A Framework for Ambient Computing 2014/CLOSER...2.1.4 Human-Computer Interaction (HCI) Is the interaction between users and computers via the user interfaces (Dnovan, 2010; Gallissot,

A Framework for Ambient Computing

Mohammed Fethi Khalfi and Sidi Mohamed BenslimaneComputer Science Department, University of Djilali Liabes, Sidi Belabess, Algeria

[email protected], [email protected]

Keywords: Pervasive Information Systems, Ambient Intelligence, Ubiquitous Computing, Smart Space.

Abstract: The proliferation of mobile computing and wireless communications is producing a revolutionary change inour information society. Ubiquitous Computing is a recent paradigm whose objective is to support users inaccomplishing their tasks, accessing information, or communicating with other users anytime, anywhere. Inother terms, Pervasive Information Systems (PIS) constitute an emerging class of Information Systems whereInformation Technology is gradually embedded in the physical environment, capable of accommodating userneeds and wants when desired. PIS differ from Desktop Information Systems (DIS) in that they encompass acomplex, dynamic environment composed of multiple artefacts instead of Personal Computers only, capable ofperceiving contextual information instead of simple user input, and supporting mobility instead of stationaryservices. In this paper, as an initial step, we present PIS novel characteristics compared to traditional desktopinformation systems; we explore this domain by o ering a list of challenges and concepts of ubiquitous com-puting that form the core elements of a pervasive environment. As a result of this work, a generic frameworkfor intelligent environment has been created. Based on various and related works concerning models anddesigns. This framework can be used to design any PIS instance.

1 INTRODUCTION

Mark Weiser was the first to describe the vision ofubiquitous computing, which has as its goal the en-hancing computer use by making many computersavailable throughout the physical environment, andmaking computers e ectively invisible to the user(Weiser, 1991). The essence of Weisers vision is thatpersons use many computers embedded in the envi-ronment, allowing technology to recede into the back-ground. The first era was defined by the mainframecomputer, a single large time-shared computer ownedby an organization and used by many people at thesame time. Second, came the era of the PC, a personalcomputer primarily owned and used by one person,and dedicated to them. The third era, ubiquitous com-puting, representative of the present time, is charac-terized by the explosion of small networked portablecomputer products in the form of smart phones, per-sonal digital assistants (PDAs), and embedded com-puters built into many of the devices we own resultingin a world in which each person owns and uses largernumbers of computers becoming integrated into ev-eryday life. In addition communications extend be-yond the classic concept man to man or man to ma-chine, to include direct communication between ma-chines, (Fig. 1).

Figure 1: Evolution of computers, from the beginning to theambient computing.

The remarkable recent progress in embedded de-vices, smart phones, wireless communications andnetworking technologies (Fig. 2) has enabled usto create pervasive computing systems and serviceswith diverse applications and global accessibility (Es-coffier, 2008), and promote their mobility in a trans-parent way without the explicit user intervention.This technological progress offers an opportunity tofocus on its main task instead of configuring andmanaging all IT equipment at their disposal and ac-

170

cess to various services offered by these objects, any-where, at any various devices (Cheikh, 2012), (Sahaand Mukherjee, 2003), (Satyanarayanan, 2001).

Figure 2: Wireless networks.

This paper is organized as follows, after review-ing the state of the pervasive computing. Section 2formulates the definition and evolution of informationsystems and the main constraints of pervasive infor-mation systems, after we present PIS novel charac-teristics compared to traditional desktop informationsystems (DIS), Section 3 identify and describe fun-damental challenges, properties and characteristics ofubiquitous computing environments that form or arepart of those environments. Section 4 reviews thebackground and related works for different architec-tural models of pervasive systems. In section 5, wedescribe out the proposed system architecture. Fi-nally, we conclude this paper in section. 6, ans wedescribed future possibilities works.

2 ISSUES AND CHALLENGES OFPERVASIVE COMPUTING:

Based on an analysis of literature in the field of ubiq-uitous computing we present an overview of funda-mental properties and characteristics of ubiquitouscomputing.

