La prolifération des objets communicants fixes et mobiles soulève la question de leur intégration dans les environnements quotidiens, par exemple dans le cadre de la e-santé ou de la domotique. Les principaux défis soulevés relèvent de l’interconnexion et de la gestion de la masse de donnée produite par ces objets intelligents. Notre premier objectif est d’adopter une démarche des couches basses vers les couches hautes pour faciliter l’intégration de ces objets à des services intelligents. Afin de développer celle-ci, il est nécessaire de d’étudier le processus de conception des objets intelligents indépendamment de considérations matérielles et logicielles, au travers de la considération de leur propriétés cyber-physiques. Pour mener à bien la réalisation de services intelligents à partir d’objets connectés, les deux axes de recherche suivant seront développés : la définition d’une méthode de conception orientée service pour les objets connectés intégrant une dimension formelle ainsi de valider le comportement de ceux-ci, l’auto-adaptation intelligente dans un contexte évolutif permettant aux objets de raisonner sur eux même au travers d’un langage déclaratif pour spécifier les stratégies d’adaptation. La validation de ces contributions s’effectuera par le biais du développement et de l’expérimentation à grandeur nature d’un service de diagnostic médical continu basé sur la collecte de données médicales en masse par des réseaux non-intrusifs de capteurs biomédicaux portables sur le corps humain. / The rapid growth of fixed and mobile smart objects raises the issue of their integration in everyday environment, e.g. in e-health or home-automation contexts. The main challenges of these objects are the interoperability, the handling of the massive amount of data that they generate, and their limited resources. Our goal is to take a bottom-up approach in order to improve the integration of smart devices to smart services. To ensure the efficient development of our approach, we start with the study of the design process of such devices regardless of specific hardware or software through the consideration of their cyber-physical properties. We thus develop two research directions: the specification of a service-oriented design method for smart devices with formal considerations in order to validate their behavior, and the proposal of a self-adaptation framework in order to handle changing operating context through self-reasoning and the definition of a declarative self-adaptation objectives specification language. The testing of these contributions will be realized through the development of a large-scale experimental framework based on a remote diagnostics case-study relying on non-invasive wearable biomedical sensors.

Comme dans de nombreux autres domaines, la construction des applications en Informatique Ambiantes (IAm) se fait par réutilisation d’entités logicielles disponibles. Pour des raisons de conductivités, de pannes, de charge de batterie mais aussi de nombreuses autres, la disponibilité de ces entités est imprévisible ce qui implique que l’auto-adaptation dynamique des applications est une nécessité. Cela passe par la spécification en parallèle des adaptations par des experts de divers domaines. Ce parallélisme de construction, peut amener des problèmes d’interférences lors de la composition dynamique de plusieurs adaptations. Dans cette thèse, par l’utilisation de graphes, nous contribuons à la définition d’un cadre formel pour la détection et la résolution de ces interférences. L’assemblage des entités logicielles repose sur des connecteurs d’assemblage qui sont utilisés dans la spécification des adaptations. Des règles de réécriture de graphe permettront de résoudre les interférences détectées, cette résolution étant guidée par la connaissance de connecteurs définis. De plus, pour pouvoir étendre dynamiquement et automatiquement notre mécanisme de gestion des interférences, nous proposons la modélisation comportementale de ces connecteurs. Ceci permet de ne pas reposer sur une connaissance à priori des connecteurs et autorise par la même d’étendre dynamiquement l’ensemble des connecteurs disponibles pour la spécification des adaptations. / Like many other fields, application construction in ubiquitous computing is done by reuse of available software entities. For reasons of conductivity, breakdown, battery charge, but also many others reasons, the availability of these entities is unpredictable. As consequence, the self-adaptation of applications becomes necessary. This requires the specification of parallel adaptations by experts from various fields. This parallel specification can cause interference problems when several adaptations are composed. In this thesis, using graph formalism, we contribute to the definition of a formal approach for the detection and the resolution of interferences. The specification of adaptation uses connectors in order to assemble software entities. Graph rewriting rules are defined to solve the detected interferences. This resolution is guided by the knowledge of defined connectors. In addition, in order to extend dynamically and automatically our interference management mechanism, we propose behavioral modeling of these connectors. This allows us extending our mechanism without an a priori knowledge of connectors and allows afterwards to extend the set of available connectors used for adaptations' specifications.

