Showing papers in "International Journal of Computers Communications & Control in 2009"

Spiking Neural P Systems with Anti-Spikes

Abstract: Besides usual spikes employed in spiking neural P systems, we consider “anti-spikes", which participate in spiking and forgetting rules, but also annihilate spikes when meeting in the same neuron. This simple extension of spiking neural P systems is shown to considerably simplify the universality proofs in this area: all rules become of the form bc → b′ or bc → lambda , where b,b′ are spikes or anti-spikes. Therefore, the regular expressions which control the spiking are the simplest possi- ble, identifying only a singleton. A possible variation is not to produce anti-spikes in neurons, but to consider some “inhibitory synapses", which transform the spikes which pass along them into anti-spikes. Also in this case, universality is rather easy to obtain, with rules of the above simple forms.

Fuzzy Logic Control System Stability Analysis Based on Lyapunov’s Direct Method

Abstract: A stability analysis method for nonlinear processes controlled by Takagi- Sugeno (T-S) fuzzy logic controllers (FLCs) is proposed. The stability analysis of these fuzzy logic control systems is done in terms of Lyapunov's direct method. The stability theorem presented here ensures sufficient conditions for the stability of the fuzzy logic control systems. The theorem enables the formulation of a new stability analysis algorithm that offers sufficient stability conditions for nonlinear processes controlled by a class of T-S FLCs. In addition, the paper includes an illustrative example that describes one application of this algorithm in the design of a stable

A Framework for Enhancing Competitive Intelligence Capabilities using Decision Support System based on Web Mining Techniques

Abstract: Nowadays Competitive Intelligence (CI) represents one of the most important pieces in strategic management of organizations in order to sustain and enhance competitive advantage over competitors. There are some studies that claim that a successful strategic management is influenced by the accuracy of external environment’s evaluation and, in the same time, in order to have correct and complete business strategies it is necessary to be sustained by competitive advantage. But till at the beginning of ’80 the things were totally different. This paper will present the evolution and the objectives of CI, the results of using CI in organizations and how can be improved the CI process using tools and techniques provided by business intelligence (BI). The study will propose a framework of a decision support system based on web mining techniques in order to enhance capabilities of organization’s competitive intelligence.

P–Lingua 2.0: A software framework for cell–like P systems

Abstract: P-Lingua is a programming language for membrane computing. It was first presented in Edinburgh, during the Ninth Workshop on Membrane Computing (WMC9). In this paper, the models, simulators and formats included in P-Lingua in version 2.0 are explained. We focus on the stochastic model, associated simulators and updated features. Finally, we present one of the first applications based on P- Lingua: a tool for describing and simulating ecosystems.

Application of Genetic Algorithms for the DARPTW Problem

Abstract: On the Dial-a-Ride with time windows (DARPTW) customer transportation problem, there is a set of requests from customers to be transported from an origin place to a delivery place through a locations network, under several constraints like the time windows. The problem complexity (NP-Hard) forces the use of heuristics on its resolution. In this context, the application of Genetic Algorithms (GA) on DARPTW was not largely considered, with the exception of a few researches. In this work, under a restrictive scenario, a GA model for the problem was developed based on the adaptation of a generic GA model from literature. Our solution applies data pre-processing techniques to reduce the search space to points that are feasible regarding time windows constraints. Tests show competitive results on Cordeau & Laporte benchmark datasets while improving processing times.

Variants of P Colonies with Very Simple Cell Structure

Abstract: We study two very simple variants of P colonies: systems with only one object inside the cells, and systems with insertion-deletion programs, so called P colonies with senders and consumers. We show that both of these extremely simple types of systems are able to compute any recursively enumerable set of vectors of

Introducing a Space Complexity Measure for P Systems

Abstract: We define space complexity classes in the framework of membrane computing, giving some initial results about their mutual relations and their connection with time complexity classes, and identifying some potentially interesting problems which require further research

First Steps Towards a CPU Made of Spiking Neural P Systems

Abstract: We consider spiking neural P systems as devices which can be used to perform some basic arithmetic operations, namely addition, subtraction, comparison and multiplica- tion by a fixed factor. The input to these systems are natural numbers expressed in binary form, encoded as appropriate sequences of spikes. A single system accepts as inputs num- bers of any size. The present work may be considered as a first step towards the design of a CPU based on the working of spiking neural P systems.