2.1 Enabling Concepts andTechnologies:

2.1.1 Wireless Networks

Wireless networks infrastructures can form the plat-form, which enable clients to transparently connectand share context with remote entities. The geo-graphic scope defines the coverage capacity of a wire-less network; we can distinguish several categories,(Fig. 3).

2.1.2 Wireless Sensor Networks (WSN)

consist of a large number of tiny devices called sensornodes that pervade an area and collaborate together

Figure 3: Pervasive Wireless Network.

to collect the information necessary to adjust systembehavior, pervasive systems usually integrate wire-less sensors, which consist of small devices capableof sensing, processing, and communicating differenttypes of sensory data. A WSN comprises a number ofcomponents, and though the terminology may changeaccording to different architectures, (Fig. 4):

Figure 4: Wireless Sensor Networks.

2.1.3 Pervasive Access Devices

Pervasive access devices constitute the front end ofPIS (Senn, 2007; Mattern and Sturm, 2003) andare likely to contain a multitude of different devicetypes that differ in size, shape (more diverse, er-gonomic, and stylistic), and functional diversity (mo-bile phones, laptops, pagers, PDAs) (fig.5, fig.6). Inessence, pervasive devices dictate the interaction be-tween the user and the pervasive system (Beigl et al.,2003). A major requirement for participation of a de-vice in a pervasive environment is connectivity. De-vices may include one or more connectivity optionsdepending on their functionality.

Figure 5: Aibo7.

A�Framework�for�Ambient�Computing

171

Figure 6: Nabaztag.

2.1.4 Human-Computer Interaction (HCI)

Is the interaction between users and computers via theuser interfaces (Dnovan, 2010; Gallissot, 2012). HCIwith ICT systems has conventionally been structuredusing a few relatively expensive access points. Thisinteraction primarily uses input from Keyboard andpointing devices which are fairly obtrusive to interactwith. Nowadays, Interaction is via natural user inter-faces and physical interactions that can involve ges-tures input, multiple touch, voice control, eye gazecontrol, etc. It can improve performance and user ex-perience through anticipated actions and user goals inrelation to the current context, past user context andgroup context.

2.2 Core Properties of PervasiveInformation Systems

2.2.1 Distributed Systems

Are considered the core of pervasive systems (Siddiqand Ali, 2010). It consists of multiple autonomouscomputers that communicate through a computer net-work (Bourcier), each device may have its own userwith individual needs, and the purpose of the dis-tributed system is to coordinate the use of shared re-sources or provide communication services to theirusers (Tanenbaum and M. Van Steen, 1999). A dis-tributed system can operate across different homoge-neous environments (Li, 2010; Dugénie; Petit) andseamlessly integrate devices with environments.

2.2.2 Mobility

Will be an important characteric of a ubiquitous sys-tem (Saadi, 2009; Louberry, 2010). There are, how-ever, different kinds of mobility schemes, such as ter-minal mobility, personal mobility, session mobilityand service mobility. Users shall be supported in sucha way that they can move from one place or terminalto another and still get a personalized service (Tigli etal,. 2009; Hoareau, 2007; Reignier, 2010; Kouici).Inthe future, the networks may also be mobile and dy-namic, and therefore, full mobility is an essential re-quirement in the ubiquitous wireless world.

2.2.3 Interoperability

Is one of the most essential requirements of ubiqui-tous software. Today, application developers’ use awide ranges of programming models, languages anddevelopment environments (Vallecillo et al.). Theneed to network the embedded products of differentvendors with cyber world applications is increasing inthe ubiquitous environment. the infrastructure for per-vasive computing must support diverse types of soft-ware component. Applications in pervasive comput-ing environments will be required to respond to noveltasks and situations, applications will increasingly beformed dynamically from available software compo-nents. This will require dynamic interoperability atthe component level, in addition to interoperabilitythat overcomes the heterogeneity of the environmentand of components.

2.2.4 Scalability

Is also crucial in pervasive computing scenarios(Duboc et al,. 2006) as smart environments may leadto highly intensive interactions between the comput-ing infrastructure and the user space (Escoffier, 2008;Najar et al., 2012; Sarr, 2010). This, in tum, canhave serious implications for the user with regards tocomputing and network resources such as bandwidth,memory and energy. The ever increasing levels ofcommunication and computing interactions betweendifferent users also adds to the pressure on resources(Bourcier) and making scalability an absolutely crit-ical requirement for pervasive systems (André; San-cho, 2010).

