On behalf of the 1st International Conference on Engineering Software and Technologies (ICSENT’12), we would like to welcome the delegates and their guests to the ICSENT’12. The ICSENT’12 was held in Hammamet, Tunisia during 15-17 December 2012. ICSENT’12 were initially created by a network of researchers and engineers both from academia and industry in the areas of Design, Analysis and Tools for Engineering Software, ICSENT’12 has been extended to the fields of Software Engineering, Computer Science and Information Technology.
The main target of ICSENT’12 is to bring together software/hardware engineering researchers, computer scientists, practitioners and people from industry to exchange theories, ideas, techniques and experiences related to all aspects of Engineering Software and technologies.
The ICSENT’12 Technical Program Committee (TPC) consists of about 100 experts in the related fields of ICSENT’12 both from academia and industry. ICSENT’12 is partnered with IAES (Institute of Advanced Engineering and Science). ICSENT’12 received 75 full-paper submissions from 13 different countries around the world. These submissions were sent to the members of the TPC and additional reviewers for review. Every submission received at least two in- dependent reviews. We would like to thank the ICSENT’12 TPC members and the reviewers for their hard work and dedication. The ICSENT’12 Technical Program includes 2 keynotes lectures which are organized into lecture sessions.
So, we would like to extend our heartfelt thanks to our co-sponsors, conference committee, program committee, reviewers, and the many other colleagues who have contributed in difference ways to the success of this event. Finally, we must thank the many authors who contributed the high-quality papers contained within these proceedings. We hope you will enjoy your stay here in Hammamet (Tunisia) and we hope you will find this conference rewarding and will continue your great support for the ICSENT’12 in the future.
Mohamed Ridda LAOUAR
ICSENT’12 Chair

The research area of software architecture and middleware has been developing rapidly. Software architecture is typically defined as the structures of the system that consist mainly of software components and interactions among them. Component is described as a unit of software that is selfcontained, self-deployable, and well-defined functionality and can be assembled with other components through its interface. The relationship of components is determined through interaction that can be described as glue to link components together. Thus, interaction is an essential part of architecture and components need to be loosely coupled. Particularly, interaction plays an important role in distributed system.
Middleware is dedicated software which is used to implement the interaction between components without involving knowledge of network protocol and hardware platform. Normally, middleware provides software architecture with mechanisms for delivering data and for supporting transparency. According to the different mechanisms, middleware can be classified as Remote Procedure Call (RPC), Object Request Brokers (including CORBA, RMI) and Message Oriented Middleware (MOM). In practice, they can be applied to different situations according to separate requirements and development environment. We will survey advanced middlewares and enterprise software architectures and discuss the challenges of these area.

The challenge posed by the increase and the diversity of information sources on the web, and secondly by the need of organizations to interoperate database systems have created not only the need to use tools for integrating data among multiple users and heterogeneous information sources, but again, these tools must overcome the limitations of current search engines by allowing not only users to ask queries more sophisticated than simple keywords, but also being able to aggregate other elements of answers from different sources to build, in the most optimized possible way by time and space research, the analytical global response to the user query. This need is becoming more and more imposed for medical information, especially with the existence of a multitude of web sources specific to medicine areas and the trend towards computerization of patient medical records.
In this distributed environment where a query involves across several heterogeneous sources, communication cost must be taken into consideration. In his conference we describe a query optimization approach using dynamic programming technique for a given set of integrated heterogeneous sources. The objective of the optimization is to minimize the total processing time including load processing, request rewriting and communication costs, to facilitate communication inter-sites and to optimize the time of data transfer from different sites. Moreover, the ability to store the data more than center site gives more flexibility in terms of Security/Safety and overload the network. In contrast to optimizers which consider a restricted search space, the proposed optimizer searches the subsets of sources and independency relationship which may be deep laniary or hierarchical trees. Especially the execution of the queries can be starting traversal anywhere over any subset and not only from a specific source.

Carbon nanotube field-effect transistors (CNTFET) modeling approach is being extensively studied as possible successors to CMOS-like. This paper introduced the analysis of the impact of diameter variations, especially logic switching threshold voltage and propagation time. The changing in diameter of CNT which is a critical parameter that can cause large swings in static and dynamic logic gates. SPICE simulations were performed on resistive-load and complementary logic inverter. The transfer characteristics as well as transient behavior were extensively studied. Finally, we have established relationships between the thresholds or propagation times and nanotube diameters.