Artificial Intelligence + Distributed Systems = Agents

Abstract: The connection with Wirth's book goes beyond the title, albeit confining the area to modern Artificial Intelligence (AI). Whereas thirty years ago, to devise effective programs, it became necessary to enhance the classical algorithmic frame- work with approaches applied to limited and focused subdomains, in the context of broad-band technology and semantic web, applications - running in open, heteroge- neous, dynamic and uncertain environments-current paradigms are not enough, be- cause of the shift from programs to processes. Beside the structure as position paper, to give more weight to some basic assertions, results of recent research are abridged and commented upon in line with new paradigms. Among the conclusions: a) Non- deterministic software is unavoidable; its development entails not just new design principles but new computing paradigms. b) Agent-oriented systems, to be effectual, should merge conventional agent design with approaches employed in advanced dis- tributed systems (where parallelism is intrinsic to the problem, not just a mean to speed up).

Task Resource Allocation in Grid using Swift Scheduler

Abstract: In nature, Grid computing is the combination of parallel and distributed computing where running computationally intensive applications like sequence alignment, weather forecasting, etc are needed a proficient scheduler to solve the problems awfully fast. Most of the Grid tasks are scheduled based on the First come first served (FCFS) or FCFS with advanced reservation, Shortest Job First (SJF) and etc. But these traditional algorithms seize more computational time due to soar waiting time of jobs in job queue. In Grid scheduling algorithm, the resources selection is NPcomplete. To triumph over the above problem, we proposed a new dynamic scheduling algorithm which is the combination of heuristic search algorithm and traditional SJF algorithm called swift scheduler. The proposed algorithm takes care of Job’s memory and CPU requirements along with the priority of jobs and resources. Our experimental results shows that our scheduler reduces the average waiting time in the job queue and reduces the over all computational time.

An LMI Technique for the Global Stabilization of Nonlinear Polynomial Systems

Abstract: This paper deals with the global asymptotic stabilization of nonlinear polynomial systems within the framework of Linear Matrix Inequalities (LMIs). By employing the well-known Lyapunov stability direct method and the Kronecker product properties, we develop a technique of designing a state feedback control law which stabilizes quadratically the studied systems. Our main goal is to derive sufficient LMI stabilization conditions which resolution yields a stabilizing control law of polynomial systems.

Fuzzy Controller Based QoS Routing Algorithm with a Multiclass Scheme for MANET

Abstract: A mobile ad hoc network (MANET) consists of a set of mobile hosts that can communicate with each other without the assistance of base stations. Due to the dynamic nature of the network topology and restricted resources, quality of service (QoS) and multicast routing in MANET are challenging tasks which attract the interests of many people. In this paper, we present a fuzzy controller based QoS routing algorithm with a multiclass scheme (FQRA) in mobile ad hoc networks. The performance of this scheduler is studied using NS2 (Network Simulator version 2) and evaluated in terms of quantitative measures such as packet delivery ratio, path success ratio and average end-to-end delay. Simulations show that the approach is efficient, promising and applicable in ad hoc networks.

High Performance Computing Systems with Various Checkpointing Schemes

Abstract: Finding the failure rate of a system is a crucial step in high performance computing systems analysis. To deal with this problem, a fault tolerant mechanism, called checkpoint/ restart technique, was introduced. However, there are additional costs to perform this mechanism. Thus, we propose two models for different schemes (full and incremental checkpoint schemes). The models which are based on the reliability of the system are used to determine the checkpoint placements. Both proposed models consider a balance of between checkpoint overhead and the re-computing time. Due to the extra costs from each incremental checkpoint during the recovery period, a method to find the number of incremental checkpoints between two consecutive full checkpoints is given. Our simulation suggests that in most cases our incremental checkpoint model can reduce the waste time more than it is reduced by the full checkpoint model. The waste times produced by both models are in the range of 2% to 28% of the application completion time depending on the checkpoint overheads.

Optimization for Date Redistributed System with Applications

Abstract: In this paper we intend to define a strategy for managing databases with mobile structures, taking into account their redistribution in the nodes of a computer network. The minimal cost of the redistribution is highlighted and some applications for medical and business databases are presented.