2.2.5 Heterogeneity

Devices present in pervasive environments can be ofvarious kinds (Duboc et al,. 2006; Shaout and Srini-vasan, 2009; Jouve, 2009): simple devices (lightor clock), mobile assistant devices (PDA or smart-phones), multimedia devices (PC or TV), etc. Thesedevices have different hardware and are based on var-ious operating systems (Flissi et al., 2005; Ferry et al,.2008; Louberry, 2010). They can have different com-munication technologies (wired/wireless) (Bourcier),and implement different network protocols (UPnP,Jini, SLP, GSM). Thus, we need to handle this het-erogeneity and provide users with an interoperablesystem. In the same network, computers with largestorage can coexist with other devices with limited re-sources (Cheung-Foo-Wo, 2009; Hoareau, 2007).

CLOSER�2014�-�4th�International�Conference�on�Cloud�Computing�and�Services�Science

172

2.2.6 Integration

Pervasive computing systems by its nature requireintegration of many different subsystems (Saha andMukherjee, 2003) with very different characteristics.These subsystems include computational facilities,communication devices, mechanical or chemical sen-sors and actuators, smart appliances, and existing con-trol systems.

2.2.7 Dynamicity

The mobility of some pervasive devices and the lim-ited resources of others increase the dynamics of per-vasive computing environments. This dynamics isperceived in terms of the number and lifetime of per-vasive functionalities a user can access to at a spe-cific time and location. (Najar et al., 2012) In particu-lar, new devices may appear in the environment whileother devices may become out of reach due to a lackof resources (e.g., battery down), or due to the rangeof radio transmissions.

2.2.8 Autonomy

Is a property of a system that allows itself to organizeand operate its own actions without human interven-tion (Bantz et al., 2003; Wikipedia, 2010). The Ubiq-uitous Computing systems are naturally autonomous,because it is always self-governing and capable of itsown independent decisions and actions. The systemcan operate without human intervention (Sarr, 2010),and both the input/output and computation of the Sys-tem are completely embedded in the device it con-trols.

2.2.9 Proaction

The system needs to be self-triggered to capture a pri-ori what its users want in order to increase the overallquality of service (Laforest, 2007).

2.2.10 Invisibility

is another key requirement for pervasive computing.In the ideal sense, invisibility implies that the under-lying technology is completely hidden from the user’sperspective (Liu, 2006). In real environments, in-visibility can exist only in the approximation sense.This could be facilitated by smart environments thatconstantly adapt to meet the user’s requirements withminimum user intervention. This problem becomesmore difficult when the user is in a dynamicallychanging environment.

2.2.11 Context Awareness

One of the most important novel characteristics thatPIS introduce is the notion of context awareness. Thisterm was formally defined and used for the first timeby (Schilit et al., 1994; Brown et al., 1997) to describeapplications that "adapt according to their locationof use, the collection of nearby people and objects,as well as the changes to those objects over time".(Abowd et al., 1999) Define context-aware comput-ing as "a system that uses context to provide relevantinformation and/or services to the user, where rele-vancy depends on the user’s task". (Pascoe, 1998)de?ne context as the subset of physical and concep-tual states of interest to a particular entity. (Jame-son, 2001) extend the previous definitions by addingthe user’s behavior and current interactions with thepervasive system. (Laerhoven and Kofi, 2001) em-phasize the importance of sensors embedded in theenvironment in order to sense the location and cur-rent movement of the user and add it to the proper-ties of a context-aware system. (Abowd et al., 1999)approach context as the user’s emotional state, focusof attention, location and orientation, date and time,objects, and people in the user’s environment. Otherdefinitions have simply provided synonyms for con-text, referring, for example, to context as the envi-ronment or situation. (Cousins and Varshney, 2009)defines context as the elements of the user’s environ-ment that the user’s computer knows about. The mostgeneric de?nition for the features of a context-awaresystem has been provided by (Dey, 2001): "a systemis context-aware if it uses context to provide relevantinformation and/or services to the user, where rele-vancy depends on the user’s task.