While the well known term digital divide refers to the gap between those who can effectively use ICT tools, such as the Internet, and those who cannot, the term e-Inclusion refers to the effective participation of individuals and communities in all dimensions of the knowledge-based society and Knowledge-Base economy through their access to ICT.
The effect of ICT in industry and in the creation of new markets for e-services is evolving up, thus the eskills is becoming a requisite for workplace and for today’s lifestyle. In this paper we show that e-Inclusion polices and strategies have to be in place to widen the youth opportunities in employment and participation in the new knowledge-based economies and societies. In addition, we propose a new e-Inclusion digital gaps model which can provide policy makers with a better understanding to the digital gaps phenomena, and assist them in designing e-Inclusion polices and strategies for bridging these digital gaps.

In this paper, a new design for the deblocking filter used in H.264 CODECs is proposed. This hardware component is designed to be used as part of a complete implementation of H.264 video coding system. In the proposed implementation we adopt a new strategy for memory management. Several on-chip memories are employed to support efficient parallel access in order to speed up the entire filtering process. Previous implementations of the filter make use of one or two elementary filters, which increases the complexity of control unit and the number of multiplexers required. Therefore, a second contribution of this work is the proposition of a new architecture based on five directional filters. These filters are obtained by subdividing each initial filter into two filters, each for one direction. An additional directional filter is used specifically to filter the left neighborhood blocks. This approach ensures parallel data processing and simplifies the control unit circuitry, reducing as well the number of multiplexers. The proposed architecture is implemented in VHDL and verified to work at 137.85MHz using Xilinx ML501 platform and at 170.95MHz using Xilinx XUPV5 platform, both incorporating Virtex5 FPGA devices. The proposed architecture takes a maximum of 71 clock cycles to process one macroblock, which is about 40% less than the best of the competing proposals.

Notre objectif principal était le cryptage d’information qui est devenu obligatoire pour empêcher d’éventuelles attaques ; il peut être réalisé selon deux modèles : La cryptographie à clé secrète et celle a clé publique. Dans cet article, on a opté pour un crypto système hybride combinant la rapidité d’un algorithme symétrique et la sécurité de l’algorithme asymétrique. Celui-ci sera programmé avec un logiciel VHDL est implémenté en hardware sur un circuit FPGA de la famille Virtex-4 permettent ainsi de combiner les techniques cryptographiques symétriques et asymétriques tout en exploitant au mieux les ressources offertes par les circuits FPGA et d’atteindre de bonnes performances en termes de sécurité et répondre aux contraintes des applications en temps réel. Enfin une description matérielle du VHDL de cette architecture ainsi que la simulation par ModelSim pour vérifier les résultats obtenus.

Two decades ago, the satellites gave us global continued images that often separated by several month or years which influence the clarity of the image . The random variation of satellite attitude and the sensor defect, creates numerous special deformations . So it necessary appeared to put automatic or semi-automatic analysis methods for these images to make their comparable, interpretable, and useful. This work is devoted to the process that geometrically aligns at best the common part of multitemporal satellite images, and that by determine the transformation that provides the most accurate match between the images. This process is known by satellite image registration and it is a crucial process in many remote sensing applications. The main algorithm consists of three phases. The first one is the automated generation of control points, and the second one is the robust estimation of mapping functions from control points. Finally, a global linear spatial transformation is applied and the remotely sensed image is efficiently registered. This paper is focused essentially on a comparative study between three novel automatic and efficient feature-based approaches for high resolution satellite images registration, which are those based on Harris corner detector ; scale invariant features named SIFT and speed up robust features called Surf . The features matching phase is carried out using a bidirectional correlation approach leading to a robust set of control points in the multi temporal remote sensing images. We used the random sample consensus (RANSAC) algorithm to handle outliers resulting from the matching process. The methods has been implemented and tested using various high resolution remote sensing imagery including landsat data and spot 5 images over test sites in Kuwait, Copenhagen and “lile de Ré”. In experimental results we demonstrate the effectiveness of the proposed approaches and show that they produces accurate results when applied to the registration of high resolution satellite images as well as multitemporal ones, and we notice that SIFT based approach perform the best.

The optical fiber presents certain manufacturing defects (heart of elliptic form) or residual stresses which make anisotropic propagation. The phenomenon of PMD is due to this optical anisotropy making a fast axe and a slow one. Contrary to the other types of dispersion in fiber, the PMD varies very quickly in a dynamic way and results with a limitat ion of bandwidth. This article is organized as follows: section 2 relates to the Theory and the modeling of PMD in an optical fiber and then aims to explain briefly the PMD phenomenon. In section 3, we represent the simulated connection. Finally in section 4, we discuss these results.