Flatness-based Control and Conventional RST Polynomial Control of a Thermal Process

Abstract: In this paper, a comparison between conventional RST polynomial control by poles placement and RST flatness-based control is proposed. These approaches were developed, in discrete-time formalism by carrying out a generation of a reference starting from a flat output and by then analyzing the tracking error in closedloop. The case of the thermal system, that we applied an output disturbance, was considered by simulation to study the effectiveness of given flatness-based robust controller with the integration of an anti-windup device, in terms of tracking trajectory and disturbance rejection.

Presentation of an Estimator for the Hurst Parameter for a Self-Similar Process Representing the Traffic in IEEE 802.3 Networks

Abstract: The hypothesis for the existence of a process with long term memory structure, that represents the independence between the degree of randomness of the traffic generated by the sources and the pattern of traffic stream exhibited by the network is presented, discussed and developed. This methodology is offered as a new and alternative way of approaching the estimation of performance and the design of computer networks ruled by the standard IEEE 802.3-2005.

A Note on the Generative Power of Axon P Systems

Abstract: Axon P systems are a class of spiking neural P systems. In this paper, the axon P systems are used as number generators and language generators. As a language generator, the relationships of the families of languages generated by axon P systems with finite and context-free languages are considered. As a number generator, a characterization of the family of finite sets can be obtained by axon P systems with only one node. The relationships of sets of numbers generated by axon P systems with semilinear sets of numbers are also investigated. This paper partially answers some open problems formulated by H. Chen, T.-O. Ishdorj and Gh. Paun.

Fuzzy Logic in Genetic Regulatory Network Models

Abstract: Interactions between genes and the proteins they synthesize shape genetic regulatory networks (GRN). Several models have been proposed to describe these interactions, been the most commonly used those based on ordinary differential equations (ODEs). Some approximations using piecewise linear differential equations (PLDEs), have been proposed to simplify the model non linearities. However they not allways give good results. In this context, it has been developed a model capable of representing small GRN, combining characteristics from the ODE’s models and fuzzy inference systems (FIS). The FIS is trained through an artificial neural network, which forms an Adaptive Nertwork-based Fuzzy Inference System (ANFIS). This network allows to adapt the membership and output functions from the FIS according to the training data, thus, reducing the previous knowledge needed to model the specific phenomenon. In addition, Fuzzy Logic allows to express their rules through linguistic labels, which also allows to incorporate expert knowledge in a friendly way. The proposed model has been used to describe the Lac Operon in E. Coli and it has been compared with the models already mentioned. The outcome errors due to the training process of the ANFIS network are comparable with those of the models based on ODEs. Additionally, the fuzzy logic approach provides modeling flexibility and knowledge acquisition advantages.

Bayesian Network Classifier for Medical Data Analysis

Abstract: Bayesian networks encode causal relations between variables using probability and graph theory. They can be used both for prediction of an outcome and interpretation of predictions based on the encoded causal relations. In this paper we analyse a tree-like Bayesian network learning algorithm optimised for classification of data and we give solutions to the interpretation and analysis of predictions. The classification of logical – i.e. binary – data arises specifically in the field of medical diagnosis, where we have to predict the survival chance based on different types of medical observations or we must select the most relevant cause corresponding again to a given patient record. Surgery survival prediction was examined with the algorithm. Bypass surgery survival chance must be computed for a given patient, having a data-set of 66 medical examinations for 313 patients.

Incremental Improvement of the Evaluation Algorithm in the Concept Map Based Knowledge Assessment System.

Abstract: The paper is devoted to the knowledge assessment system that has been developed at the Department of Systems Theory and Design of Riga Technical University for the last four years. The system is based on concept maps that allow displaying the knowledge structure of a particular learner in the form of a graph. Teacher’s created concept maps serve as a standard against which learner’s concept maps are compared. However, it is not correct to compare teacher’s and learners’ concept maps by examining the exact equivalence of relationships in both maps, because people construct knowledge in different ways. Thus, an appropriate mechanism is needed for the flexible evaluation of learners’ concept maps. The paper describes the algorithm implemented in the concept map based knowledge assessment system and its evolution through four prototypes of the system.

