Meaningful OLAP analyses often need to examine correlated data, i.e., data coming from multiple facts. Constellation schemes provide for this kind of need. In order to help the decisional designer construct efficient data marts, we propose a method that discovers and constructs constellation schemes directly from the relational database of the enterprise information system. For this, we define the concept of parallel-relations on which we base this construction and, a set of construction rules. Parallel-relations materialize interrelated business activities and therefore are good candidate for constellation construction. The construction rules extract multidimensional concepts; they build facts and measures on parallel-relations and, dimensions and their attributes on relations connected to parallel-relations. Our rules have the merit to produce dimensional attributes organized in hierarchies and, they keep track of the origin of each multidimensional component in the generated data mart schema. This trace is fundamental for the ETL processes.
Keywords: Decisional support system, Parallel-relations, Fact constellation construction, relational data source.
Assia Tebib and Mahmoud Boufaida
The e-government implies an interoperability between the various existing administrative services so that the work of institutions can be more transparent for the citizen, more reliable and economic. This interoperability is the key problem in the development of the e-government services. In this context, we define an approach to implement the mechanism of
interoperability of the information system of the egovernment. We propose an applicatif protocol based on TCP/IP which allows the technical interoperability by ensuring the data exchange between the Public Administration.
A case study was developed to validate the approach suggested and some interfaces developed were presented.
Keywords: E-government, Interoperability, TCP/IP, Protocol.
Mounira Zerari and Mahmoud Boufaida
Workflow Management Systems (WfMS) support the definition, management and execution of business processes inside organizations. However, they suffer from lack of flexibility. They are inappropriate to support easily some mechanisms such as cooperative interactions and exchanging data. Indeed, one of the most important requirements emerging from new
workflow applications that must be permanent for the process definition is the ability of communicating with outside organizations (partners, clients, and suppliers) and incorporating them in a process. In the last years, Web Services are emerged as a major evolution in technology to provide relationships among interenterprises and enterprises-costumers. Hence, we try to exploit this to integrate a web service in WfMS. However, those enterprises rely on the wide variety of
languages and technologies which are heterogeneous inpermanent evolution. Consequently our proposed approach is based on principles of MDA (Model Driven Architecture).
Keys words: Business process, Models, mapping, B2B.
In this work, we propose a solution to solve the problem of ambiguity in interbusiness portals. We define by ambiguity, any type of terminological variation defining the same object or aspect, that can be due to an ambiguity related on the language (synonymy and polysemy), or to the difference in terminological choice of the various companies. To minimize the appearance of this problem with an aim of improving collaboration of the companies, we propose a solution based on a semantic homogenisation of the terminology used on the portal. This semantic homogenisation would be concretized by the development of ontology.
Keywords: knowledge portal, semantic ambiguity, ontology, and equivalence.
Mounira Ben Abdallah, Nadia Ben Saïd, Jamel Feki and Hanene Ben-Abdallah
A multidimensional pattern (MP) is a generic solution of typical analytical requirements in a given domain. It can be instantiated to specify particular OLAP (On- Line Analytical Processing) requirements, to build the corresponding data mart (DM) schema and eventually to load the schema from a data source. This paper presents a toolset, called MP-builder, that implements our MP construction method. It illustrates the functionalities of MP-builder through an MP analyzing the "sales" fact in the commercial domain.
Keywords: Decisional support system, multidimensional pattern, data mart design, commercial domain fact.
Djamila Hamdadou, Yahya Slimani and Bouziane Beldjilali
Our study falls under the perspective which aims at optimizing the quality of decision brought to the spacetime decision-making process. Our aim is to claim with an extensible, generic, deterministic and multicriterion model based on the axiomatic of models representing decision strategies and authorizing interaction between criteria. The suggested approach is constructive, interactive and based on uncertainty theories (fuzzy logic, possibility theory, fuzzy integrals) and linear programming. We define a new approach as well for the description of available information as for their use and suggest replacing the additivity property in the performance aggregation phase by a more reliable property: the growth using non- additive aggregation operators resulting from the capacity theory and largely known as fuzzy measurements. The latter allow evaluating space compatibility between the available data by defining a weight on each subset of criteria and fuzzy integral, more specifically; the Choquet's integral is an aggregation operator able to consider the interaction among these criteria.We elaborate, in this paper a spatial decision support system. The latter is based on a combined use of Geographical Information Systems (GIS) and Multicriterion Analysis Methods namely the (ordinal and nominal) sorting approaches to claim the territorial (spatial) context analysis. This study allows the professionals to carry out a diagnostic
and proposes adapted actions in the resolution of two Territory Planning problems: The first relates to the search of a surface better satisfying certain criteria and the second consists in realizing the land use plan.
Keywords: Spatial Decision Support System (SDSS), Territory Planning (TP),Multicriterion Analysis (MCA), Geographical
Information System (GIS), Fuzzy Measurement.
MUMTAZ MOHAMMED ALI AL-MUKHTAR
The spam has now become a significant security issue and a massive drain on financial resources. In this paper, a
spam filter is introduced, which works at the server side. The proposed filter is a combination of antispam techniques. The
integrated solution create a spam filtering system which is more robust and effective than each of the comprising
techniques. The task of proposed filter is to minimize the ability of the spammers to distract the network by the spam. That is done by blocking the spam messages at the server level. A server-based solution is normally more advantageous than protecting e-mail users individually. Such a solution gives more control to administrators.
Keywords: e-mail spam, spam filtering, machine learning, blacklisting, RBL.
This paper proposes a Reliable Distributed Firewall System (RDFS), which is a client-server network paradigm. The system
consists of two elements: a Distributed Firewall-Client, which captures every transmitted or received packet, originated from
or received by the client machine, and then applies the filtering rules on these packets. The second element is the Firewall-
Controller designed as a user friendly GUI, which manages the Firewall-Clients on all the machines. It can read, write or
modify the rules for each client individually through authenticated and secured communication channels. Each Firewall-Client
uses the Firewall-Hook Driver on Windows platform as the firewall application. The proposed distributed system addresses the shortcomings of the conventional firewall as being the networks' bottleneck.