The feature extraction plays a very important for face recognition. In this paper, a model is proposed for the automatic recognition of faces, by combining two approaches to intelligence Artificial. The images used are represented as pixel intensity, with a front view of the face. Each face will be described by a feature vector representing the three main components of the face are: eyes, mouth and the nose. [Brunelly and Poggio(1993)]. [Chellappa et al (1995)] Be used for the extraction of characteristic points of the treatments on the histogram of the projections of the image, the distances between these points will construct the set of feature vectors used as input for training and testing SVM multi_classe in end SVM establish its identification and final decision.

During recent decades embedded systems have become more widely used in modern society of today. These systems incorporate processors, which have become more powerful and more cost efficient. Now, embedded systems are of interest in areas where not used before, allowing for the development of ever more distributed and complex control systems. To handle the increasing complexity of such systems, industry is constantly looking for new software development tools and strategies. In this paper, we present an approach to design such systems while using an instance of a healthcare control application. Such application would support both patients and their healthcare providers in the process of treatment. It deals with the management of the conditions of patients at home while making available information on such conditions at the remote clinical center. The designed system is built around networked embedded devices and integrates the web technology.

L’objectif de ce travail est de concevoir et réaliser un outil logiciel nommé WUMA (Web Usage Mining Analyser) qui permettra au « Webmaster », qui peut être considéré comme un QoS Manager, d’avoir l’ensemble des connaissances sur le site web qu’il gère en vue de son amélioration. Cette dernière peut concerner l’adaptation du site web aux comportements de ses utilisateurs et cibler la bonne organisation de la structure du site en prenant les bonnes décisions concernant ce qu’il faut mettre comme contenu dans le site et sur le choix de la bonne forme et le meilleur agencement des pages web. Il s’agit en fait d’extraire de l’information à partir du serveur hébergeant le site Web et prendre les décisions d’amélioration de la qualité de service (QoS). Cet intérêt nous a poussés à créer et mettre en exploitation un site éducatif appartenant à l’université de la formation continue de Bordj Bou Arreridj en Algérie (www.ufcbba.net) pour mener notre étude d’amélioration de la qualité de service.

The business environment around us is a on a continuous change on a very fast pace as Cloud computing is taking over as new business environment. To cope up with the issues of process changes, such as adaptation of new or different rules and regulations at the time of an expansion of a business extends a need of its business processes to be reengineered. This article discusses in detail about three fundamental concepts of Business Process Reengineering or BPR, Systems Reengineering and Service-Oriented Architecture or also widely known as SOA. This study provides background of these concepts as well as provides the artifacts researched and work by several researchers and business analysts.

The requirement engineering process aims to provide a requirements specification document that defines the system to be developed. In this paper we focus our work on the negotiation phase for the requirements engineering process in cooperative information systems (CIS). The negotiation includes the discussion of requirements conflicts and the search of a compromise approved by all stakeholders. In order to improve this phase we propose a new negotiation approach which overcomes the following problems: requirements could be in conflicts, and the existing conflicts cannot be explicitly well expressed i.e. conflicts are often unknown. In this paper we detail the different phases of our approach with their models and finally we validate this approach by a case study.

The heterogeneity of data sources and their diversity on the web are the main difficulties encountered by users. The data integration solutions take some research on approaches to mediation, data warehouses or semi-materialized approaches, but their main difficulty lies in the automatic interpretation of the meaning and semantics of heterogeneous data autonomous. The work presented in this paper focuses on the integration of data sources based on ontology with a hybrid or semimaterialized, this approach offers a compromise between response time and the time to update the data.
In this paper, we defined in first part, an algorithm for automatic integration of data sources based ontological (HybExtendOnto). This algorithm is used to create the data warehouse schema and the virtual (mediating) based on the following principle : All existing concept in a number greater than or equal to N number of data sources or being referenced by at least N sources will be included in the materialized, otherwise it will be integrated in the virtual. In second part, we have modeled our integration architecture based hybrid ontological, the results obtained so far are deemed satisfied.

Ontology matching is a key interoperability enabler for the Semantic Web since it takes the ontologies as input and determines as output correspondences between the semantically related entities of those ontologies. We present in this paper a graph based approach to tackle the ontology matching problem. The objective is to address the combinatorial aspects related to this issue. More precisely, our approacha consists in modeling the problem of extracting an alignment (matching) which satisfies cardinality constraints, as minimizing some cost on feasible flow problem defined on a bipartite graph. The found solution represents the best alignment which maximizes the global similarity between the entities of the two ontologies.