2.2.12 Context Management

Users in pervasive environments must access to de-vice functionalities. Moreover, they need a represen-tation of the different elements present in the environ-ment (Dey, 2001). The first requirement of a perva-sive system thus is the context management. It de-mands to represent the context enable users to getan overview of functionalities and to enable users tomanage this representation. Moreover, context man-agement also involves to be continuously aware of en-vironmental changes and to relate this according tothe context representation.

2.2.13 Adaptability

Adaptation is required in order to overcome the in-trinsically dynamic nature of pervasive computing.Mobility of users, devices and software components

A�Framework�for�Ambient�Computing

173

can occur, leading to changes in the physical and vir-tual environments of these entities. Moreover, appli-cations can be highly dynamic, with users requiringsupport for novel tasks and demanding the ability tochange requirements on the fly (Bourcier). It shouldbe the role of the infrastructure for pervasive comput-ing to facilitate adaptation, which may involve adapt-ing individual software components and/or reconfig-uring bindings of components by adding, removingor substituting components. Dynamic adaptation caninvolve complex issues such as managing the adap-tation of software components that are used simulta-neously by applications with different (and possiblyconflicting) requirements, and maintaining a consis-tent external view of a component that has behaviorthat evolves over time.

2.2.14 Security

The goal of security is also to guarantee the well func-tioning of pervasive computing systems. Ubiquitouscomputing has been designed in order to be deployedanywhere and be accessed by the majority of users.The user mobility and the proliferation of lightweightdevices make complex security problems. Recent re-search on pervasive computing focuses on buildinginfrastructures for managing smart spaces, connect-ing new devices, and providing useful applicationsand services. Privacy, trust, and security issues insuch environments, Contextual data are generally per-sonal data are static data such as user preferences andhabits, and dynamic data such as its location and thetasks it performs. Thus, these data have confidential-ity and exchange of data between these applicationsmust comply with the security measures defined bythe user.

2.2.15 Service Discovery

Is an important and challenging issue in pervasive en-vironments. Service discovery is the task of locatingwhich services are available (Brown, 1996), usuallyfollowed by a selection activity whose main goal isthe choice of the most appropriate (Zhu et al., 2005),or best solution based on user-defined metrics (e.g.accessibility, cost, available bandwidth, load, etc.) inthe presence of multiple instances of the same servicescattered throughout the network.

2.2.16 Quality of Service Management

(UIT) describes the quality of service as a set of re-quirements on the collective behavior or multiple ob-jects. The aim is to provide users with applications

that work best and as long as possible on their fa-vorite mobile device and that whatever the changesin the environment (Louberry, 2010).

3 RELATED WORK

In this section, we briefly present some researchprojects related to this topic, which are also basedon the ubiquitous concepts. We are positioning ourcontribution in relation to this work. Though muchprogress has been achieved, current architectures pro-vide very few of abstraction and often generic andlimited support. Several studies on the architectureof pervasive systems are proposed, even if they dif-fer, all of these works are in fact complementary.(Satyanarayanan, 2001; Moitra, 2004; Afyouni, 2010;Achour et al., 2012; Em and Yoo, 2005; Siddiqand Ali, 2010; Cousins and Varshney, 2009; Contiet al., 2012; Zhou et al., 2010) Adopt an architec-turally model based on new technologies (infrastruc-ture networks and communicating devices). (Sahaand Mukherjee, 2003) detailed description of ubiqui-tous computing, In their opinion, the sensors provideinformation about the context of ubiquitous systems,what makes them different from traditional systems.(Saha and Mukherjee, 2003) Introduced a model ofthe pervasive computing environment by differenti-ating between devices, pervasive networking, perva-sive middleware, and pervasive applications. (Satya-narayanan, 2001) explains that ubiquitous computingbased on distributed systems and mobile computing,adding features such as smart spaces and invisibil-ity. (Hoh, 2006) In his initial work presents a user-centric model. The authors (Cheung-Foo-Wo, 2009;Afyouni, 2010; Achour et al., 2012; Hoh, 2006) il-lustrate the importance of smart spaces where dig-ital and physical world are related in a natural andtransparent to the user. Work (Satyanarayanan, 2001;Afyouni, 2010; Em and Yoo, 2005; Siddiq and Ali,2010) integrate the field of intelligent interfaces thatallows users to control and interact with objects in-tuitively. Finally (Moitra, 2004; Em and Yoo, 2005;Hoh, 2006) address the security which is a major chal-lenge in building pervasive environment.Most of the approaches proposed in the literature arespecific to an application or to a particular domain.They are not sufficiently generic to be reusable inother domains. The key contribution of the paperis a reference model architectural has been proposedfor analyzing the functionalities and key of pervasivecomputing. Tables 2 offer an overview of the propos-als that have inspired our proposed model.

