La caractérisation du couvert nival en forêt boréale est un élément important pour la compréhension des régimes climatiques et hydrologiques. Depuis plusieurs années, l’utilisation des micro-ondes passives est étudiée pour l’estimation de l’équivalent en eau de la neige (SWE : Snow Water Equivalent) à partir de capteurs satellitaires. Les algorithmes empiriques traditionnels étant limités en forêt boréale, le couplage d’un modèle de transfert radiatif (MTR) micro-onde passive (qui prend en compte les contributions du sol, de la neige, de la végétation et de l’atmosphère) avec un modèle de neige pour l’inversion du SWE semble une avenue prometteuse. La thèse vise donc à coupler un MTR avec le schéma de surface du modèle climatique canadien (CLASS) dans une perspective d’application opérationnelle pour les estimations de SWE à partir de données satellitaires micro-onde à 10.7, 19 et 37 GHz. Dans ce contexte, certains aspects centraux du MTR, dont l’effet de la taille des grains ainsi que la contribution de la végétation sont développés et quantifiés. Le premier aspect étudié dans la thèse concerne l’adaptation du modèle d’émission micro-onde passive DMRT-ML (Dense media radiative transfer theory – multi layer) pour l’intégration d’une nouvelle métrique représentant la taille des grains (surface spécifique des grains de neige: SSA). L’étude basée sur des mesures radiométriques et de neige in situ, montre la pertinence de l’utilisation de la SSA dans DMRT-ML et permet d’analyser le sens physique de l’adaptation nécessaire pour amener le modèle à simuler les températures de brillance (T[indice inférieur B) de la neige avec une erreur quadratique moyenne minimale de l’ordre de 13 K. Dans un contexte du couplage entre le modèle de neige de CLASS et DMRT-ML, un modèle d’évolution de la SSA est ensuite implémenté dans CLASS. Les SSA simulées par le module développé sont validées avec des données in situ basées sur la réflectance de la neige dans l’infrarouge à courte longueur d’onde pour différents types d’environnement. Au niveau de la contribution de la végétation, le modèle γ-ω a été étudié à partir de différentes bases de données (satellite, avion et au sol) en forêt boréale dense. L’étude montre l’importance de la considération de la diffusion (ω) pour l’estimation de l’émission de la végétation, paramètre auparavant généralement négligé aux hautes fréquences. Ensuite, des relations entre les transmissivités et certains paramètres structuraux de la forêt, dont l’indice de surface foliaire (LAI), ont été établies pour des forêts boréales en été. Des valeurs d’albédo de diffusion (ω) ainsi que les paramètres définissant la réflectivité du sol (QH) en forêt boréale ont aussi été inversées. Finalement, les simulations de T [indice inférieur] B issues du couplage du MTR (DMRT-ML, modèle γ-ω, et modèle atmosphérique) avec CLASS (dont les SSA simulées) ont été comparées avec les données AMSR-E sur une série temporelle continue de sept ans. Les premières comparaisons montrent une différence entre les paramètres de végétation (γ-ω) d’été et d’hiver, ainsi qu’une importante contribution des croûtes de glace dans la neige au signal. Les simulations du modèle ajusté montrent une bonne correspondance avec les observations d’AMSR-E (de l’ordre de 3 à 7 K selon la fréquence et la polarisation). Des tests de sensibilité montrent par contre une faible sensibilité du MTR/CLASS au SWE pour des forêts denses et des couverts nivaux épais. Le MTR-CLASS développé pourrait permettre l’assimilation de températures de brillance satellitaires en forêt boréale dans des systèmes opérationnels pour l’amélioration de paramètres de surface, dont la neige, dans les modèles météorologiques et climatiques.

<p>Patients with systemic lupus erythematosus (SLE) have elevated levels of interferon (IFN)-α in blood and IFN-α-producing cells in tissues. In the present thesis, we investigate the mechanisms behind the upregulated IFN-α-production in SLE and also show that the IFN-α system is activated in primary Sjögren’s syndrome (pSS), with IFN-α-producing cells in the major affected organ, the salivary glands. The IFN-α is a type I IFN, a family of cytokines counteracting especially viral infections, by acting directly on infected cells, and via many immunomodulatory effects. The latter may also contribute to autoimmune processes.</p><p>The type I IFNs are usually produced upon recognition of microbial structures. In SLE, however, DNA-containing immune complexes (ICs) that induce IFN-α production are found. Many autoantibodies in SLE and pSS are directed to nucleic acids or to DNA/RNA-binding proteins. We show that also RNA in complex with autoantibodies from SLE or pSS patients (RNA-IC) induces IFN-α-production. The RNA could be either in the form of RNA-containing material released from apoptotic or necrotic cells or as a pure RNA-containing autoantigen, the U1 small nuclear ribonucleoprotein particle. </p><p>The IFN-α-production induced by RNA-IC occurred in plasmacytoid dendritic cells (PDCs), also termed natural IFN-producing cells (NIPCs), via binding to Fcγ-receptor IIa, endocytosis and triggering of Toll-like receptors (TLRs), probably TLR7 and TLR9. The RNA-IC may also have other effects, and we found that they induce prostaglandin E2 (PGE2) production in monocytes and tumor necrosis factor (TNF)-α in both monocytes and NIPC/PDC. The PGE2 downregulated the IFN-α induction in NIPC/PDC, and the IFN-α induction was increased in monocyte-depleted cell cultures. </p><p>The findings presented in this thesis aids in the understanding of the mechanisms behind the activated IFN-α system in SLE and other autoimmune diseases, and shows that also pSS is one of these diseases.</p>