Our society is increasingly demanding situation-aware smarter software (SASS) systems, whose goals change over time and depend on context situations. A system with such properties must sense their dynamic environment and respond to changes quickly, accurately, and reliably, that is, to be context-aware and self-adaptive. The problem addressed in this dissertation is the dynamic management of context information, with the goal of improving the relevance of SASS systems' context-aware capabilities with respect to changes in their requirements and execution environment. Therefore, this dissertation focuses on the investigation of dynamic context management and self-adaptivity to: (i) improve context-awareness and exploit context information to enhance quality of user experience in SASS systems, and (ii) improve the dynamic capabilities of self-adaptivity in SASS systems. Context-awareness and self-adaptivity pose signi cant challenges for the engineering of SASS systems. Regarding context-awareness, the rst challenge addressed in this dissertation is the impossibility of fully specifying environmental entities and the corresponding monitoring requirements at design-time. The second challenge arises from the continuous evolution of monitoring requirements due to changes in the system caused by self-adaptation. As a result, context monitoring strategies must be modeled and managed in such a way that they support the addition and deletion of context types and monitoring conditions at runtime. For this, the user must be integrated into the dynamic context management process. Concerning self-adaptivity, the third challenge is to control the dynamicity of adaptation goals, adaptation mechanisms, and monitoring infrastructures, and the way they a ect each other in the adaptation process. This is to preserve the eff ectiveness of context monitoring requirements and thus self-adaptation. The fourth challenge, related also to self-adaptivity,concerns the assessment of adaptation mechanisms at runtime to prevent undesirable system states as a result of self-adaptation. Given these challenges, to improve context-awareness we made three contributions. First, we proposed the personal context sphere concept to empower users to control the life cycle of personal context information in user-centric SASS systems. Second, we proposed the SmarterContext ontology to model context information and its monitoring requirements supporting changes in these models at runtime. Third, we proposed an effi cient context processing engine to discover implicit contextual facts from context information speci fied in changing context models. To improve self-adaptivity we made three contributions. First, we proposed a framework for the identi cation of adaptation properties and goals, which is useful to evaluate self-adaptivity and to derive monitoring requirements mapped to adaptation goals. Second, we proposed a reference model for designing highly dynamic self-adaptive systems, for which the continuous pertinence between monitoring mechanisms and both changing system goals and context situations is a major concern. Third, we proposed a model with explicit validation and veri cation (V&V) tasks for self-adaptive software, where dynamic context monitoring plays a major role. The seventh contribution of this dissertation, the implementation of Smarter-Context infrastructure, addresses both context-awareness and self-adaptivity. To evaluate our contributions, qualitatively and quantitatively, we conducted several comprehensive literature reviews, a case study on user-centric situation-aware online shopping, and a case study on dynamic governance of service-oriented applications. / Graduate

When the human mind searches concept space for solutions to a given condition we have a choice between conventional and creative thinking. But what are the probabilities of improving a given situation using creative thinking compared with conventional thinking? To answer this question we are extending the meaning of creativity beyond human creativity. We view creativity as an optimised search strategy applicable to the larger set of all teleonomic systems and term this creativity teleonomic creativity. We argue that an analog process is common to all manifestations of creativity within teleonomic systems and describe this process and its cause. In order to show this process and to make quantitative comparisons, we utilise the metaphor of an adaptive fitness landscape and simple statistical techniques. The term fitness in our case describes the condition of a well-defined property being suitable for a purpose, rather than an overall evaluation of many complex interactions measuring reproductive success. We define creativity as the successful attempt of either individuals or populations to gain higher fitness via exploration of global fitness peaks as opposed to the exploitation of a currently occupied local peak. We then show mathematically how the inclusion of creativity in a search can dramatically increase the chances of finding appropriate solutions. We also recognise that creative behaviour is most successful when the environmentis unstable. We note the existence of a strategic meta-parameter that allows self-adaptation when tuned via a feedback loop from the environment. We show that creativity can be understood as a random process with an optimal setting for the standard deviation that maximises the probability of hitting a target of higher fitness. We support our claims with computer simulations and observe several occurrences of teleonomic creativity in nature. In addition we measure the entropy of a teleonomic system via the phase-space of internal variables and observe a sudden entropy increase during the onset of creative behaviour in a teleonomic system. Our investigations also enable us to rationalise the processes, conditions and phenomena surrounding human creativity such as mistakes, madness, serendipity, humor, analogy making and interpret the function of creativity promoters and inhibitors. Our findings may also allow us to incorporate creativity into artificial computer models. We speculate that creativity is an emerging property of any teleonomic system and as such ubiquitous in nature.