CLOSER�2014�-�4th�International�Conference�on�Cloud�Computing�and�Services�Science

174

Table 1: Summaries of research work, (-) and (+) corre-spond to an absence and a presence of the layer

HC

I

Dev

ices

Smar

tSpa

ce

Infr

astr

uctu

re

Secu

rity

Debashis SAHA, 2003 + + - + -Punnet GUPTA, 2004 - + + + +Natalia V,Em, 2005 + + - + +S Hoh, 2006 - + + + -Karlene C, 2009 - + - + -Shahid SIDDIQ, 2010 + + - + +LAYFOUNI, 2010 + + + + -Jiehan ZHOU, 2010 - + - + -Fatma ACHOUR, 2012 - + + + -Marco CONTI, 2012 - + + + +Our model + + + + +

4 ARCHITECTURAL MODELPROPOSED

The global architecture of PIS is illustrated in Fig.7. It is based on five layers. The smart interactionLayer is called for to combine many individual ac-tivity interactions for smart devices and smart envi-ronments to support the core properties of UbiquitousComputing. In other words smart interaction is a uni-fied and continuous interaction model between Ubiq-uitous Computing applications and their UbiquitousComputing infrastructure, physical world and humanenvironments. The smart devices layer can be consid-ered as a single entry to access sets of popular mul-tiple application services on the device or remotelyon servers. Most of them usually have one specifieduser. The smart environment layer comprises a setof networked devices connected to the physical envi-ronment. Different from smart devices, the devicesincluding a smart environment normally complete asingle pre-defined task. Integrated environment com-ponents automatically react to or anticipate user in-teraction using iHCI (implicit human computer inter-action). The infrastructure layer contains all physi-cal hardware such as devices and network equipment.This layer collects the raw data which are provided bythe physical devices. The security layer must ensurebasic security services. Privacy refers to the protec-tion of the users’ identities and information from non-authorized parties. Confidentiality is required to pro-tect the users’ information in the whole system, accesscontrol; it is a restriction to access or performance ofan action on some resources, Authentication: used to

identify and authenticate an entity in order to verifyits source or to estimate the trust level that it can begranted.

Figure 7: Our system architecture for pervasive computing.

4.1 Infrastructure Layer

Represents the level of communication channels, dataand sensor networks, and broadband internet access,but also the intelligent design of databases, storage,and network applications (Fig. 8). The term infos-tructure reflect the change in complexity over the tra-ditional view of infrastructures, but also to emphasizethe increased interdependency with the other levels ofthe value chain. Pervasive infostructure is all aboutaccess to information and services beyond the tradi-tional client-server paradigm. The merger of com-puting and communications is leading the way to-wards pervasiveness, and infrastructures are the levelwhere we place the property of this merger. Systeminfrastructure provides the systems needed for usingthe services (Cheung-Foo-Wo, 2009). Infrastructureproviders have advantage about ubiquitous computingas products and services are provided through multi-ple channels (Moitra, 2004).

Figure 8: Layer Infrastructure for the Intelligent Environ-ment.

A�Framework�for�Ambient�Computing

175

4.2 Smart Devices Layer

Smart devices, personal computer, mobile phone,tend to be multipurpose ICT devices (Sartor), oper-ating as a single portal to access sets of popular mul-tiple application services that may reside locally onthe device or remotely on servers. There is a rangeof forms for smart devices. Smart devices tend to bepersonal devices, having a specified owner or user.In the smart device model, the locus of control anduser interface resides in the smart device. Devices areoften designed to be multi-functional because theseease access to, and simplify the interoperability of,multi functions at run time. The main characteristicsof smart devices are as follows (Fig. 9).

