Nous étudions le comportement asymptotique de statistiques ou fonctionnelles liées à des processus à longue mémoire saisonnière. Nous nous concentrons sur les lignes de Donsker et sur le processus empirique. Les suites considérées sont de la forme $G(X_n)$ où $(X_n)$ est un processus gaussien ou linéaire. Nous montrons que les résultats que Taqqu et Dobrushin ont obtenus pour des processus à longue mémoire dont la covariance est à variation régulière à l'infini peuvent être en défaut en présence d'effets saisonniers. Les différences portent aussi bien sur le coefficient de normalisation que sur la nature du processus limite. Notamment nous montrons que la limite du processus empirique bi-indexé, bien que restant dégénérée, n'est plus déterminée par le degré de Hermite de la fonction de répartition des données. En particulier, lorsque ce degré est égal à 1, la limite n'est plus nécessairement gaussienne. Par exemple on peut obtenir une combinaison de processus de Rosenblatt indépendants. Ces résultats sont appliqués à quelques problèmes statistiques comme le comportement asymptotique des U-statistiques, l'estimation de la densité et la détection de rupture.

Ultra wide band (UWB) is an extremely promising wireless technology for researchers and industrials. One of the most interesting is its high data rate and fading robustness due to selective frequency fading. However, beside such advantages, UWB system performance is highly affected by existing narrowband interference (NBI), undesired UWB signals and tone/multi-tone noises. For this reason, research about NBI cancellation is still a challenge to improve the system performance vs. receiver complexity, power consumption, linearity, etc. In this work, the two major receiver sections, i.e., analog (radiofrequency or RF) and digital (digital signal processing or DSP), were considered and new techniques proposed to reduce circuit complexity and power consumption, while improving signal parameters. In the RF section, different multiband UWB low-noise amplifier key design parameters were investigated like circuit configuration, input matching and desired/undesired frequency band filtering, highlighting the most suitable filtering package for efficient UWB NBI cancellation. In the DSP section, due to pulse transmitter signals, different issues like modulation type and level, pulse variety, shape and color noise/tone noise assumptions, were addressed for efficient NBI cancelation. A comparison was performed in terms of bit-error rate, signal-to-interference ratio, signal-to-noise ratio, and channel capacity to highlight the most suitable parameters for efficient DSP design. The optimum number of filters that allows the filter bandwidth to be reduced by following the required low sampling rate and thus improving the system bit error rate was also investigated.

The main topic of the paper is on-line filtering for non-Gaussian dynamic (state space) models by approximate computation of the first two posterior moments using efficient numerical integration. Based on approximating the prior of the state vector by a normal density, we prove that the posterior moments of the state vector are related to the posterior moments of the linear predictor in a simple way. For the linear predictor Gauss-Hermite integration is carried out with automatic reparametrization based on an approximate posterior mode filter. We illustrate how further topics in applied state space modelling such as estimating hyperparameters, computing model likelihoods and predictive residuals, are managed by integration-based Kalman-filtering. The methodology derived in the paper is applied to on-line monitoring of ecological time series and filtering for small count data. (author's abstract) / Series: Forschungsberichte / Institut für Statistik

The topic of the paper is filtering for non-Gaussian dynamic (state space) models by approximate computation of posterior moments using numerical integration. A Gauss-Hermite procedure is implemented based on the approximate posterior mode estimator and curvature recently proposed in 121. This integration-based filtering method will be illustrated by a dynamic trend model for non-Gaussian time series. Comparision of the proposed method with other approximations ([15], [2]) is carried out by simulation experiments for time series from Poisson, exponential and Gamma distributions. (author's abstract) / Series: Forschungsberichte / Institut für Statistik

Diese Arbeit ist ein Überblick über die Ideen und Methoden der dynamischen stochastischen Modellierung von normalverteilten und nicht-normalverteilten Prozessen. Nach einer Einführung der allgemeinen Modellform werden Aussagemöglichkeiten wie Filtern, Glätten und Vorhersagen diskutiert und das Problem der Identifikation unbekannter Hyperparameter behandelt. Die allgemeinen Ausführungen werden an zwei Fallstudien, einer Zeitreihe des mittleren jährlichen Grundwasserspiegels und einer Zeitreihe von Tagesmittelwerten von SO2-Emissionen illustriert. (Autorenref.) / Series: Forschungsberichte / Institut für Statistik

This thesis describes theory and algorithms for use with Radial Basis Functions (RBFs), emphasising techniques motivated by three particular application areas. In Part I, we apply RBFs to the problem of interpolating to integral data. While the potential of using RBFs for this purpose has been established in an abstract theoretical context, their use has been lacking an easy to check sufficient condition for finding appropriate parent basic functions, and explicit methods for deriving integral basic functions from them. We present both these components here, as well as explicit formulations for line segments in two dimensions and balls in three and five dimensions. We also apply these results to real-world track data. In Part II, we apply Hermite and pointwise RBFs to the problem of surface reconstruction. RBFs are used for this purpose by representing the surface implicitly as the zero level set of a function in 3D space. We develop a multilevel piecewise technique based on scattered spherical subdomains, which requires the creation of algorithms for constructing sphere coverings with desirable properties and for blending smoothly between levels. The surface reconstruction method we develop scales very well to large datasets and is very amenable to parallelisation, while retaining global-approximation-like features such as hole filling. Our serial implementation can build an implicit surface representation which interpolates at over 42 million points in around 45 minutes. In Part III, we apply RBFs to the problem of animation control in the area of motion synthesis---controlling an animated character whose motion is entirely the result of simulated physics. While the simulation is quite well understood, controlling the character by means of forces produced by virtual actuators or muscles remains a very difficult challenge. Here, we investigate the possibility of speeding up the optimisation process underlying most animation control methods by approximating the physics simulator with RBFs.