Patients with systemic lupus erythematosus (SLE) have elevated levels of interferon (IFN)-α in blood and IFN-α-producing cells in tissues. In the present thesis, we investigate the mechanisms behind the upregulated IFN-α-production in SLE and also show that the IFN-α system is activated in primary Sjögren’s syndrome (pSS), with IFN-α-producing cells in the major affected organ, the salivary glands. The IFN-α is a type I IFN, a family of cytokines counteracting especially viral infections, by acting directly on infected cells, and via many immunomodulatory effects. The latter may also contribute to autoimmune processes. The type I IFNs are usually produced upon recognition of microbial structures. In SLE, however, DNA-containing immune complexes (ICs) that induce IFN-α production are found. Many autoantibodies in SLE and pSS are directed to nucleic acids or to DNA/RNA-binding proteins. We show that also RNA in complex with autoantibodies from SLE or pSS patients (RNA-IC) induces IFN-α-production. The RNA could be either in the form of RNA-containing material released from apoptotic or necrotic cells or as a pure RNA-containing autoantigen, the U1 small nuclear ribonucleoprotein particle. The IFN-α-production induced by RNA-IC occurred in plasmacytoid dendritic cells (PDCs), also termed natural IFN-producing cells (NIPCs), via binding to Fcγ-receptor IIa, endocytosis and triggering of Toll-like receptors (TLRs), probably TLR7 and TLR9. The RNA-IC may also have other effects, and we found that they induce prostaglandin E2 (PGE2) production in monocytes and tumor necrosis factor (TNF)-α in both monocytes and NIPC/PDC. The PGE2 downregulated the IFN-α induction in NIPC/PDC, and the IFN-α induction was increased in monocyte-depleted cell cultures. The findings presented in this thesis aids in the understanding of the mechanisms behind the activated IFN-α system in SLE and other autoimmune diseases, and shows that also pSS is one of these diseases.

<p>One of the major reasons for political instability in Sub-Saharan Africa originates from the way elections are conducted. Most African countries have quite a handful of electoral malpractices which lead to political instability, civil wars and low economic growth.  Electronic voting might be a solution to the election problems and thus bring in a stable political atmosphere which attracts investors.</p><p>This thesis looks at the prospects and challenges of implementing e-voting in Ghana and Sub-Saharan Africa as a solution to the current manual paper-based system and proposes a framework and requirements which can be used as guidelines for its adoption and implementation.</p><p>The thesis has been conducted by studying literature on electronic voting and Diffusion of Innovations Theory. Furthermore, a case study has been conducted on the Ghanaian electoral process, by conducting interviews with the Danquah Institute of Ghana and through questionnaires to some Ghanaian citizens in both the rural and urban areas of Ghana.</p><p>The main results from the thesis show that the adoption and subsequent implementation of e-voting in SSA countries, can only be possible if the governments show strong committment and support by securing funds through donor organizations, and providing the necessary IT infrastructure and other resources needed to support the project. Voter education too is an absolute necessity. The adoption of e-voting is mainly based on how the innovation will be diffused and the perceived benefits that will be derived from the investment. Hence, particular attention should be paid to the various communication channels, especially the media, through which messages are passed across to the citizens.</p>

My thesis deals with register allocation. During this phase, the compiler has to assign variables of the source program, in an arbitrary big number, to actual registers of the processor, in a limited number k. Recent works, for instance the thesis of F. Bouchez and S. Hack, have shown that it is possible to split in two different decoupled step this phase: the spill - store the variables into memory to release registers - followed by the registers assignment. These works demonstrate the feasibility of this decoupling relying on a theoretic framework and some assumptions. In particular, it is sufficient to ensure after the spill step that the number of variables simultaneously live is below k.My thesis follows these works by showing how to apply this kind of approach when real-world constraints come in play: instructions encoding, ABI (application binary interface), register aliasing. Different approaches are proposed. They allow either to ignore these problems or to directly tackle them into the theoretic framework. The hypothesis of the models and the proposed solutions are evaluated and validated using a thorough experimental study with the compiler of STMicroelectronics. Finally, all these works have been done with the constraints of modern compilers in mind, the JIT (just-in-time) compilation, where the compilation time et the memory footprint of the compiler are key factors. We strive to offer solutions that cope with these criteria or improve the result until a given budget is reached. We, in particular, used the SSA (static single assignment) form to define algorithm like tree scan that generalizes linear scan based approaches proposed for JIT compilation.