Figure 9: The main characteristics of smart devices.

4.3 Smart Interface Layer

Smart Interaction is called for to combine many in-dividual activity interactions for smart devices (Mi-raoui, 2009) and smart environments to support thecore properties of Ubiquitous Computing. In otherwords, smart interaction is a unified and continuousinteraction model between Ubiquitous Computing ap-plications and their Ubiquitous Computing infrastruc-ture, physical world and human environments.

4.4 Smart Space Layer

A smart environment comprises a set of net-worked devices connected to the physical environ-ment (Satyanarayanan, 2001). It allows inhabitedservices and tasks with different objects available(Fontaine, 2006). The collection of information fromthe physical environment system using communicat-ing objects via using iHCI (implicit human computerinteraction) or objects capable of capturing informa-tion (sensors). The collected information is then in-terpreted, filtered and aggregated by various applica-tions to enrich the contextual information in order to

provide appropriate services. The intelligence of thesystems depend on their way to exploit the contextdata acquired. The data received from the environ-ment and the user must be interpreted to perform theappropriate action (Fig. 10).

Figure 10: Smart Space.

4.5 Security layer

In pervasive computing, services are more open, ac-cessible, distributed, and close to the user. This prox-imity introduces new threats and vulnerabilities forthe systems. Now, systems and services are assimi-lated to supermarkets open for everybody (Bourcier),in which merchandise is directly accessible to clientsin a self-service way. Of course, such displayedgoods are vulnerable and can be easily stolen, butnew sophisticated guard systems appeared to protectthe merchandise. new techniques were developed toadapt the security requirements to the environment.

5 CONCLUSION ANDPERSPECTIVES

In this paper, we have validated our approach in thedomain of pervasive computing from a complete do-main analysis. Based on the concepts of UbiquitousComputing, we define a list of challenges and con-cepts that form the core elements of a pervasive en-vironment. Pervasive information systems (PIS) con-stitute an emerging class of information systems (IS)where information technology (IT) is gradually em-bedded in the physical environment (Kouruthanas-sis and Giaglis, 2006). Therefore, PIS introduce theproperty of context awareness as a result of the perva-sive artifacts capability to collect, process, and man-age environmental or user-related information on areal-time basis. other research issues that we wouldlike to work on are: context awareness in pervasivecomputing.

CLOSER�2014�-�4th�International�Conference�on�Cloud�Computing�and�Services�Science

176

REFERENCES

Mark Weiser, The computer for the 21st century, Scien-tic American, tome 265, nr 3, pages 66-75, septembre1991.

Ansem Ben Cheikh, E-CARe: une méthode d’ingénieriedes Systèmes d’Information ubiquitaires, 2012.

Clement Escoffier, iPOJO: Un modèle à composant à ser-vice flexible pour les systèmes dynamiques, 2008.

Debashis Saha and Amitava Mukherjee, "Pervasive com-puting: a paradigm for the 21st century", IEEE Com-puter, 25-31, March 2003.

M. Satyanarayanan, Pervasive Computing: Vision andChallenges IEEE Personal Communications, 2001.

Puneet Gupta Deependra Moitra, Evolving a pervasiveIT infrastructure: a technology integration approach,2004, p33.

Cristiano da Costa André, A context aware service basedsoftware infrastructure for pervasive computing.

Adnan Shaout, Hari Srinivasan , Framing Pervasive Com-puting, 2009.

Imad Afyouni, Indoor space modelling, mobility, and posi-tioning in ubiquitous information systems, 2010.

Jinshan Liu, Découverte de services sensible à la qualitéde service dans les environnements de l’informatiquediffuse, 2006.

Salma Najar, Manuele Pinheiro, Luiz Angelo and CarineSouveyet, Analyse des Mécanismes de Découverte deServices avec Prise, 2012.