Modeling of Errors Realized by a Human Learner in Virtual Environment for Training

Abstract: This study focuses on the notion of erroneous actions realized by human learners in Virtual Environments for Training. Our principal objective is to develop an Intelligent Tutoring System (ITS) suggesting pedagogical assistances to the human teacher. For that, the ITS must obviously detect and classify erroneous actions produced by learners during the realization of procedural and collaborative work. Further, in order to better support human teacher and facilitate his comprehension, it is necessary to show the teacher why learner made an error. Addressing this issue, we firstly modeling the Cognitive Reliability and Error Analysis Method (CREAM). Then, we integrate the retrospective analysis mechanism of CREAM into our existing ITS, thus enable the system to indicate the path of probable cause-effect explaining reasons why errors have occurred.

Development Journey of QADPZ - A Desktop Grid Computing Platform

Abstract: In this paper we present QADPZ, an open source system for desktop grid computing, which enables users of a local network or Internet to share resources. QADPZ allows a centralized management and use of the computational resources of idle computers from a network of desktop computers. QADPZ users can submit compute-intensive applications to the system, which are then automatically scheduled for execution. The scheduling is performed according to the hardware and software requirements of the application. Users can later monitor and control the execution of the applications. Each application consists of one or more tasks. Applications can be independent, when the composing tasks do not require any interaction, or parallel, when the tasks communicate with each other during the computation. The paper describes both QADPZ functionality and the process of design and implementation, with focus on requirements, architecture, user interface and security. Some future work ideas are also presented.

Mutation Based Testing of P Systems

Abstract: Although testing is an essential part of software development, until re- cently, P system testing has been completely neglected. Mutation testing (mutation analysis) is a structural software testing method which involves modifying the pro- gram in small ways. In this paper, we provide a formal way of generating mutants for systems specified by context-free grammars. Furthermore, the paper shows how the proposed method can be used to construct mutants for a P system specification.

Applying RBF Neural Nets for Position Control of an Inter/Scara Robot

Abstract: This paper describes experimental results applying artificial neural networks to perform the position control of a real scara manipulator robot. The general control strategy consists of a neural controller that operates in parallel with a conventional controller based on the feedback error learning architecture. The main advantage of this architecture is that it does not require any modification of the previous conventional controller algorithm. MLP and RBF neural networks trained on-line have been used, without requiring any previous knowledge about the system to be controlled. These approach has performed very successfully, with better results obtained with the RBF networks when compared to PID and sliding mode positional controllers.

A Novel Fuzzy ARTMAP Architecture with Adaptive FeatureWeights based on Onicescu’s Informational Energy

Abstract: Fuzzy ARTMAP with Relevance factor (FAMR) is a Fuzzy ARTMAP (FAM) neural architecture with the following property: Each training pair has a rel- evance factor assigned to it, proportional to the importance of that pair during the learning phase. Using a relevance factor adds more flexibility to the training phase, allowing ranking of sample pairs according to the confidence we have in the infor- mation source or in the pattern itself. We introduce a novel FAMR architecture: FAMR with Feature Weighting (FAM- RFW). In the first stage, the training data features are weighted. In our experiments, we use a feature weighting method based on Onicescu's informational energy (IE). In the second stage, the obtained weights are used to improve FAMRFW training. The effect of this approach is that category dimensions in the direction of relevant features are decreased, whereas category dimensions in the direction of non-relevant feature are increased. Experimental results, performed on several benchmarks, show that feature weighting can improve the classification performance of the general FAMR algorithm.