Made available in DSpace on 2014-12-17T15:48:08Z (GMT). No. of bitstreams: 1 JoseANM_DISSERT.pdf: 1910810 bytes, checksum: b18d20e7060e889bfdf95282aa19c128 (MD5) Previous issue date: 2013-05-17 / Self-adaptive software system is able to change its structure and/or behavior at runtime due to changes in their requirements, environment or components. One way to archieve self-adaptation is the use a sequence of actions (known as adaptation plans) which are typically defined at design time. This is the approach adopted by Cosmos - a Framework to support the configuration and management of resources in distributed environments. In order to deal with the variability inherent of self-adaptive systems, such as, the appearance of new components that allow the establishment of configurations that were not envisioned at development time, this dissertation aims to give Cosmos the capability of generating adaptation plans of runtime. In this way, it was necessary to perform a reengineering of the Cosmos Framework in order to allow its integration with a mechanism for the dynamic generation of adaptation plans. In this context, our work has been focused on conducting a reengineering of Cosmos. Among the changes made to in the Cosmos, we can highlight: changes in the metamodel used to represent components and applications, which has been redefined based on an architectural description language. These changes were propagated to the implementation of a new Cosmos prototype, which was then used for developing a case study application for purpose of proof of concept. Another effort undertaken was to make Cosmos more attractive by integrating it with another platform, in the case of this dissertation, the OSGi platform, which is well-known and accepted by the industry / Sistemas de software autoadaptativos s?o caracterizados por terem a capacidade de alterar sua estrutura e/ou comportamento em tempo de execu??o em resposta a mudan?as ocorridas em seus requisitos, seu ambiente de execu??o ou em seus componentes. Uma das maneiras de se alcan?ar a autoadapta??o ? a utiliza??o de uma sequ?ncia de a??es (conhecidas como planos de adapta??o) que normalmente s?o definidas em tempo de desenvolvimento. Esse tipo de adapta??o foi adotado pelo Cosmos - um framework proposto para dar suporte ? configura??o e ao gerenciamento de recursos em ambientes distribu?dos. De maneira a lidar com a variabilidade inerente a sistemas autoadaptativos, como por exemplo, o aparecimento de novos componentes que permitam o estabelecimento de configura??es que n?o foram consideradas em tempo de desenvolvimento, o presente trabalho procura dar ao Cosmos a possibilidade de utilizar planos de adapta??o gerados em tempo de execu??o. Para tal, foi necess?rio realizar a reengenharia do mesmo, de maneira a permitir sua integra??o com um mecanismo capaz de gerar planos de adapta??o dinamicamente. Nesse contexto, o presente trabalho se concentrou na reengenharia do Cosmos. Dentre as mudan?as realizadas no Cosmos, podemos destacar altera??es no metamodelo utilizado para representar componentes e aplica??es, que foi redefinido com base em uma linguagem de descri??o arquitetural. Essas altera??es foram propagadas para a implementa??o de um novo prot?tipo do Cosmos, que foi utilizado para o desenvolvimento de aplica??es definidas para fins de prova de conceito. Outro esfor?o empreendido consistiu em tornar o uso do Cosmos mais atrativo ao viabilizar sua integra??o com outras plataformas. Especificamente, no presente trabalho, com a plataforma OSGi, uma plataforma bem aceita pela ind?stria

Self-adaptive systems are capable of autonomously adjusting their behavior at runtime to accomplish particular adaptation goals. The most common way to realize self-adaption is using a feedback loop(s) which contains four actions: collect runtime data from the system and its environment, analyze the collected data, decide if an adaptation plan is required, and act according to the adaptation plan for achieving the adaptation goals. Existing approaches achieve the adaptation goals by using formal methods, and exhaustively verify all the available adaptation options, i.e., adaptation space. However, verifying the entire adaptation space is often not feasible since it requires time and resources. In this thesis, we present an approach which uses machine learning to reduce the adaptation space in self-adaptive systems. The approach integrates with the feedback loop and selects a subset of the adaptation options that are valid in the current situation. The approach is applied on the simulator of a self-adaptive Internet of Things application which is deployed in KU Leuven, Belgium. We compare our results with a formal model based self-adaptation approach called ActivFORMS. The results show that on average the adaptation space is reduced by 81.2% and the adaptation time by 85% compared to ActivFORMS while achieving the same quality guarantees.