This is done by decentralizing the packets filtering processes and making them work independently. This would increase the
system availability and at the same time protect against internal attacks, which is unfeasible using the conventional firewall
setup. In addition, the system can be configured for fail-over mode, by imposing a dual controller. This would enhance the
overall system availability remarkably. Policy optimization techniques for rules and security policies are developed to reduce
the processing requirement per packet, thus faster filtering speed can be achieved.
Keywords: Firewall, Network Security, Security Management, Firewall-Hook Driver, Dual-Redundancy.
KHALID MOHAMMAD NAHAR, OSAMA MAHMOUD ABU ABBAS AND MOHAMMAD AHMAD TUBISHAT
Encryption is the process of translating data into a secret code. This paper intends to introduce a new encryption system (HRO system). This system is a combination of three known algorithms: hash based encryption algorithm, RSA algorithm, and one-time pad algorithm. The purpose of HRO system is to expedite the encryption and decryption processes and to reduce the vulnerability of the system to external attacks. HRO system uses hash function and random number generator to increase security. Some conclusions that are arrived at concern the performance of the algorithm, hash function, random number generator and the time of encryption and decryption.
Keywords: Encryption, Decryption, Hash Based Encryption, RSA, One-time pad, Collisions.
Dr. Hana'a M. Salman
Natural language text watermarking, means embedding the watermark into a text document, using the natural language components as a carrier, in such a way that, modifications are imperceptible to the readers and the embedded information is robust against possible attacks. The watermark embedding scheme can either embed the watermark directly into host data or to a transformed version of the host data. In this paper a transform watermark embedding scheme is used for watermarking natural language text is proposal by using the spectrum method. Also explore various ways of natural language text watermarking.
The results show that, the proposal technique is a successful one in implementing methods like, natural language text watermarking using spectrum transformation
Keywords: Natural Language Watermarking, Spectrum Method, Transform Method.
SALAM A. ISMAEEL, WESAM A. SHOKER and THAHA A. TAHA
An anomaly based intrusion detection systems needs to be able to learn user's or system's behavior because
users and systems behavior changes over time in today's dynamic environment. In this research experimenting with
user's behavior will used as parameters in anomaly intrusion detection. The proposed intrusion detection system is
uses a back propagation neural network to learn user's behavior.
The neural network will check if it able to classify normal behavior correctly, and detect known and unknown
attacks without using a huge amount of training data.
The experiments were separated into three parts. The first preliminary experiment was conducted to see when
the neural network was properly trained to classify sessions correctly. In this experiment, both known and unknown
attacks were used. The next experiment was conducted to test the neural network with a small traffic, known and
unknown attacks. Unknown attacks are the most threatening attacks, because these attacks are not known or not
expected. In the final experiment, the classification rate was 82% on known attacks.
Keywords: Intrusion detection; Anomaly IDS; Back propagation training algorithm; Neural Network.
OSAMA MAHMOUD ABU ABBAS, KHALID MOHAMMAD NAHAR AND MOHAMMAD AHMAD TUBISHAT
Data encryption is the translation of data into a form that unintelligible without a deciphering mechanism. This paper is intended to introduce a new approach for encryption, ARAE cipher system. It uses Arabic letters and their diacritics for encrypting English messages and vice versa. A pseudo random generator is used to generate integer numbers to represent each character in Arabic language. The same numbers are used again after sorting them to represent the English characters. The conclusions that are extracted indicate the efficiency of ARAE system according to security and time performance.
Keywords: Encryption, Decryption, Pseudo Random Generator, Arabic Characters, RSA, Sorting Algorithm.
SATTAR J ABOUD, MAMOUN S. AL RABABAA and MOHAMMAD A AL-FAYOUMI
The security on many public key encryption schemes relied on the intractability of finding the integer factoring problem. However, there is a great deal of research related to the RSA factoring scheme compared with the other similar schemes such as Rabin factoring algorithm. This paper will present a new technique for factoring the Rabin scheme. The suggested algorithm aims to factoring the Rabin modulus using a new idea that based on a new idea of attack. The new idea claimed to be more efficient than the well known algorithm that is Pollard rho algorithm since it is faster and takes less running time.
Keywords: Public key cryptography, RSA scheme, Rabin scheme, integer factoring problem, Pollard rho algorithm.
Manal Badi Mahmood and Sabah Hussein Ali
One of the basic parts which constitute the study of spatial information is the map projection. All projections have mathematical formulas which define the relationship between the latitude/longitude graticule on the earth and its representation on the map sheet, or the relationship between the geographic coordinates (latitude and longitude) of points and their projected coordinates
(grid or rectangular coordinates) on the map. Adopting this concept, the mathematical derivations of the map projection are used in this paper as the basis for the design of software application which will be called MLM (Multi-Layer Mapping). The map projection model used here is a product of two conformal projection, these are :
1- Projection of ellipsoid onto a sphere.
2- Projection of the sphere onto a plane.
The actual geographical coordinates of the earth ellipsoid are used in these projections within the MLM in order to display the interesting layers which constitutes the geographical map of the study area on the computer screen, the MLM dose also an inverse calculation of projection, i.e. when a user selects a point on the screen, it displays the corresponding values of the geographical
coordinates for the selected point. The MLM provides the following main functions: display the grid of angles (latitude and longitude) for the study area, display multilayer features choosing by the user, producing geographical spatial information for any point on the map, computing of the distance between any two selected points on the stereographic plane, finally the printing of the study map.
Keyword: stereographic projections, conformal projection, geographical coordinate, GIS, spatial data, ellipsoid.