Fatma Achour, Anis Jedidi and Faiez Gargouri, TheGeneric Model for Pervasive Information System,2012.

Daniel Cheung-Foo-Wo, Adptation dynamique par tissaged’aspects d’assemblage, 2009.

Moeiz Miraoui, Architecture logicielle pour l’informatiquediffuse, 2009.

Natalia V. Em, Keedong Yoo, Extracting requirements forubiquitous computing technology-based is deployingfactor analysis, 2005.

Kouruthanassis, P. E. and Giaglis, G. M., A design theoryfor pervasive information systems, 3rd Int. Workshopon Ubiquitous Computing, pp. 62-70, 2006.

Wilfried Jouve, Approche déclarative pour la génération decanevas logiciels dédiés à l’informatique ubiquitairep7, 2009.

Didier Hoareau, Composants ubiquitaires pour reseaux dy-namiques, p8-9, 2007

L. Duboc, et al., "A framework for modelling and analysisof software systems scalability," published in the pro-ceedings of the International conference on SoftwareEngineering, 2006, pp 949-952.

D. F. Bantz, C. Bisdikian, D. Challener, J. P. Karidis, S.Mastrianni, A. Mohindra, D. G. Shea and M. Vanover,"Autonomic personal computing". IBM Systems Jour-nal, 42(1):165-176, 2003.

A. K. Dey, "Understanding and Using Context". Personaland Ubiquitous Computing, 5(1):4-7, 2001.

Boxiao Li, Design and Devleopment of an "Intelligent En-vironment", p10-11, 2010.

Jared Dnovan, Human computer interaction, 2010.Wikipedia. Autonomy. http://en.wikipedia.org/wiki/

Autonomy, 2010.P. Senn, "Objets communicants et nanotechnologie",

Journées Scientifiques du CNRFS Nanosciences etRadioélectricité, 2007.

F. Mattern and P. Sturm, "From Distributed Systemsto Ubiquitous Computing - The State of the Art,Trends, and Prospects of Future Networked Systems",In Klaus Irmscher and Klaus-Peter Fähnrich, editors,Proceedings KIVS, pp. 3-25, Springer- Verlag, Febru-ary 2003.

M. Beigl, A. Krohn, T. Zimmer and Christian Decker, Sit-uation Aware Context Communication", Internationalconference on Ubiquitous Computing, October 12-15,Seattle, USA, 2003.

A. S. Tanenbaum and M. Van Steen. Distributed Systems:Principles and Paradigms. Prentice Hall, 1999.

Idrissa Sarr, Routage des Transactions dans les Bases deDonnées à Large Echelle, 2010.

Rachid Saadi, The Chamelon: Un système de sécurité pourutilisateurs nomades en environnement pervasif et col-laboratif, 2009, p2-15.

Jean-Yves Tigli, Stéphane Lavirotte, Gaëtan Rey, VincentHourdin, Daniel Cheung- Foo-Wo, Eric Callegari, andMichel Riveill. WComp Middleware for Ubiqui-tous Computing: Aspects and Composite Event-basedWeb Services. Annals of Telecommunications (AoT),64(3-4):197-214, Apr 2009.

F. Zhu, M. W. Mutka and L. M. Ni. Service Discovery inPervasive Computing Environments. IEEE PervasiveComputing, 4(4):81-90. IEEE Educational ActivitiesDepartment, 2005.

Frédérique Laforest, De l’adaptation à la prise en compte ducontexte Une contribution aux systèmes d’informationpervasifs, p10, 2007

Areski Flissi, Christophe Gransart and Philippe Merle,Une infrastructure à composants pour des applicationsubiquitaires. UbiMob 2005: 45-48.

Nicolas Plouznikoff, Jean-Marc Robert, ApplicationsUbiquitaires, Caractéristiques, enjeux et défis del’informatique portée, 2005.

Antonio Vallecillo, Juan Hernandez and José M. Troya,Component Interoperability.

Nicolas Ferry, Stéphane Lavirotte, Gaëtan Rey and Jean-Yves Tigli, Adaptation Dynamique d’Applications auContexte en. Informatique Ambiante, 2008.