Les réseaux de capteurs suscitent un engouement croissant du fait du grand nombre d'applications mais également des défis inhérents à ce genre de réseaux. Le standard IEEE 802.15.4 a été proposé afin de standardiser les couches physique et MAC. Dans ce travail nous avons dans un premier temps proposé une variante multi-canal pour le standard 802.15.4 permettant de résoudre le problème de collisions de supertrames. Pour cela nous proposons de construire un réseau en arbre avec la particularité que chaque cellule du réseau utilise un canal différent permettant ainsi une réduction conséquente des interférences et augmente la capacité du réseau. Nous avons également introduit un nouveau mécanisme de construction de topologie, d'allocation de canal et d'ordonnancement de supertrames nécessaire au bon fonctionnement d'une telle solution. Dans un deuxième temps nous avons analysé l'impact des différents paramètres de la méthode d'accès du standard. Nous avons mis en exergue les faiblesses de la méthode d'accès dont les performances baissent drastiquement pour des réseaux trop importants. Partant de ce constat, nous avons proposé des mécanismes d'auto-adaptation pour la méthode d'accès du standard. Ces derniers permettent d'adapter dynamiquement la taille des fenêtres de contention en fonction des conditions de trafic observées sur le canal. Le calcul des valeurs optimales est exécuté par chaque coordinateur pour résoudre le problème de surdité. Ces mécanismes sont distribués et convergent rapidement même en cas de trafic en rafales. / An increasing interest has been observed in Wirless Sensor Network that can be explained by wide range of WSN applications as well as by the challenges involving the constraints of this type of networks. The IEEE 802.15.4 standard has been proposed with the objective of standardizing the physical and MAC layers. In this work, we have firstly proposed an alternative multichannel scheme for the IEEE 802.15.4 standard. It is able to solve the problem of superframe collisions based on a cluster-tree topology approach in which each cluster uses a different channel allowing the reduction of the interference and increased network capacity. We have also difined a novel mechanism for topology construction, channel allocation, and superframe scheduling. Secondly, we have analyzed the impact of different parameters on the medium access control. We have shown the weakness of the medium access method proposed in the standard. For instance, we have observed a decrease in performance when the standard is applied in large networks. Motivated by the analysis and its results we have then proposed auto-adaptive mechanisms for the medium access control. They allow to dynamically adapt the size of the contention window according to the observed traffic conditions. Each coordinator computes the optimal values to avoid deafness. The proposed mechanisms rapidly converge even in the case of bursty traffic.