Dr. Abdallah Alashqur
The increase in power and capacity of hardware systems coupled with the decrease of hardware costs made it possible
for institutions and corporations to store larger quantities of data in their database and warehouse systems than ever
before. Multi-terabyte databases are becoming more widely spread than in the past. This creates a need to improve the
performance of data retrieval and data manipulation operations in such large databases. Techniques such as bitmap
indexes, materialized views, and partitioning have been incorporated in many state-of-the-art database management
systems. In this paper, we provide an overview of the different partitioning techniques that have been introduced in the
literature, then we present the results of an analysis that quantitatively demonstrates the positive impact that partitioning
can have on query performance in database and data warehouse systems.
Key Words: database fragmentation, distributed database, disk reads, Oracle, indexing.
Yasser. A. Shehata and Paul. White
Best subsets regression is often used to identify a good regression model. The standard approach to assess statistical significance for a best subsets regression model is flawed. A computationally intensive randomization algorithm which corrects the problem is outlined and implemented. Simulation studies show that this procedure corrects a non-trivial problem that exists independent of
sample size and is a procedure that is robust to the presence of influential observations. This procedure leads to a simple decision rule even with correlated predictors unlike the use of a single probe. The proposed method is shown to retain power in a non-null situation.
Keywords: Best subset regression, randomization, probe variable, Type I error, bias.
Dr. Salah A. Salih
Dr. Abdul-RAZZAK T. Ziboon
Aseel Abbas Salman
Soil is an essential part of any terrestrial ecosystem. Physical and chemical properties have studied for many years for agriculture and soil conservation. These studies usually require field sampling and laboratory analysis that are time-consuming. Remotely sensed data are an alternative that provide reliable information at low cost save in efforts and time. The objective of this study was to evaluate the usefulness of Landsat ETM+ data and GIS techniques to classify soil in wet land area (Marsh region) south of IRAQ.
In this study, satellite remote sensing data have been processed and manipulated in computerized GIS manner to build-up digital information database in order to detect the soil classification and some of its chemical properties that affect on the spectral response of soil in the overall study region. The main results of this study show that the selected visible bands in the digital visual
interpretation process are considered as the best for identifying soil-mapping units. The digital map of unsupervised classification gives good presentation of some of the main land cover classes and merges the others, whereas the supervised classification gives good presentation of the main land cover classes with overall accuracy of (99.7%).
QASEM A. AL-RADAIDEH
Dataset partitioning problem involves the vertical partitioning of the classification datasets into suitable subsets that preserve or enhance the classification quality of the original datasets. Typical classification model needs to be constructed for each
subset and all generated models are then combined to form the classification model. This paper presents a dataset partitioning approach for rough set based classification. In this approach, the dataset is partitioned into two mutually exclusive subsets.
Local reduct set is generated for each attribute subset which is then combined and used to generate the set of classification rules. A preliminary experimental result using the partitioning approach over some standard medical datasets showed that the approach preserves the classification accuracy.
Key words: Data Mining; Rough Sets; Feature Set Decomposition; Classification.
Djamal BENNOUAR1, Tahar KHAMMACI2 and Abderrezak HENNI
Nowadays, Modeling of component's port is typically based on interfaces, which heavily constrain the definition of an application's architecture. This is mainly due, to the fact that, software architecture imported only the general concepts of its fundamental elements from related fields such as computer and network architecture, and did not show interests on how these concepts are organized and used in these fields in the process of defining miscellaneous architecture. To limit the interface constrains we have defined a port model, inspired from these related fields where the activity of defining architecture has reached a high degree of maturity. The port model is completely
independent from interface concept. It allows the free manipulation of its internal structure and the specification of various controls over port and component. With these capabilities, the port model opens a new way, not supported by nowadays software architecture tools, to specify any topology an architect can imagine. In addition, the port represents one of the fundamentals elements supporting the aspect orientation of our approach to software architecture. The aspect orientation is supported through aspect ports, representing aspect's join point, which must be connected to specific aspect components.
Keywords: Software Architecture, Component, Port, ArchJava, Aspect.
F. BELALA, F. LATRECHE1 and M. BENAMMAR
The primary purpose of an ADL (Architecture Description Language) is to specify the structural composition of a software system in terms of system's components and connectors through the means of a formal representational language. Many ADLs have emerged recently, none of them addresses formal analysis and verification of distributed architecture with a tractable model and an efficient mechanisable technique. In this paper, we explore the possibility of using Rewriting Logic model (via its Maude language) for specifying SADL architectural systems, showing how to conceive a behavior specification of systems using Maude concepts and rules. With them we do not only obtain a high
level specification of SADL architecture system behavior, but we are also in a position to formally reason about and prototype the specification design produced and prototype it.
Keywords: ADLs, Properties Formal Analysis, Rewriting Logic.
Laïd Kahloul and Allaoua Chaoui
Code mobility technologies attract more and more developers and consumers. Numerous domains are concerned, many platforms are developed and interest applications are realized. However, developing good software products requires modeling, analyzing and proving steps. The choice of models and modeling languages is so critical on these steps. Formal tools are powerful in analyzing and proving steps. However, poorness of classical modeling language to model mobility requires proposition of new models. The objective of this
paper is to provide a specific formalism "labeled reconfigurable nets" and to show how this one seems to be adequate to model different kinds of code mobility.
Keywords: code mobility, modeling mobility, labeled reconfigurable nets, mobile agent.
Saleh Alhazbi and Aman Jantan
Componentware seems to be a promising methodology for software development in order to cope with software complexity. With componentware, the software
development is shifted from building every thing from scratch into just assembling existing components. Therefore, Components must be integrated through well-defined infrastructure. This paper presents a component model and a framework for composing component-based systems based on message-pattern interaction among the components.
Keywords : Component-based system, message-based interaction pattern, connectors.
