En présence de plans de sondage complexes, les méthodes classiques d’estimation de la variance présentent certains défis. En effet, les estimateurs de variance usuels requièrent les probabilités d’inclusion d’ordre deux qui peuvent être complexes à obtenir pour certains plans de sondage. De plus, pour des raisons de confidentialité, les fichiers externes de microdonnées n’incluent généralement pas les probabilités d’inclusion d’ordre deux (souvent sous la forme de poids bootstrap). En s’inspirant d’une approche développée par Ohlsson (1998) dans le contexte de l’échantillonnage de Poisson séquentiel, nous proposons un estimateur ne requérant que les probabilités d’inclusion d’ordre un. L’idée est d’approximer la stratégie utilisée par l’enquête (consistant du choix d’un plan de sondage et d’un estimateur) par une stratégie équivalente dont le plan de sondage est le plan de Poisson. Nous discuterons des plans proportionnels à la taille avec ou sans grappes. Les résultats d’une étude par simulation seront présentés. / In a complex design framework, standard variance estimation methods entail substantial challenges. As we know, conventional variance estimators involve second order inclusion probabilities, which can be difficult to compute for some sampling designs. Also, confidentiality standards generally prevent second order inclusion probabilities to be included in external microdata files (often in the form of bootstrap weights). Based on Ohlsson’s sequential Poisson sampling method (1998), we suggest a simplified estimator for which we only need first order inclusion probabilities. The idea is to approximate a survey strategy (which consists of a sampling design and an estimator) by an equivalent strategy for which a Poisson sampling design is used. We will discuss proportional to size sampling and proportional to size cluster sampling. Results of a simulation study will be presented.

We consider transformations of a given random walk on a countable group determined by Markov stopping times. We prove that these transformations preserve the Poisson boundary. Moreover, under some mild conditions, the asymptotic entropy (resp., rate of escape) of the transformed random walks is equal to the asymptotic entropy (resp., rate of escape) of the original random walk multiplied by the expectation of the corresponding stopping time. This is an analogue of the well-known Abramov's formula from ergodic theory.

Using historical data from the past two seasons of the National Hockey League, three different prediction models based on Poisson regression are developed. The aim is to determine whether taking into account the recent form of a team as well as data from how they have previously performed against their opponent can help make better predictions of how many goals they will score against this opponent and thereby calculate the likelihood of each outcome. The three models are evaluated using different measures, for example comparing the odds yielded by the models against the odds of bookmakers. Different ways to account for recent form are discussed. The paper concludes that using recent form and head-to-head data will indeed improve predictions.

In this thesis a method is presented for the modelling the effects of the excluded volume (ion-ion) and ion excess polarisability (ion-solvent) interactions in an electrolyte at a smooth planar electrode. The impact of these interactions is studied in terms of the equilibrium state of single and mixed electrolytes, the dynamic response of single electrolytes to a time-dependent applied potential, and their effect on the reaction rate, for both steady and time-dependent applied potentials. For reacting systems, the reaction rate is modelled using a modified form of the Frumkin-Butler-Volmer equation, in which the interactions are explicitly accounted for. At equilibrium, the model offers improvement over models which only account for the excluded volume interaction, in terms of both the predicted electrolyte structure and the electrical properties of the electrode. For example accounting for the polarisability interaction is shown to limit and then reverse the growth in the differential capacitance at the point of zero charge as the bulk concentration increases, an effect is not seen when only the excluded volume interaction is accounted for. Another example is for mixed electrolytes, in which accounting for the polarisability interaction leads to better agreement with experimental data regarding the composition of the double layer. For the response of an electrolyte to a potential step, the two interactions both lead to peaks in the time taken to reach equilibrium as a function of the potential. The effect of the domain length on the equilibration time is qualitatively discussed, together with the differences between the two interaction models. The response to a time-dependent potential is analysed through simulated electrochemical impedance spectroscopy and consideration of the capacitance dispersion effect. Between this and the potential step response data it is shown that the interactions themselves have little direct effect on the dynamic processes beyond the way in which they limit the ion concentrations in the double layer and alter the differential capacitance of the system. The investigation of the effect of the ion interactions on the reaction rate shows that both terms can either increase or decrease the rate, relative to a system with no interactions, depending on the details of the reaction and the applied potential. This is linked to the changes in the electric field within the double layer, which are caused by the interactions, and how this affects the reactant flux in that region. In terms of simulated EIS, deviations are observed relative to the equivalent circuit for the system, the reasons for which are discussed.