L'Internet des Objets (IdO) couvre peu à peu tous les aspects de notre vie. À mesure que ces systèmes deviennent plus répandus, le besoin de gérer cette infrastructure complexe comporte plusieurs défis. En effet, beaucoup de petits appareils interconnectés fournissent maintenant plus d'un service dans plusieurs aspects de notre vie quotidienne, qui doivent être adaptés à de nouveaux contextes sans l'interruption de tels services. Cependant, ce nouveau système informatique diffère des systèmes classiques principalement sur le type, la taille physique et l'accès des nœuds. Ainsi, des méthodes typiques pour gérer la couche logicielle sur de grands systèmes distribués comme on fait traditionnellement ne peuvent pas être employées dans ce contexte. En effet, cela est dû aux capacités très différentes dans la puissance de calcul et la connectivité réseau, qui sont très contraintes pour les appareils de l'IdO. De plus, la complexité qui était auparavant gérée par des experts de plusieurs domaines, tels que les systèmes embarqués et les réseaux de capteurs sans fil (WSN), est maintenant accrue par la plus grande quantité et hétérogénéité des logiciels et du matériel des nœuds. Par conséquent, nous avons besoin de méthodes efficaces pour gérer la couche logicielle de ces systèmes, en tenant compte les ressources très limitées. Cette infrastructure matérielle sous-jacente pose de nouveaux défis dans la manière dont nous administrons la couche logicielle de ces systèmes. Ces défis peuvent entre divisés en : Intra-nœud, sur lequel nous faisons face à la mémoire limitée et à la puissance de calcul des nœuds IdO, afin de gérer les mises à jour sur ces appareils ; Inter-noeud, sur lequel une nouvelle façon de distribuer les mises à jour est nécessaire, en raison de la topologie réseau différente et le coût en énergie pour les appareils alimentés par batterie ; En effet, la puissance de calcul limitée et la durée de vie de chaque nœud combiné à la nature très distribuée de ces systèmes, ajoute de la complexité à la gestion de la couche logicielle distribuée. La reconfiguration logicielle des nœuds dans l'Internet des objets est une préoccupation majeure dans plusieurs domaines d'application. En particulier, la distribution du code pour fournir des nouvelles fonctionnalités ou mettre à jour le logiciel déjà installé afin de l'adapter aux nouvelles exigences, a un impact énorme sur la consommation d'énergie. La plupart des algorithmes actuels de diffusion du code sur l'air (OTA) sont destinés à diffuser un microprogramme complet à travers de petits fragments, et sont souvent mis en œuvre dans la couche réseau, ignorant ainsi toutes les informations de guidage de la couche applicative. Première contribution : Un moteur de modèles en temps d'exécution représentant une application de l'IdO en cours d'exécution sur les nœuds à ressources limitées. La transformation du méta-modèle Kevoree en code C pour répondre aux contraintes de mémoire spécifiques d'un dispositif IdO a été réalisée, ainsi que la proposition des outils de modélisation pour manipuler un modèle en temps d'exécution. Deuxième contribution : découplage en composants d'un système IdO ainsi qu'un algorithme de distribution de composants efficace. Le découplage en composants d'une application dans le contexte de l'IdO facilite sa représentation sur le modèle en temps d'exécution, alors qu'il fournit un moyen de changer facilement son comportement en ajoutant/supprimant des composants et de modifier leurs paramètres. En outre, un mécanisme pour distribuer ces composants en utilisant un nouvel algorithme appelé Calpulli est proposé. / The Internet of Things (IoT) is covering little by little every aspect on our lives. As these systems become more pervasive, the need of managing this complex infrastructure comes with several challenges. Indeed, plenty of small interconnected devices are now providing more than a service in several aspects of our everyday life, which need to be adapted to new contexts without the interruption of such services. However, this new computing system differs from classical Internet systems mainly on the type, physical size and access of the nodes. Thus, typical methods to manage the distributed software layer on large distributed systems as usual cannot be employed on this context. Indeed, this is due to the very different capacities on computing power and network connectivity, which are very constrained for IoT devices. Moreover, the complexity which was before managed by experts on several fields, such as embedded systems and Wireless Sensor Networks (WSN), is now increased by the larger quantity and heterogeneity of the node’s software and hardware. Therefore, we need efficient methods to manage the software layer of these systems, taking into account the very limited resources. This underlying hardware infrastructure raises new challenges in the way we administrate the software layer of these systems. These challenges can be divided into: intra-node, on which we face the limited memory and CPU of IoT nodes, in order to manage the software layer and ; inter-node, on which a new way to distribute the updates is needed, due to the different network topology and cost in energy for battery powered devices. Indeed, the limited computing power and battery life of each node combined with the very distributed nature of these systems, greatly adds complexity to the distributed software layer management. Software reconfiguration of nodes in the Internet of Things is a major concern for various application fields. In particular, distributing the code of updated or new software features to their final node destination in order to adapt it to new requirements, has a huge impact on energy consumption. Most current algorithms for disseminating code over the air (OTA) are meant to disseminate a complete firmware through small chunks and are often implemented at the network layer, thus ignoring all guiding information from the application layer. First contribution: A models@runtime engine able to represent an IoT running application on resource constrained nodes. The transformation of the Kevoree meta-model into C code to meet the specific memory constraints of an IoT device was performed, as well as the proposition of modelling tools to manipulate a model@runtime. Second contribution: Component decoupling of an IoT system as well as an efficient component distribution algorithm. Components decoupling of an application in the context of the IoT facilitates its representation on the model@runtime, while it provides a way to easily change its behaviour by adding/removing components and changing their parameters. In addition, a mechanism to distribute such components using a new algorithm, called Calpulli is proposed.