MAHMOUD KHRAIWESH and ASIM EL SHEIKH
Requirements management measures help organizations to understand, control and assess requirements management process. The goal of this paper is to validate a set of requirements management measures. The measures were defined for the five specific practices of Requirements Management Key Process Area (KPA) in Capability Maturity Model Integration (CMMI) by applying the Goal Question Metrics (GQM) paradigm to the five specific practices. We have made a questionnaire to prove the validity and reliability of the defined measures and confirm that they really measure the five specific practices. The questionnaire was filled by practitioners in six institutions specialized in software development. The collected data were analyzed by cronbach alpha in SPSS.
Keywords: Software Requirements, Requirements Management, Measures, Measures Validation.
Khader M. Titi , Saad Al-Aani and Riyadh Al-Shalabi
Nowadays, new information and communication technologies have become major resources and basis for learning in higher
education. Technologies have several potentials to support different instructional strategies and provide an efficient way of delivering electronic course material and improving comprehension. The contemporary universities need to increase lifelong learning opportunities to its students any time, any place and at any rate to be successful in the global learning marketplace .
The use of e-leaning in the learning process has grown significantly in the last few years, however, it is a relatively insecure, hence, most learning organization haven't yet taken into considerations any new strategy for securing elearning process . Additionally, implementing elearning is complex. Implementing e-learning is about project management, change management and risk and security management . The main aim of this research is to provide a new mechanism to protect the e-learning material from unauthorized distribution, to protect the ecourse material from being altered or modified from any intruders and to provide new tools, strategies and mechanism to make the e-learning process more secure and trustee.
Key words: Security, e-learning, Encryption.
Makhlouf Derdour and Nacira Ghoualmi-Zine
The uncertain future of the electronic market places (e-MP) brings to wonder about the development of these inter-organizational information systems that joins three types of actor mainly: the operator, the customers and the suppliers. Our work aims to propose an architecture based agent to improve management of e-MP: to add or to withdraw an agent of a system without
altering the other components and without to restart again the market place. It will allow us to modify the traditional approach of the commerce by integrating more intelligence, autonomy and automation; which are the most important features of the software agent. We present the necessity of the inescapable consumer/supplier's interfaces easily implemented with the advent of the generalization the object approach and the principle of encapsulation that are associated to them. This can be conceived henceforth like an interaction of software entities, by communication interfaces/programming that they propose. In conception stage we present the existing use cases diagrams in e-MP illustrated by UML. To implement our system we choose Jade platform (Java Agent
Develops Frame-work) and Java language. Since our work is based on a Multi-Agents system where agents must communicate between them, we model the exchanged information with KQML language (Knowledge Query Modeling Language) that permits to
define a uniform communication between different agents of the e-MP system. We translate the model of information in XML vocabulary and message's transfer messages by RMI/IIOP/Java event according to the type of communication. Results are:-The widening of the EMP's offer, -Agent's integration to measure the performances of the enterprise based upon E-MP, -The security of the exchanges, -The interoperability between E-MP, -flexibility of e-MP.
Keywords: Electronic Place of Market (E-PM), multiagents system (SMA), Customer, Seller, E-Commerce.
AMJAD MAHMOOD, TAHER S. K. HOMEED, KHALID E. JASSIM and
ABDULLATIF M. SEMIHAN
The diversity and availability of new training media are growing and providing valuable tools for organizations to train and develop their employees. Of particular growing interest is the medium of computer-based training (CBT). The main purpose of this research study is to demonstrate the problems and possibilities of implementing an on budget, on time, quality, easy to use and easy to run CBT courses. Bahrain-Civil Aviation Affaires (CAA) organization has been selected as a case study for this research. The research investigates CAA employees' attitudes towards CBT in terms of their interest in getting involved in it and the extent of
preparation they have to use it. In other words, the research identifies and empirically tests factors that may influence CAA employees' use and acceptance of CBT as a tool for delivering training in future. The results revealed that successful implementations of CBT will depend on several factors and it concluded that there is strong opportunity to implement CBT in CAA.
Keywords: Computer-Based Training, CBT, e-learning, Web-Based Training.
The aim of this paper is to design and implement a framework for e-learning and web teaching system in delivering (courses, lessons, examples, exercises, exercise solutions, self evaluation test, send and receive report) via web technology and database system. We proposed an e-learning system which is divided into three interfaces (1-Administrator 2-Instructor 3- Student). This system has been developed to provide elearning solutions over Internet or intranet and represents the new technology of education, significantly improving the learning process especially for universities. In the proposed system we used the
1- HTML (Hypertext Markup Language). 2- SQL Server 2000 (Structured Query Language). 3- ASP (Active Server Page).4- ADO (ActiveX Data Object). Proposed system design consists of:-
1- System structure for e-learning system.
2- Flow diagrams for (1-Administrator interface 2-Instructor interface 3-Student interface 4-E-learning system)
3- Flowcharts for implementation
(1- Administrator interface 2- Instructor interface 3- Student interface).
4- Database diagram for proposed e-learning system.
Keywords: : E-learning,E-learning web tools, system design.
This paper investigates the move to a new era of learning by using mobility technology. M-learning is distinct in its focus on learning across contexts and learning with mobile devices. It takes advantage of learning opportunities offered by using mobile devices and wireless transmission to enhance the existing learning management systems (LMS) that can potentially provide important opportunities for learning and collaborative interaction. Moodle is one of the most attractive LMS around the world due to its
outstanding features and reliability. Although portable technology offered anywhere-anytime discussions, the technical problems encountered when using the WAPenabled phones hindered participation in the ediscussions. In addition to presenting technical aspects of the WAP, the paper also introduce the advantages and disadvantages of using mobile technology in the learning process. Furthermore, this paper discusses the suitability and feasibility of using WAP technology devices for distance learning in real-time.
Keywords: e-learning, m-learning, WAP, LMS.