Poisson process is a common model for count data. However, a global Poisson model is inadequate for sparse data such as the marked salmon recovery data that have huge extraneous variations and noise. An empirical Bayes model, which enables information to be aggregated to overcome the lack of information from data in individual cells, is thus developed to handle these data. The method fits a local parametric Poisson model to describe the variation at each sampling period and incorporates this approach with a conventional local smoothing technique to remove noise. Finally, the overdispersion relative to the Poisson model is modelled by mixing these locally smoothed, Poisson models in an appropriate way. This method is then applied to the marked salmon data to obtain the overall patterns and the corresponding credibility intervals for the underlying trend in the data. / Science, Faculty of / Statistics, Department of / Graduate

In x-ray imaging, the x-ray radiation must be the minimum necessary to achieve the required diagnostic objective, to ensure the patients safety. However, low-dose acquisitions yield images with low quality, which affect the radiologists image interpretation. Therefore, there is a compromise between image quality and radiation dose. This work proposes an image restoration framework capable of restoring low-dose acquisitions to achieve the quality of full-dose acquisitions. The contribution of the new method includes the capability of restoring images with quantum and electronic noise, pixel offset and variable detector gain. To validate the image processing chain, a simulation algorithm was proposed. The simulation generates low-dose DBT projections, starting from fulldose images. To investigate the feasibility of reducing the radiation dose in breast cancer screening programs, a simulated pre-clinical trial was conducted using the simulation and the image processing pipeline proposed in this work. Digital breast tomosynthesis (DBT) images from 72 patients were selected, and 5 human observers were invited for the experiment. The results suggested that a reduction of up to 30% in radiation dose could not be perceived by the human reader after the proposed image processing pipeline was applied. Thus, the image processing algorithm has the potential to decrease radiation levels in DBT, also decreasing the cancer induction risks associated with the exam. / Em programas de rastreamento de câncer de mama, a dose de radiação deve ser mantida o mínimo necessário para se alcançar o diagnóstico, para garantir a segurança dos pacientes. Entretanto, imagens adquiridas com dose de radiação reduzida possuem qualidade inferior. Assim, existe um equilíbrio entre a dose de radiação e a qualidade da imagem. Este trabalho propõe um algoritmo de restauração de imagens capaz de recuperar a qualidade das imagens de tomossíntese digital mamária, adquiridas com doses reduzidas de radiação, para alcançar a qualidade de imagens adquiridas com a dose de referência. As contribuições do trabalho incluem a melhoria do modelo de ruído, e a inclusão das características do detector, como o ganho variável do ruído quântico. Para a validação a cadeia de processamento, um método de simulação de redução de dose de radiação foi proposto. Para investigar a possibilidade de redução de dose de radiação utilizada na tomossíntese, um estudo pré-clínico foi conduzido utilizando o método de simulação proposto e a cadeia de processamento. Imagens clínicas de tomossíntese mamária de 72 pacientes foram selecionadas e cinco observadores foram convidados para participar do estudo. Os resultados sugeriram que, após a utilização do processamento proposto, uma redução de 30% de dose de radiação pôde ser alcançada sem que os observadores percebessem diferença nos níveis de ruído e borramento. Assim, o algoritmo de processamento tem o potencial de reduzir os níveis de radiação na tomossíntese mamária, reduzindo também os riscos de indução do câncer de mama.

Cette thèse étudie deux problèmes de nature géométrique et topologique associés au crochet de Poisson sur les variétés symplectiques. Le premier problème porte sur la notion de submersion symplectique que nous introduisons dans le présent texte et qui généralise la notion de symplectomorphisme. Il s'avère qu'une submersion symplectique est un morphisme de Poisson : il s'agit d'une application entre variétés symplectiques qui préserve le crochet de Poisson. Notre intérêt pour ces fonctions réside dans le fait que le théorème de non-tassement de Gromov porte sur l'aire minimale possible pour les images des submersions symplectiques (allant d'une boule symplectique vers le plan symplectique) obtenues comme compositions d'un plongement symplectique dans l'espace symplectique euclidien de dimension 2n et de la projection standard vers le plan de coordonnées conjuguées (p_1, q_1). Nous investiguons le problème inverse dit « de représentabilité » : nous obtenons des conditions nécessaires et suffisantes pour qu'une submersion symplectique comme ci-dessus se factorise comme précédemment à travers un plongement ou une immersion symplectique dans l'espace euclidien. Nous montrons par ailleurs qu'il existe une submersion symplectique qui ne se factorise pas de la sorte à travers une immersion et qu'il existe aussi une submersion symplectique qui se factorise de la sorte à travers une immersion, mais pas à travers un plongement. Le deuxième problème porte sur la conjecture du crochet de Poisson de Polterovich. Étant donné une variété symplectique (M, omega) et un recouvrement U de M, nous pouvons définir l'invariant pb(F) associé à une partition de l'unité F subordonnée à U, qui est une sorte de norme sur les crochets de Poisson entre les paires de fonctions de la partition. En dénotant e(U) l'énergie de disjonction de Hofer maximale d'un ouvert du recouvrement U, la conjecture demande s'il existe une constante positive C indépendante de U et de F telle que le produit de pb(F) et de e(U) soit supérieur à C. Cette conjecture a été établie récemment par Buhovski-Logunov-Tanny dans le cas des surfaces ; en nous inspirant de travaux antérieurs de Buhovski-Tanny, nous avons aussi démontré la conjecture pour les surfaces de genre plus grand que 1. Nous exposons notre approche dans le second chapitre de cette thèse. À l'aide des submersions symplectiques, nous généralisons nos méthodes afin d'attaquer la conjecture en dimensions supérieures ; nous obtenons ainsi une nouvelle preuve d'un théorème de Polterovich et de Buhovski-Tanny concernant l'invariant pb pour des recouvrements formés de petits ouverts. Afin de rendre cette thèse aussi accessible et auto-suffisante que possible, nous débutons par une introduction à la topologie symplectique. Des annexes recueillent les faits plus particuliers que nous utilisons tout au long de ce travail. / This thesis studies two problems of geometric and topological nature associated to the Poisson bracket on symplectic manifolds. The first problem concerns the notion of "symplectic submersion" that we introduce here and which generalizes the concept of symplectomorphism. A symplectic submersion turns out to be a Poisson morphism, namely a map between symplectic manifolds which preserves the Poisson bracket. Our interest in those maps stems from the fact that Gromov's nonsqueezing theorem is a statement about the minimal area possible for the images of the symplectic submersions (going from a symplectic ball to a symplectic plane) which are compositions of a symplectic embedding into the Euclidean symplectic space and of the standard projection onto the plane of conjugated variables (p_1, q_1). We investigate the inverse "representability" problem: we give necessary and sufficient conditions for a symplectic submersionas above to factorize in the previous way either through a symplectic embedding or through a symplectic immersion into Euclidean space. We show moreover that there exists a symplectic submersion which does not factorize in this way through an immersion, and also that there exists a symplectic submersion which does factorize in this way through an immersion, but not through an embedding. The second problem concerns Polterovich's Poisson bracket conjecture. Given a symplectic manifold (M, omega) and an open cover U of M, we can define the invariantpb(F) of a partition of unity F subordinated to U, which is a sort of norm on the pairwise Poisson brackets of the functions in F. Denoting e(U) the maximal Hofer displacement energy of a set in U, the conjecture asks whether there exists a positive constant C independent of U and F such that the product of pb(F) and e(U) is greater than C. This conjecture was proved recently by Buhovsky-Logunov-Tanny in the case of surfaces; based on earlier work of Buhovsky-Tanny , we also proved the conjecture for surfaces of genus one and above. We present our approach in the second chapter of this thesis. Using symplectic submersions, we generalize our methods in order to tackle the conjecture in higher dimensions; in particular, we obtain a new proof of a theorem of Polterovich and Buhovsky-Tanny about the pb invariant of covers made up of small open sets. In order to make this thesis as accessible and self-contained as possible, we first give an introduction to symplectic topology. The appendices also collect the more specialized facts we use throughout this work.