Despite separation of concern provided by the MVC architectural style, there still remain some inter – component dependencies which constitute the liabilities of the style. In this paper we highlight the drawbacks of the MVC architectural style and propose an ontology driven solution to address these problems. In the proposed solution the ontology is used as an active run time component. The ontology is generic and can be used for any application based on MVC style. We demonstrate the usage of this style through an example implementation based on both the traditional MVC style and the ontology – driven MVC style. We evaluated the impact of ontology on the quality attributes of the MVC style. We observe that the proposed solution apart from addressing the drawbacks of MVC style also enhances maintainability and reusability of an information system based on the proposed variant style.

In this paper, we propose a semantic representation model of online multilingual educational resources, that aims to improve their search, access and reuse. This model is based on a set of ontologies, whose concepts are used for indexing learning resources that are available in a system of e-Learning. The ontological concepts and relations that we propose, provide a means of guided navigation in these resources. We also show how these ontologies can be helpful in performing multilingual search of specific learning resources and in facilitating their share and reuse for building new educational resources.

Component-Based Software Engineering (CBSE) is a knowledge-intensive discipline where all activities imply the utilization and the transfer of knowledge between collaborators. A better utilization, transfer and application of this knowledge are indispensable factors to face the challenges of CBSE. In this perspective, Knowledge Management (KM) aims at fulfilling these factors. Knowledge Management Systems (KMS) are the technological infrastructure making KM concrete. Similar to software engineering techniques, knowledge engineering techniques are used to model the KMS.
In this paper, we present our approach to model KMS for CBSE organizations in an organizationindependent context. Our approach is based on CommonKADS, one of the most adopted knowledge modeling techniques or the design of KMS.

MDA (Model Driven Architecture) aims to provide a conceptual, technological and methodological framework, in which models are central to the activities of software engineering. The classic use of MDA is to describe systems as models and their development as model transformations. Using MDA to describe the functioning itself of a system is not common or easy to perform. In this paper, we suggest the modeling of connectionist systems by models and their functioning by model transformations. A model is used to model a system in a given state among the different states that exist during its functioning. The interest is to generate from a particular model, a new model just by applying a model transformation. The generated model describes the system in the next state.

The design of component-based software architectures is performed regardless of the existing components in the warehouses, which does not promote the reuse. In this paper we present a support tool to assist the architect in the design phase of component-based software architecture, the tool implements an approach exploiting external descriptions of components, architectural elements and quality attributes. A new characterization model based on facets is proposed for components, it supports quality attributes of the ISO 9126 standard enhanced with metrics of measurement proposed to assist the architect in the evaluation of the quality of the architecture being designed.

A denial-of-service attack (DoS attack) is an attempt to prevent a service from functioning efficiently or at all, temporarily or indefinitely. This attack is typically launched against application servers or web servers and can be done by overloading the network with useless information. The intrusion detection systems are powerful tools for the detection of attempted DOS (Denial Of Service) attacks. However, they suffer from a number of problems such as high rate of false positives and negatives. In this paper, we present a new adaptive mechanism for DOS intrusion detection based on the use of agents. This self-learning mechanism ensures detection of DOS attacks with a reduction of false alarms.

Human security means protecting fundamental freedoms that are the essence of life. It also means protecting the individual against serious threats or generalized. It’s necessary to build on the strengths and aspirations of each individual, but it also means creating systems that give people environmental elements essential to their survival, their lives and their dignity. In this context, we present in this paper a new concept to create an adaptive system to save lives and help the individual in all kinds of danger, the system called Disaster Emergency system “DES”.

This paper deals with the synthesis of fuzzy tracking con- troller via virtual reference model for a nonlinear systems with distur- bance. First, a Takagi-Sugeno fuzzy model represents the nonlinear sys- tem with distrubance. Second, we combine an integral control schemes, and feedback fuzzy control whith H1 performance, to develop a robust feedback tracking controller. Third, the concept of virtual reference is used to to simplify the design of model reference. In light of this con- cept, the design procedure is split into two steps: i) Determine the virtual desired variables from the generalized kinematics; and ii) Determine the control law. The fuzzy controller is designed to track a reference signal and guaranteed a minimum disturbance attenuation level for the closed- loop system. The gains of fuzzy controller are obtained by solving a set Linear matrix inequalities (LMIs) techniques. Finally, we present an applocation on Permanent Magnet synchronous Machine. Satisfactory results validate the claims.