Haoua Cheribi, Salima Hacini and Zizette Boufaïda
Mobile agent applications are various. They include electronic commerce, personal assistance or parallel processing. The use of mobile agent paradigm provides several advantages. However, it has introduced some serious problems and has emphasized existing security issues. Current researches efforts in the mobile agent security area follow two aspects: (i) protection of the
hosts from malicious mobile agents, (ii) protection of the mobile agent from malevolent hosts. This paper focuses on the second point. It deals with the protection of mobile agent from eavesdropping attacks. The proposed approach is based on a dynamic adaptability policy supported by a reflexive architecture. The idea relies on the fact that mobile agent behave differently
and in unforeseeable manner during its life cycle. This ability complicates analysis attempts and protects it. In order to show the feasibility of the proposed security strategy, we propose to illustrate it throw an e-business application, implemented in Java language using Jade platform.
Keywords: Mobile agent protection, Malicious hosts, Eavesdropping attacks, Dynamic adaptability, Reflexivity, e-business.
Brahim BOULEBTATECHE, Mouldi BEDDA and Hadj Ahmed ABASSI
Reinforcement Learning (RL) is a class of model-free learning control methods that can solve Markov Decision Process (MDP) problems. However, one difficulty for the application of RL control is its slow convergence, especially in MDPs with continuous state
space. In this paper, a modified structure of RL is proposed to accelerate reinforcement learning control. This approach combines supervision technique with the standard Q-learning algorithm of reinforcement learning. The a priori information is provided to the RL learning agent by a direct integration of a human operator commands (a.k.a. human advices) or by an optimal LQ-controller, indicating preferred actions in some particular situations. It is shown that the convergence speed of the supervised RL agent is greatly improved compared to the conventional Q-Learning algorithm. Simulation work and results on the cart-pole balancing
problem and learning navigation tasks in unknown grid world with obstacles are given to illustrate the efficiency of the proposed method.
Keywords: Supervised Reinforcement Learning, Autonomous Agents, LQ-controller, Machine Learning.
AHMAD K. AHMAD and FADHIL A. ALI
A computer aided design (CAD) by using optimization methods for designing magnetic lens, by mixing the dynamic programming procedure and artificial intelligence technique. CADION package has been designed and written in Java expert system shell ( JESS 6.1 ) and Visual Basic 6 ( VB6 ) for optimizing and analyzing full calculation processes. The optimized axial magnetic flux density according to the constraints has been used in the design of pole pieces.
Keywords: magnetic lenses, dynamic programming, artificial intelligence, aberration.
Dr. Bayez Al-Sulaifanie and Dr. Ahmed M. Al-Kababji
Automatic speaker recognition systems use machines to recognize a person from a spoken phrase. These systems can operate in two modes: to identify a particular person or to verify a person's claimed identity. Personal identity verification is an essential
requirement for controlling access to protected resources or in forensic applications. One of the still challenging fields in speaker
recognition is to verify a person in a multi-speaker environment. The multi-speaker recognition task has been in the NIST (National Institute of Standards and Technology) evaluation plan since 1999. However, the researches done in this field are not sufficient and there are very few as compared the single speaker recognition systems. In this paper, a new speaker recognition system was proposed and tested for one and multi-speaker task. The system was constructed of a wavelet decomposition front end followed by a linear predictive coding cepstral feature extractor. The matching process was accomplished by a probabilistic neural network (PNN) with a background model as an imposter reference. The proposed system was able to reach an equal error rate
(EER) of 2.35% for a one speaker male gender dependent system and an EER of 3.35% for a one speaker female gender dependent system. For the two speakers and at a target to imposter ratio (TIR) of 3db the male gender dependent system had an EER of 8.7%, while for female system the EER value was 17%. The TIMIT corpus was used as the system evaluation database.
Keywords: Speaker recognition, Speaker verification, Wavelet, Probabilistic neural network, Multi speaker.
Dr.Yhya R. M. Kuraz, Prof.Dr.Waleed Ameen Mahmoud Al-Jawher
Wavelet transform is a useful tool for function estimation and signal processing; nevertheless wavelets are limited to small dimensions. The combinations of wavelet transform (WT) and neural networks (NN) will lead to overcome these limitations of large dimension efficiently. Wavelet networks (WN)s have been developed for their abilities of self-learning and self organizing. WNs have been successfully demonstrated to have potential in many applications. Although, controller design can be considered as signal
approximation, and wavelet transform is a powerful tool in this field, yet its applications in the process control and identification areas have not been investigated deeply . In this paper a new structure of Fuzzy Wavelet Network (FWN) is proposed to identify
Multi-input Multi-output complex nonlinear systems. It was found that the FWN performance depends on the selection of mother wavelet basis function and the associated number of wavelons. The FWN is used to replace the linearization feedback
of a robot arm that has four inputs and four outputs. Thus the FWN was employed as an identifier and it gave good results and fast convergence for the non parametric function under consideration in comparison with conventional Neural Network as well as, it was
shown that one set of data is sufficient during off-line learning. In Two Flexible Joints Robot Manipulation under consideration, the static learning of FWN structure can achieve robustness behavior in dynamic control. The stiffness of the FWN to avoid change of robot parameters was up to ±50% of its nominal values.
Keywords: identification, wavelet network, fuzzy logic, fuzzy wavelet network and robot manipulation.
Abdelmalek AMINE, Zakaria ELBERRICHI, Michel SIMONET and Mimoun MALKI
The classification of textual documents has been the subject of many studies. Technologies like the web and numerical libraries facilitated the exponential growth of available documentation. The classification of textual documents is very important since it allows the users to effectively and quickly fly over and understand better the contents of large corpora. Most classification
approaches use the supervised method of training, more suitable with small corpora and when human experts are available to generate the best classes of data for the training phase, which is not always feasible. The non supervised classification or "clustering"
methods make emerge latent (hidden) classes automatically with minimum human intervention, There are many, and the SOM (self Organized Maps) by kohonen is one the algorithms for non-supervised classification that gather a certain number of similar
objects in groups without a priori knowledge. In this paper, we propose non supervised classification of textual document using the SOM of kohonen and based on a conceptual representation of the texts.