Germán Sancho, Adaptation d’architectures logiciellescollaboratives dans les environnements ubiquitaires.Contribution à l’interopérabilité par la sémantique,2010, p20.

Christine Louberry, Adaptation au Contexte pour la Gestionde la Qualité de Service,2010.

UIT. X.902, Technologies de l’information, traite-ment reparti ouvert, modèle de référence fonde-ments. Recommandation x.902 (11/95), Union Inter-nationale des Télécommunications.

Patrick Reignier, Intelligence Ambiante Pro-Active de laSpécification à l’Implémentation. 2010.

A�Framework�for�Ambient�Computing

177

Nabil Kouici, Gestion des déconnexions pour applicationsréparties à base de composants en environnementsmobiles.

Johann Bourcier, Auto-Home: une plate-forme pour la ges-tion autonomique d’applications pervasives, 94.

Pascal Dugénie, Espaces collaboratifs ubiquitaires sur uneinfrastructure à ressources distribuées, p52.

Mathieu Petit, Approche spatiale pour la caractérisation ducontexte d’exécution d’un système.

Fabien Sartor, Modélisation de l’interopérabilité d’objetscommunicants et de leur coopération Application à ladomotique.

Shahid Siddiq and Aasim Ali, Gap Analysis between Ubiq-uitous Computing Requirements and Features of anOpen Source Operating System (Openmoko) for HandHeld Devices, International Journal of Electrical &Computer Sciences IJECS-IJENS Vol: 10 No: 05,2010.

Mehdi Loukil, Gestion de contexte pour l’optimisation del’acces et l’adaptation des services sur des environ-nements heterogenes, 137, 2012.

Mathieu Gallissot, Modéliser le concept de confort dansl’habitat intelligent: du multisensoriel au comporte-ment, 2012, p16.

Emeric Fontaine, Programmation d’espace intelligent parl’utilisateur final, p12, 2006

Schilit, N. B., Adams, N. and Want, R., (1994). Contextaware computing applications. In Workshop on Mo-bile Computing Systems and Applications. Sta Cruz,EtatsUnis, décembre 1994. p 8590.

Brown, J. P., Bovey, D. J. and Chen, X., (1997). Con-textaware Applications: from the Laboratory to themarketplace. In Institute of Electrical and Electron-ics Engineers (IEEE) Personnal Communications. v.4, p. 5864.

P. J. Brown. The Stick-e Document: a framework for creat-ing Context aware applications. In Electronic Publish-ing 96, (1996), Document Manipulation and Typog-raphy, 24-26 September 1996, Pala-Alto, California,USA, pp.259-272

Pascoe. Adding generic contextual capabilities to wearablecomputers. In Proceedings of 2nd International Sym-posium on Wearable Computers, October 1998, pp.92-99

Kristof Van Laerhoven and Kofi Aidoo, Personal and Ubiq-uitous Computing vol. 5 issue 1 February, 2001. p. 46- 49.

Abowd, G. D., Dey, A. K., Brown, P. J., Davies, N., Smith,M., and Steggles, P., 1999. Towards a Better Under-standing of Context and Context- Awareness ( LectureNotes In Computer Science, Vol. 1707, pp. 304- 307),1999

Karlene C. Cousins and Upkar Varshney Designing Ubiqui-tous Computing Environments to Support Work LifeBalance, 2009,117-122.

S. Hoh, J. S. Tan, M. Hartley, Context-aware systems aprimer for user-centred services, BT Technology Jour-nal 186 o Vol 24 No 2 o April 2006

Marco Conti, Sajal K. Das and Chatschik Bisdikian, Look-ing ahead in pervasive computing: Challenges and op-

portunities in the era of cyber-physical convergence,Pervasive and Mobile Computing 8 (2012) 2-21.

Jiehan Zhou, Ekaterina Gilman, Mika Ylianttila and JukkaRiekki, Service Computing: Visions and Challenges,2010 10th IEEE International Conference on Com-puter and Information Technology, p1335-1339.

Anthony Jameson, "Modeling Both the Context and theUser", Personal Technologies, 5 (2001).

CLOSER�2014�-�4th�International�Conference�on�Cloud�Computing�and�Services�Science

178