This research investigates what rural villagers perceive as important and develops recommendations for improved electrification project implementation centring on user-perceived values (UPVs). UPVs capture more than the basic definition of value in the sense that they include benefits, concerns, feelings and underlying drivers that vary in importance and act as the main motivators in the lives of project beneficiaries as perceived at a given time. Low access to energy continues in rural sub-Saharan Africa despite significant investment by the development community. One fundamental reason is that energy infrastructure adoption remains low, as evidenced by the lack of project sustainability. To counter this, the challenge for energy project developers is to achieve sustainable long-term interventions through the creation of value for beneficiaries, rather than the traditional approach of focusing on short-term project outputs. The question of what is valuable to people in rural communities has historically not played into the design and diffusion of energy infrastructure development projects. This research drew on design and marketing approaches from the commercial sector to investigate the UPVs of rural Ugandans. To better understand the UPVs of rural villagers a new method, consisting of a UPV game and UPV framework, was developed. This method is suitable for capturing, understanding and mapping what rural populations perceive as important. Case study analyses were carried out in seven villages across rural Uganda. The case studies included the UPV game supplemented by non-energy-specific and energy-specific interviews with villagers. Additionally, interviews with experts were conducted to verify the UPV framework and to identify the gap between experts’ opinion and villagers’ perception of what is important. The research demonstrates the effectiveness of the UPV game in deducing the values of rural villagers. The findings demonstrate a disconnect in the ability to accurately capture and design projects which resonate with and respond to the UPVs of recipients of rural electrification projects. A comparison between the villagers’ statements and experts’ opinion regarding what is most valuable to rural communities reveals striking differences that point to a fundamental misunderstanding of rural community UPVs which are likely to be contributing to widespread electrification project failure.

One of the major reasons for political instability in Sub-Saharan Africa originates from the way elections are conducted. Most African countries have quite a handful of electoral malpractices which lead to political instability, civil wars and low economic growth.  Electronic voting might be a solution to the election problems and thus bring in a stable political atmosphere which attracts investors. This thesis looks at the prospects and challenges of implementing e-voting in Ghana and Sub-Saharan Africa as a solution to the current manual paper-based system and proposes a framework and requirements which can be used as guidelines for its adoption and implementation. The thesis has been conducted by studying literature on electronic voting and Diffusion of Innovations Theory. Furthermore, a case study has been conducted on the Ghanaian electoral process, by conducting interviews with the Danquah Institute of Ghana and through questionnaires to some Ghanaian citizens in both the rural and urban areas of Ghana. The main results from the thesis show that the adoption and subsequent implementation of e-voting in SSA countries, can only be possible if the governments show strong committment and support by securing funds through donor organizations, and providing the necessary IT infrastructure and other resources needed to support the project. Voter education too is an absolute necessity. The adoption of e-voting is mainly based on how the innovation will be diffused and the perceived benefits that will be derived from the investment. Hence, particular attention should be paid to the various communication channels, especially the media, through which messages are passed across to the citizens.

No description available.

Conducting seamless Live Virtual Constructive (LVC) simulation remains the most challenging issue of Modeling and Simulation (M&S). There is a lack of interoperability, limited reuse and loose integration between the Live, Virtual and/or Constructive assets across multiple Standard Simulation Architectures (SSAs). There have been various theoretical research endeavors about solving these problems but their solutions resulted in complex and inflexible integration, long user-usage time and high cost for LVC simulation. The goal of this research is to provide an Agile Roadmap for the Live Virtual Constructive-Integrating Training Architecture (LVC-ITA) that will address the above problems and introduce interoperable LVC simulation. Therefore, this research describes how the newest M&S technologies can be utilized for LVC simulation interoperability and integration. Then, we will examine the optimal procedure to develop an agile roadmap for the LVC-ITA. In addition, this research illustrated a case study using an Adaptive distributed parallel Simulation environment for Interoperable and reusable Model (AddSIM) that is a component based integrated simulation engine. The agile roadmap of the LVC-ITA that reflects the lessons learned from the case study will contribute to guide M&S communities to an efficient path to increase interaction of M&S simulation across systems.

No description available.