Keywords: Non-supervised classification, descriptors, Similarity, self-organizing maps of Kohonen, Wordnet.
Raida El Mansouri
Business process models describe how a business works, or more specifically, how they accomplish missions, activities, or tasks. The automated control and coordination of business processes is made possible by task control constructs that model behaviors like
concurrency, asynchronism, and choice. However, there is a real danger of introducing control flow anomalies and behavioral inconsistencies like deadlock, livelock, imperfect termination, and multiple task repetitions . Petri Nets provide a powerful formal modeling method based on solid mathematical fundament while having graphical representation of system models as net
diagrams and provide various analysis techniques such as reachability tree, incidence matrix and invariant analysis method, through which properties of the Petri Net model such as liveness, reachability and deadlock can be analyzed. This paper proposes an approach to illustrate the use of the Petri Net INA (Integrated Net Analyzer)  environment for formalizing business process
specifications and using analytical techniques to support verification studies. The first step is automated.
Keywords: Business process modeling, Petri Nets, INA, Verification, Graph transformation, Meta Modeling.
O. Badawy, U. Aburawash and Y. Hassan3, M. Abdeen
The need for intelligent systems has grown in the past decade because of the increasing demand on humans and machines for better performance. The researchers of AI have responded to these needs with the development of intelligent hybrid systems. This paper describes the modeling language for interacting hybrid systems in which we will build a new hybrid model of cellular automata and multi-agent technology. Simulations with complex behavior will be model social dynamics. Therefore, in our approach, cellular automata form a useful framework for the multi-agent simulation model and the model will be used for traffic and computer network systems. In this paper, we will analyze, describe, and design environments in which agents can operate effectively and interact with each other productively. The environment or cellular automata grid will provide a computational infrastructure for such interactions to take place. The infrastructure will include protocols for agents to communicate and interact with each other. The rules that control the agents' growth, death, and behavior have been designed locally for each agent.
M.Badeche and M.Benmohamed
In real time applications of augmented reality, it is always matter of tracking; one of the most promising techniques in this sense is tracking based pattern. In this paper we describe a method based on the discrete Kalman filter for a real-time tracking of a 2D
pattern for augmented reality. Objective being to make Real-time tracking of a 2D pattern through a video stream on the simple personal computer; the process can be resumed as follow: in each frame we do an estimation by Kalman filter of predict corner location, after we proceed to correct this estimation by a measure around predict location, with Harris corner detection. the results obtains are very satisfactory, because all the process precede on real time and succeed to accurately track the pattern through video stream.
KEYWORDS: Augmented reality, corner tracking, Realtime tracking, Kalman filter, corner detection.
I. Jarrah, M. Shdefat and A. Rjoub
An implementation of an efficient Iris recognition algorithm is presented in this paper. Its based on reducing the number of instructions and increasing its speed operation to be used properly by portable and handheld equipment, so that battery life is
increased. This is achieved by using techniques minimize the number of instructions in the code. The proposed technique is used in the three stages of the paper: Segmentation, Normalization and finally Encoding. Comparison results with other tools verified the validity of the proposed technique.
Keywords: Digital Image Processing, Iris Recognition, Hough Transformation, Low Power Design Techniques, Low Power Code.
Face recognition is a biometric authentication method that has become more significant and relevant in recent years. It is becoming a more mature technology that has been employed in many large scale systems such as Visa Information System, surveillance access control and multimedia search engine. Generally, there are three categories of approaches for recognition, namely global
facial feature, local facial feature and hybrid feature. Although the global facial-based feature approach is the most researched area, this approach is still plagued with many difficulties and drawbacks due to factors such as face orientation, illumination, and the presence of foreign objects. This paper presents an improved offline face recognition algorithm based on a multi-local feature
selection approach for greyscale images. The approach taken in this work consists of five stages, namely face detection, facial feature (eyes, nose and mouth) extraction, moment generation, facial feature classification and face identification. Subsequently, these stages were applied to 3065 images from three distinct facial databases, namely ORL, Yale and AR. The experimental results obtained have shown that recognition rates of more than 89% have been achieved as compared to other global-based features and
local facial-based feature approaches. The results also revealed that the technique is robust and invariant to translation, orientation, and scaling.
Keywords: Face Recognition, Facial Feature Extraction, Localization, Neural Network, Genetic Algorithm (GA).
MOHAMED FEZARI, Hamza Attoui
Abstract: An approach based on spotted word recognition system is applied to control a set of autonomous robots. The methodology adopted is based on hybrid techniques used in speech recognition which are zero crossing and extremes with dynamic time warping followed by a decision system based on independent methods test results. To test the approach on a real application, a PC interface was designed to control the movement of five mobile robots using a radio frequency transmission. Also, it was checked whether the basic commands used would be enough to control the robots. Some experiments are tested by actually running a set of powered-toy vehicle and each robot is secured from frontal collision by some sensors.
Keywords: Voice command, hybrid technique, spotted words, Robots.
Mr. AHMED M. HASSAN
This paper describes a new method of removing additive white noise of known variance from photographic images. This method is based on a characterization of statistical properties of natural images represented in complex wavelet decomposition. Specifically, we decompose the noisy image into wavelet subbands, denoising the subbands using wavelet. The proposed method is more efficient compared to use the wavelet techniques only. In this paper six methods are used to estimate the threshold value.
Keyword: WN (Wavelet Network), GSI (Gray Scale Image), DWT (Discrete Wavelet Transform), LOD (Level of Decomposition), NN (Neural Network).