[pt] A amostragem de pontos por discos de Poisson preserva a uniformidade espacial e as propriedades de ruído azul do seu espectro de Fourier. Esse padrão de amostragem é bem popular por ser visualmente agradável o que favorece algumas aplicações. Diversos estudos se dedicam à amostragem de um único conjunto de pontos distribuídos por discos de Poisson, caracterizando uma única classe de pontos sobre domínios planares ou sobre domínios de superfícies poligonais. Uma recente técnica de amostragem sobre domínios planares estende esse método para múltiplas classes de maneira que cada classe de pontos e a união das classes sejam distribuídas por discos de Poisson. Nossa principal contribuição estende este método de amostragem em múltiplas classes sobre domínios planares para superfícies poligonais preservando a boa qualidade em cada classe e na união das classes. A independência entre os pontos de classes distintas permite ainda atributos independentes por classe e com isso apresentamos uma aplicação da distribuição de distintos objetos sobre superfícies. / [en] Poisson disk sampling preserves the spatial uniformity and blue noise properties of Fourier spectrum. This sampling pattern is very popular for its high visual quality that favors some applications. Several works are dedicated to a single set of Poisson disk sampling characterizing one class of points on the plane or on polygonal surfaces. A recent sampling process on the plane extends this method to multiple classes such that each class as well as their union keep Poisson disk proprieties. Our main contribution extends this method of multi-class Poisson disk sampling on the plane to arbitrary polygonal surfaces preserving the good quality in each class and in the union of the classes. The independence between points of different classes also allows independent attributes for each class and thus we present an application to distribute different objects on surfaces.

No description available.

Using drones for cellular coverage enhancement is a recent technology that has shown a great potential in various practical scenarios. However, one of the main challenges that limits the performance of drone-enabled wireless networks is the limited flight time. In particular, due to the limited on-board battery size, the drone needs to frequently interrupt its operation and fly back to a charging station to recharge/replace its battery. In addition, the charging station might be responsible to recharge multiple drones. Given that the charging station has limited capacity, it can only serve a finite number of drones simultaneously. Hence, in order to accurately capture the influence of the battery limitation on the performance, it is required to analyze the dynamics of the time spent by the drones at the charging stations. In this thesis, we first use tools from queuing theory and stochastic geometry to study the influence of each of the charging stations limited capacity and spatial density on the performance of a drone-enabled wireless network. We then extend our work to rural areas where users are greatly impacted by low income, high cost of backhaul connectivity, and limited resources. Considering the limitation of the electricity supply scarcity in some rural regions, we investigate the possibility and performance enhancement of the deployment of renewable energy (RE) charging stations. We outline three practical scenarios, and use simulation results to demonstrate that RE charging stations can be a possible solution to address the limited on-board battery of UAVs in rural areas, specially when they can harvest and store enough energy.