thomas SchUtzle, Angel Sinigersky, Stoyan Maleshkov
Abstract: Industrial CAE applications pose ever growing requirements for the performance of data management systems. Our research focuses on novel techniques for integration of readily available database systems into the structural optimization software Tosca. Such techniques are using domain specific knowledge in order to improve performance in terms of speed and disk usage of the database. The present paper proposes usage of hierarchical databases like HDF5 for storing and handling of optimization data and results in order to extend the capabilities of the optimization system TOSCA Structure. Both performance and software maintenance aspects of our novel approach are discussed.
Key words: CAE software, database techniques, HDF5, hierarchical data storage, structural optimization
Elizabeth Mihaylova, Milena Lazarova, Ognyan Nakov, Nadejda Angelova
Abstract: The paper presents a comparison of Java ThreadPool with .NET ThreadPool. A new NET ThreadPool is suggested that utilizes Java ThreadPool features. Based on two developed testing tools the performance parameters of using the thread pools was experimentally evaluated for different number of pools, threads, tasks and time per task.
Key words: parallel programming, thread pool, optimization, .NET ThredPool, Java ThredPool
Abstract: The increased activity of cyber-crime in the last years requires close cooperation between the CERTs from different countries and automation of processes of exchange of information on computer incidents.
Key words: Security incident, CERT, electronic data interchange
Daniela Gotseva, loannis Dimakopoulos
Abstract: This paper presents a brief overview to ‘Al (Artificial Intelligence) today. The authors show the way in which knowledge can be represented and how to make reasoning. Paper explains the rule engines, production rule systems, hybrid reasoning system, and expert systems. All explanations are from modern programming languages and technologies.
Key words: Artificial Intelligence, Rete Algorithm, Knowledge Representation, Hybrid Reasoning Systems, Expert Systems, JBoss, Java.
Ventzislav Nikolov, Stephan Panov, Emilian Janev
Abstract: The subject of the article is a model for building test environment by using a directed graph to navigate in the subject area. Each of the graph vertices represents a test unit. Test unit attributes reflect the relative weight of the current unit, its complexity and the semantic relations between the topics of the subject area on which the student knowledge will be evaluated. The formalization of the semantic dependency between the test units is achieved by using AND/OR connections between vertices. Graph navigation is adaptive and depends on the primary student knowledge and his/her current results. In this respect the graph navigation is goal/question driven process.
Keywords: machine learning, directed graph, AND/OR vertex, learning tree.