Dr. Mohamed Abdullah Hussain, Dr. Mumtaz Mohamed Ali and Dr. Wrya Mohamed Ali
Internet based process had widespread in the industry field in the recent years. Little documentation can be found on this due to company polices to protect
their projects. In our region the case is different and it is rare to find internet based process control. Simplicity in any system is an asset to have. Especially, in our
region in which we lack hardware and software company dealers. It is best to depend on free software as far as is possible in addition to on simple easy
repairable systems. In this research a hardware interface circuit and a software system has been designed to control the temperature and level of a liquid tank (water has been considered in our case study). The main advantage of the designed interface circuit is its simplicity and low cost. The same can be true for the software system in which we used Java servlet to accomplish the communication task. Java servlets have many advantages over the CGI based methods used
previously that has common problems of speed and data persistence. In security aspects a password has been added to all control action commands to secure our process control tasks. The designed hardware and software systems could be used for remote or local control with no amendment and for both manual and automatic control.
Keywords: Remote Instruments Control, Network based Control, Network based Instruments Control, Remote Process Control, Internet based Process Control.
Basil Sh. Mahmood and Shefa A. Dawwd
Abstracat: Convolutional neural network (CNN) is a well-known robust image recognition model. It is a multi-layer architecture where the successive layers are designed to learn progressively higher-level features, until the last layer which produces categories. In order to apply this model to robot vision or various intelligent systems, its VLSI implementation with high performance is required. In this paper, a hardware implementation of one of the most complex and powerful
image recognition convolutional neural networks(the neocognitron) has been considered. The original neocognitron has been modified to reduce its complexity whether it is hardwarely or softwarely implemented while its principle working ideas has been kept. As a result to these modifications and simplifications, a relatively small FPGA hardware model of 200,000 gates has been used to implement a relatively complex design. The system is tested by using the Optical
Recognition Library(ORL) face database for face recognition problem and its performance is compared with the results obtained using advanced software system designed specifically for face recognition. The recognition rate achieved from both software and hardware versions were equal to 93%. A speed up of (88) is achieved for the parallel architecture implemented in an FPGA as compared with the computer software. Also a performance of 1GCPS is achieved and seems
reasonable when it is compared to the today available neuro-hardwares.
Keywords: Convolutional neural network implementation, neocognitron, neural hardware implementation.
K. BOUAMRANE, N. TAGHEZOUT and F. AMRANI, A M. DJEBBAR
To offer to the companies the means to choose the diagnosis tool to implement or to buy represent an important financial profit in a competing context.
Thus, to ensure the company that the proposed tool is most adequate according to there production nature forces to set up the means of identifying the
needs for each company. Development of the Multicriterion Diagnosis System MDS is a response to the problem. The MDS is an integrated tool for the choice of the most relevant diagnosis system. This tool was developed by using the multicriterion approach Electra III. While basing itself on a set of criteria as well as a
set of diagnosis tools thoroughly selected and implemented for the circumstance, namely: expert system diagnosis which models knowledge in the form of rules, Markov chain whose models knowledge in the form of probabilistic graphs realized using an iterative algorithm Baum Welch as well as the Viterbi algorithm for the generation of the failures effects sequence and Bayesians networks which formalizes the knowledge of experts in the form of a graphic model representing
conditional independences between a set of variables. Lastly, the model of failure was carried out using the current tool for diagnosis FMEA
(failures Mode and effects Analysis). The developed tool allows on the one hand, to guide the maintenance expert to choose the diagnosis
system to be adopted, and on the other hand, to carry out a fast and effective diagnosis. In order to validate our work, we applied the tool to an industrial process of an Algerian company of the iron and steel sector. This enabled us to show the effectiveness of this type of tool in term of availability and reduction of the downtime of the machines.
Keywords: Multicriterion Diagnosis system (MDS), Electra III, expert system, Markov chain, Baum Welch algorithm, Viterbi Algorithm, Bayesian network, FMEA.
Hamid BESSALAH, Fadila MOSTEFAI and Zahia BRAHIMI
In this paper, architecture for hardware implementation of the Advanced Encryption Standard (AES) Algorithm is presented. Where, encryption, decryption and key
schedule are all implemented using small resources of only 3383 Slices and 8 Block RAMs. So our implementation fits easily in a Xilinx VirtexII XC2V2000-
4FF896 FPGA. The proposed implementation can encrypt and decrypt data streams with a throughput of 235 Mbps, and a new way of implementing MixColumns
and InvMixColumns transformations using shared logic resources is presented.
Keywords: AES, Decryption, Encryption, FPGA, ECB,Images.
Maher M. F. Algreer
This paper proposed a new hybrid structure called a FPD2 controller for a simple robotic system. The controller consists of a fuzzy PD ( FPD) controller part, that has a simple rules base nine rules only, and conventional D controller (CD) part. The CD is added to the FPD to improve the transited performance and to avoid the derivative kick with out add more rules. The performance of the proposed controller is compared with the response of the conventional PD controller, showing
that the FPD2 has smaller overshot and less settling time over its CPD.
Keywords: Fuzzy Logic Control, Conventional PID Control, Robot System, Fuzzy PD.
F.Z. LAALLAM and M. SELLAMI
ABSTRACT:The activity of industrial supervision and control of the gas turbines is a very complex spot and requires a great experiment. This experiment is acquired with the passing of years what makes departure of an expert in retirement a great loss of the know-how. In the same way, these experts are not available constantly and in each site. The problem thus consists in capturing this know-how and allows experiment to be cumulate with an aim of the construction of a system of assistance to the diagnosis. The work presented in this article relates to ontological engineering and more particularly the use of ontologies in the Knowledge-based systems. After the stages of conceptualization and ontologisation , which allowed the capture of knowledge of the domain; we proceed in this article to the Operationalisation of conceived ontology. The modeling of the inference is based on the specification of the operational objective by scenario of use. The JessTab tool is used for the integration of the contexts of use in JESS. Thus we have obtained our system of assistance to the diagnosis "OntoTurb-Expert" of which the goal is not to solve the problem of diagnosis automatically, but to help the user by providing him suitable information and by leaving him the responsibility for a contextual evaluation of this information
Key words: Operationalisation, ontology, industrial gas turbine, scenario of use, diagnosis of breakdowns, JessTab.