An Immuno-Genetic Hybrid Algorithm

Abstract: The construction of artificial systems by drawing inspiration from nat- ural systems is not a new idea. The Artificial Neural Network (ANN) and Genetic Algorithms (GAs) are good examples of successful applications of the biological metaphor to the solution of computational problems. The study of artificial immune systems is a relatively new field that tries to exploit the mechanisms of the natural immune system (NIS) in order to develop problem- solving techniques. In this re- search, we have combined the artificial immune system with the genetic algorithms in one hybrid algorithm. We proposed a modification to the clonal selection algo- rithm, which is inspired from the clonal selection principle and affinity maturation of the human immune responses, by hybridizing it with the crossover operator, which is imported from GAs to increase the exploration of the search space. We also in- troduced the adaptability of the mutation rates by applying a degrading function so that the mutation rates decrease with time where the affinity of the population in- creases, the hybrid algorithm used for evolving a fuzzy rule system to solve the well- known Wisconsin Breast Cancer Diagnosis problem (WBCD). Our evolved system exhibits two important characteristics; first, it attains high classification performance, with the possibility of attributing a confidence measure to the output diagnosis; sec- ond, the system has a simple fuzzy rule system; therefore, it is human interpretable. The hybrid algorithm overcomes both the GAs and the AIS, so that it reached the classification ratio 97.36, by only one rule, in the earlier generations than the two other algorithms. The learning and memory acquisition of our algorithm was ver- ified through its application to a binary character recognition problem. The hybrid algorithm overcomes also GAs and AIS and reached the convergence point before

The Impact of Java Applications at Microarchitectural Level from Branch Prediction Perspective

Abstract: The portability, the object-oriented and distributed programming mod- els, multithreading support and automatic garbage collection are features that make Java very attractive for application developers. The main goal of this paper consists in pointing out the impact of Java applications at microarchitectural level from two perspectives: unbiased branches and indirect jumps/calls, such branches limiting the ceiling of dynamic branch prediction and causing significant performance degrada- tion. Therefore, accurately predicting this kind of branches remains an open problem. The simulation part of the paper mainly refers to determining the context length in- fluence on the percentage of unbiased branches from Java applications, the prediction accuracy and the usage degree obtained using a Fast Path-Based Perceptron predictor. We realize a comparison with C/C++ application behavior from unbiased branches perspective. We also analyze some Java testing programs, built using design pat- terns or including inheritance, polymorphism, backtracking and recursivity, in order to determine the features of indirect branches, the arity of each indirect jump and the prediction accuracy using the Target Cache predictor.

3D Virtual Spaces Supporting Engineering Learning Activities

Abstract: Virtual environments constitute the support platform for various teaching and learning activities. Instead of a local application for this purpose, this paper explores the effects of using distributed virtual reality environments in the educational process. The architecture of the presented system is based on the recently developed web-based technology called AJAX (Asynchronous Javascript And XML), implemented on a Linux operating system configured to run Apache with PHP and MySQL support; i.e., LAMP architecture, which contributes to the portability and ease of installation of the application. The platform is designed to support the integration of lesson modules such as the EngView environment which is discussed in more detail in this contribution. Pedagogical, technical, and implementation-related aspects are presented in conjunction with the virtual environment used in the engineering training curriculum. Statistical information resulted from the training shows a significant increase in task completion time when the virtual setup is used.

Quality Control of Statistical Learning Environments and Prediction of Learning Outcomes through Reproducible Computing

Abstract: This article introduces a new approach to statistics education that al- lows us to accurately measure and control key aspects of the computations and com- munication processes that are involved in non-rote learning within the pedagogi- cal paradigm of Constructivism. The solution that is presented relies on a newly developed technology (hosted at www.freestatistics.org) and computing framework (hosted at www.wessa.net) that supports reproducibility and reusability of statistical research results that are presented in a so-called Compendium. Reproducible com- puting leads to responsible learning behaviour, and a stream of high-quality commu- nications that emerges when students are engaged in peer review activities. More importantly, the proposed solution provides a series of objective measurements of actual learning processes that are otherwise unobservable. A comparison between actual and reported data, demonstrates that reported learning process measurements are highly misleading in unexpected ways. However, reproducible computing and ob- jective measurements of actual learning behaviour, reveal important guidelines that

An algorithm for initial fluxes of metabolic P systems

Abstract: A central issue in systems biology is the study of efficient methods inferring fluxes of biological reactions by starting from experimental data. Among the different techniques proposed in the last years, the theory of Metabolic P systems, which is based on the Log-Gain principle, proved to be helpful for deducing biologi- cal fluxes from temporal series of observed dynamics. According to this approach, the algebraic systems provided by the Log-Gain principle determine the reaction fluxes underlying a system dynamics when initial fluxes are known. Here we propose a heuristic algorithm for estimating the initial fluxes, that is tested in two case studies.