Pour rester performants et compétitifs, les systèmes de production doivent être capables de s’adapter pour faire face aux changements tels que l’évolution de la demande des clients. Il leur est essentiel de pouvoir déterminer quand et comment s’adapter (capacités, etc.). Malheureusement, de tels problèmes sont connus pour être difficiles. Les systèmes de production étant complexes, dynamiques et spécifiques, leurs gestionnaires n’ont pas toujours l’expertise nécessaire ni les prévisions suffisantes concernant l’évolution de leur système. Cette thèse vise à étudier la contribution que peut apporter l’apprentissage automatique à l’auto-adaptation des systèmes de production. Dans un premier temps, nous étudions la façon dont la littérature aborde ce domaine et en proposons un cadre conceptuel dans le but de faciliter l’analyse et la formalisation des problèmes associés. Ensuite, nous étudions une stratégie d’apprentissage à partir de modèles qui ne nécessite pas d’ensemble d’apprentissage. Nous nous intéressons plus précisément à une nouvelle approche basée sur la programmation génétique linéaire visant à extraire des connaissances itérativement à partir d’un modèle de simulation pour déterminer quand et quoi faire évoluer. Notre approche est implémentée à l’aide d’Arena et μGP. Nous l’appliquons à différents exemples qui concernent l’ajout/retrait de cartes dans un système à flux tiré, le déménagement de machines ou encore le changement de politique de réapprovisionnement. Les connaissances qui en sont extraites s’avèrent pertinentes et permettent de déterminer en continu comment chaque système peut s’adapter à des évolutions. De ce fait, elles peuvent contribuer à doter un système d’une forme d’intelligence. Exprimées sous forme d’un arbre de décision, elles sont par ailleurs facilement communicables à un gestionnaire de production. Les résultats obtenus montrent ainsi l’intérêt de notre approche tout en ouvrant de nombreuses voies de recherche. / Manufacturing systems must be able to continuously adapt their characteristics to cope with the different changes that occur along their life, in order to remain efficient and competitive. These changes can take the form of the evolution of customers demand for instance. It is essential for these systems to determine when and how to adapt (e.g., through changes in capacities). Unfortunately, such issues are known to be difficult. As manufacturing systems are complex, dynamic and specific in nature, their managers do not always have all the necessary expertise nor accurate enough forecasts on the evolution of their system. This thesis aims at studying the possible contributions of machine learning to the self-adaptation of manufacturing systems. We first study how the literature deals with self-adaptation and we propose a conceptual framework to facilitate the analysis and the formalization of the associated problems. Then, we study a learning strategy relying on models, which presents the advantage of not requiring any training set. We focus more precisely on a new approach based on linear genetic programming that iteratively extracts knowledge from a simulation model. Our approach is implemented using Arena and μGP. We show its benefits by applying it to increase/decrease the number of cards in a pull control system, to move machines or to change the inventory replenishment policy. The extracted knowledge is found to be relevant for continuously determining how each system can adapt to evolutions. It can therefore contribute to provide these systems with some intelligent capabilities. Moreover, this knowledge is expressed in the simple and understandable form of a decision tree, so that it can also be easily communicated to production managers in view of their everyday use. Our results thus show the interest of our approach while opening many research directions.