Qais Thanon Algwari and Sabah Waad Altaie
This work describes a design of control system to orientation the solar cells using Atmel Flash microcontroller (AT89C51). The solar elevation angles over the course of a year are calculated for any position depending on longitude and latitude using C++ program. The software is convert the angles into its equivalent value of steps according to the resolution of the step motors and save the result in Hex file. The Hex file is loaded on EEPROM by EPROM programmer. The solar cells holder step motors will receive the data of sun position and will move to it.
Keywords: Sun position, micro-controller, AT89C51, C++.
Saad H. Al-Tak and Sufyan T. Faraj
Fieldbus applications suffer from severe environmental conditions. These conditions may affect the communication process among different nodes. Because
of the real-time nature of such applications, timing behaviour must be well designed and studied. PROFIBUS as one of the widely applied fieldbus
protocols is considered here. This paper evaluates inaccessibility overheads in PROFIBUS protocol in the presence of transient faults. It introduces a novel analytical model for the inaccessibility of PROFIBUS message/token transmission in the presence of transient faults. Different error scenarios are suggested to produce bestcase (BC) and worst-case (WC) error overhead evaluation that are based on the integration of single bit errors together with burst errors into a bounded fault
arrival model. The introduced error components are included in the worst-case response time (WCRT) of PROFIBUS message cycles. This work is essential to
assess the real-time behaviour of the protocol under the incidence of errors.
Keywords: Fieldbus, Fault Model, Inaccessibility, PROFIBUS, Transient Faults, WCRT.
Belabbas Yagoubi, Nadia Hadi and Meriem Meddeber
Workload and resource management are two essential functions provided at the service level of the Grid software. To improve the global throughput of these
environments, effective and efficient load balancing algorithms are fundamentally important. Although load balancing problem in classical distributed systems has
been intensively studied, new challenges in Grid computing still make it an interesting topic, and many research projects are under way. This is due to the Grid
characteristics and to the complex nature of the problem itself. This paper presents a task load balancing model in Grid environment. First we propose a tree-based model to represent Grid architecture in order to manage workload. This model is characterized by three main features: (i) it is hierarchical; (ii) it supports heterogeneity and scalability; and, (iii) it is totally independent from any Grid physical architecture. Second, we develop a hierarchical load balancing strategy to balance tasks among Grid resources. The main characteristics of the proposed strategy are: (i) It uses a task-level load balancing; (ii) It privileges local
tasks transfer to reduce communication costs; (iii) It is a distributed strategy with local decision making.
Keywords: Grid computing, Load balancing, Workload, Hierarchical model, Tasks transferring.
Dr. Ahmed T. Sadik and Dr. Rana F. Ghani
The aim of this paper is to improve the Dijkstra algorithm which is widely used in the internet routing. Quantum computing approach is used to improve the work of Dijkstra algorithm for network routing by exploiting the massive parallelism existing in the quantum environment and to deal with the demands of
continuous growing of the internet. This algorithm is compared according to the number of iterations and time complexity with Dijkstra's algorithm and the result shows that the quantum approach is better in finding the optimal path with better time complexity when it is implemented in quantum computer.
Keywords: Quantum Computing, Routing Algorithms, Computer Network.
N.TAGHEZOUT, A. RIAD and K.BOUAMRANE
Real world manufacturing environments are highly dynamic because of frequently changing situations. Deterministic mechanisms using a centralized control mechanism cannot handle the system dynamics. Agent-based approaches are particularly suitable for dynamic manufacturing scheduling. This paper proposes a negotiation approach based on multiagent system for complex manufacturing systems. The local scheduling and control function in dynamic distributed
environment is addressed by a new negotiation protocol based on a hybrid model. The agent negotiation protocol that we have developed facilitates the establishment of the solution program in real time on the basis of hybridization between the Contract Net Protocol and a particular approach which is based on artificial intelligence. The purpose of this protocol is to assign operations dynamically to the production system resources in order to accomplish the proposed task. The article describes the common resource sharing through a number of scenarios of negotiation between the initiator agents and participants; this description treats the cases of conflicts, renegotiation and waiting.
Keywords: Multi Agent System, Negotiation, Decision Support System (DSS), Dynamic Scheduling, Contract net Protocol, ISP (Integrated Station of Production).
RASHID JAYOUSI1, JIHAN AWAWDA and BADIE SARTAWI
This paper review previous models developed for mobile communication system. It outlines the mobility behaviors, communication algorithm, nodes movement model, routing protocols, and node localization technique used by such models. In this paper we present our study on the effect of the different communication parameters on the mobility behavior. Such parameters are capacity, delay, node power consumption, node localization accuracy, and fault tolerance. Based on the
models investigated we propose a variant model that can increase flexibility and scalability, decrease node power consumption; achieve high transmission rate with lower delays, and achieve fault tolerance nodes. We report preliminary results of the simulation experiment we have conducted based on the model presented in this paper.
Keywords: Mobile communication systems, Mobility, Fault tolerance.
Dr. Qutaiba I. Ali and Salah A. Jaro Alabady
As applications become more distributed, the design and management of security services in networked systems play an increasingly significant role. This paper deals with the plan and design of a typical security system for a large cooperative network. The main aim of the design is to protect the network against internal and external threats as well as various types of attacks. The design includes the ability of the administrator to control and manage the network from different locations inside the network and remotely from outside the network. First of all, the current security state of the network is examined, then, complete network security architecture is proposed. This architecture is based on supplying the network with 11 security methods against internal threats and 6 security methods against external threats. These methods have both software and hardware nature and work in all network layers. The affectivity of the suggested security solutions is tested against different attacks and proves its ability to resist these situations.
Keywords: Network Security, Network Management, Firewall, AAA Server, VP, IDS.