In this thesis, a survey on pseudospectral methods for differential equations is presented. Properties of the classical orthogonal polynomials required in this context are reviewed. Differentiation matrices corresponding to Jacobi, Laguerre,and Hermite cases are constructed. A fairly detailed investigation is made for the Hermite spectral methods, which is applied to the Schr&ouml / dinger eigenvalue equation with several potentials. A discussion of the numerical results and comparison with other methods are then introduced to deduce the effciency of the method.

Vlasov-Maxwell equilibria are characterised by the self-consistent descriptions of the steady-states of collisionless plasmas in particle phase-space, and balanced macroscopic forces. We study the theory of Vlasov-Maxwell equilibria in one spatial dimension, as well as its application to current sheet and flux tube models. The ‘inverse problem' is that of determining a Vlasov-Maxwell equilibrium distribution function self-consistent with a given magnetic field. We develop the theory of inversion using expansions in Hermite polynomial functions of the canonical momenta. Sufficient conditions for the convergence of a Hermite expansion are found, given a pressure tensor. For large classes of DFs, we prove that non-negativity of the distribution function is contingent on the magnetisation of the plasma, and make conjectures for all classes. The inverse problem is considered for nonlinear ‘force-free Harris sheets'. By applying the Hermite method, we construct new models that can describe sub-unity values of the plasma beta (βpl) for the first time. Whilst analytical convergence is proven for all βpl, numerical convergence is attained for βpl = 0.85, and then βpl = 0.05 after a ‘re-gauging' process. We consider the properties that a pressure tensor must satisfy to be consistent with ‘asymmetric Harris sheets', and construct new examples. It is possible to analytically solve the inverse problem in some cases, but others must be tackled numerically. We present new exact Vlasov-Maxwell equilibria for asymmetric current sheets, which can be written as a sum of shifted Maxwellian distributions. This is ideal for implementations in particle-in-cell simulations. We study the correspondence between the microscopic and macroscopic descriptions of equilibrium in cylindrical geometry, and then attempt to find Vlasov-Maxwell equilibria for the nonlinear force-free ‘Gold-Hoyle' model. However, it is necessary to include a background field, which can be arbitrarily weak if desired. The equilibrium can be electrically non-neutral, depending on the bulk flows.

Neste trabalho formulações chamadas de "Sistemas de Parâmetros Concentrados': são usadas na obtenção de uma solução em forma fechada de um problema bidimensional de solidificação de ligas. Os sistemas concentrados são obtidos a partir do sistema diferencial original que descreve a distribuição de temperatura, resultando em modelos matemáticos mais simples que relacionam a temperatura no contorno do meio com uma nova temperatura média gerada por um processo de integração . A determinação de equações que relacionam a temperatura média e do contorno geram diferentes abordagens sendo a principal delas relacionada ao uso de fórmulas de integração numérica de Hermite, que propiciam a introdução de informações do contorno no modelo simplificado. Aqui o modelo bidimensional é abordado em meio simples e composto sendo que as equações unidimensionais simplificadas obtidas pela integração são tratadas pela Técnica das Transformadas Integrais Generalizadas (GITT). Resultados numéricos, obtidos com o software matemático Maple são apresentados. / In this work the so-called "lumped analysis'' is used to obta.in a closedform solution to a multidimensional solidification problem. The lumped systems provide simpler model, than the original one by using an integration scheme that results in average variables. Different approaches are basead on the choice of auxiliary equations to relate the average and the original variable after the integration process. Here the bidimensional problem is solved in homogeneous end in a composite medium and the resulting one-dimensional equation is solved by the Generalized Integral Tranform Technique. Numerical results are obta.ined by the symbolic software Maple.

Neste trabalho formulações chamadas de "Sistemas de Parâmetros Concentrados': são usadas na obtenção de uma solução em forma fechada de um problema bidimensional de solidificação de ligas. Os sistemas concentrados são obtidos a partir do sistema diferencial original que descreve a distribuição de temperatura, resultando em modelos matemáticos mais simples que relacionam a temperatura no contorno do meio com uma nova temperatura média gerada por um processo de integração . A determinação de equações que relacionam a temperatura média e do contorno geram diferentes abordagens sendo a principal delas relacionada ao uso de fórmulas de integração numérica de Hermite, que propiciam a introdução de informações do contorno no modelo simplificado. Aqui o modelo bidimensional é abordado em meio simples e composto sendo que as equações unidimensionais simplificadas obtidas pela integração são tratadas pela Técnica das Transformadas Integrais Generalizadas (GITT). Resultados numéricos, obtidos com o software matemático Maple são apresentados. / In this work the so-called "lumped analysis'' is used to obta.in a closedform solution to a multidimensional solidification problem. The lumped systems provide simpler model, than the original one by using an integration scheme that results in average variables. Different approaches are basead on the choice of auxiliary equations to relate the average and the original variable after the integration process. Here the bidimensional problem is solved in homogeneous end in a composite medium and the resulting one-dimensional equation is solved by the Generalized Integral Tranform Technique. Numerical results are obta.ined by the symbolic software Maple.