La combinaison des avancées médicales et technologiques a rendu possible le monitoring à distance des patients, introduisant le paradigme de la e-santé (e-health en anglais). Il comprend des services ou des systèmes de convergence de la médecine, la santé et l’information, avec l’e-surveillance et la télé-médecine. Le type de dispositif médical "healthcare" doit assurer une sûreté de fonctionnement optimale, que ce soit pour une analyse médicale ou l’application d’un traitement. Le niveau de sûreté et de performance est critique pour la santé du patient.Maintenant que ces dispositifs s’appliquent à un nombre croissant de patients, tous différents et vivant dans des conditions différentes, ils doivent être capables de s’adapter en termes de performances et s’adapter à l’activité, à l’environnement et au patient lui-même. Une stratégie globale et réutilisable d’auto-adaptation à été définie dans le but de mettre en place méthodiquement et de façon optimale le processus et les étapes de l’auto-adaptation. Une des pathologies importantes actuellement est le glaucome, une neuropathie oculaire détériorant le nerf optique induisant une perte du champ visuel progressive et irréversible. Le diagnostique et le suivi de cette pathologie se fait par une mesure de la Pression Intra-Oculaire(PIO). La méthode utilisée actuellement est contraignante et requière la manipulation experte d’un dispositif de mesure ponctuelle. De nouveaux capteurs biométriques ont vu le jour permettant une mesure continue et portable amenant un meilleur diagnostic ainsi qu’un suivi plus simple. Notre cas d’étude sera donc centré sur l’utilisation d’un capteur biométrique de mesure de la PIO qui servira de plate-forme de validation aux solutions proposées sachant qu’il devra être auto-adapté aux différentes anatomies, physiologies, biomécaniques et environnements du patient. Tous les indicateurs perturbateurs ou informateurs de la mesure sont regroupés en deux catégories, ceux liés à l’environnement et ceux liés aux conditions d’utilisation. Un état de l’art est proposé pour la première catégorie et un modèle analytique est proposé pour la deuxième. L’état de l’art montre que les expérimentations in vivo sont nécessaires pour compléter les connaissances sur les indicateurs majeurs dans le cas d’une mesure de la PIO faite par un capteur biométrique. Le modèle analytique inclut le couplage électromagnétique entre les deux antennes du capteur, ainsi que des phénomènes perturbateurs. Il suggère de nouveaux indicateurs indirects (Pente de la baseline et le coefficient du second ordre de la courbe de phase) pour l’adaptation aux phénomènes perturbateurs. / Thanks to the combination of medical and technological advances, it becomes feasible and relevant to operate remotely monitoring of patients, introducing e-health paradigm. It includes services or systems that are at the convergence of medicine, health and information, including e-monitoring and telemedicine. Among existing medical devices, the "healthcare" type must ensure optimum operational reliability whether it is for a medical analysis or a treatment’s application. The level of safety and performance are critical for the patient’s health. Since these devices apply to a growing number of patients, all different from each other and living under different conditions, the device must therefore be able to adapt it self to the patient’s needs, activity, his environment but also to the patient himself. A general reusable strategy of self-adaptation, aiming to define methodically and optimally the steps and processes of self-adaptation, have been defines. Diseases recognized as being of the current centuries diseases require regular medical follow. Among them is glaucoma, an ocular neuropathy that deteriorates the optic nerve leading to a progressive and irreversible loss of the vision field. To diagnose and monitor it, a measurement of the intraocular pressure (IOP) is mandatory. Current techniques are constraining and require the manipulation of punctual measurement tools by a doctor. New biometric devices have emerged allowing continuous and portable measurements to allow a better diagnostic and a simplified monitoring. Our case study will therefore rely on the use of a biometric IOP measurement device which will be used as a validation platform for the proposed solutions knowing that it must be self-adapted to the different anatomies, physiologioes, biomechanics and environments of the patient. All the disturbing or informative mesurement indicators are regrouped in two types, the environmental and the variability in operation. A state of the art is proposed for the first type and an analytical model for the second one. The state of the art shows that In-vivo experimentations are mandatory to complete the knwoledge of major indicator influencing the IOP measurement with a biometric sensor. The analytical model include the magnetic coupling between antennas and the impact of some disturbances. It suggests a new ideal indirect indicator (the slope of the baseline and second coefficient order of the Bode phase plot) for the disturbances adaptation. This model is valid for sensors with different dimensions of antennas.