Dissertations / Theses on the topic 'Q-space'

Le but général de cette thèse est de proposer de nouvelles méthodes d'acquisition et de traitement du signal en imagerie par résonance magnétique (IRM) de diffusion, dans le but d'ouvrir de nouvelles perspectives dans la reconstruction de la structure de la matière blanche \emph{in vivo}. L'IRM de diffusion est une technique d'imagerie non invasive qui mesure localement, en chaque voxel, la diffusion des molécules d'eau. Le déplacement de ces dernières étant contraint par la présence de tissus, le fait de pouvoir caractériser la diffusion des molécules d'eau apporte des informations sur la nature, l'orientation, la microstructure des tissus biologiques sous-jacents. La forte anisotropie observée dans la matière blanche fait de l'IRM de diffusion un outil privilégié pour l'étude de la connectivité cérébrale. Une des premières techniques d'acquisition et de reconstruction, appelée IRM du tenseur de diffusion, est maintenant utilisée de manière routinière en clinique, pour le diagnostique de certaines maladies neurologiques, ou encore en planification préopératoire. L'IRM du tenseur de diffusion repose sur un modèle de diffusion gaussien cependant, qui est limité quand il s'agit de décrire des configurations de tissus complexes à l'intérieur d'un voxel, par exemple quand plusieurs faisceaux de fibres se croisent. Dès lors, on a cherché ces dernières années à développer des techniques qui ne reposent pas sur un modèle a priori, afin de décrire de manière plus précise le déplacement des molécules d'eau, et dépasser les limitations du modèle tensoriel. La plupart de ces techniques, dites à haute résolution angulaire, sollicitent un temps d'acquisition généralement long, et mettent en jeu des problèmes de reconstruction non triviaux. Dans la première partie de cette thèse, nous décrivons la structure microscopique des tissus de la matière blanche du cerveau, et présentons la physique de formation des images en IRM de diffusion. Nous faisons un état de l'art des méthodes de reconstruction, et des techniques d'acquisition proposées à ce jour. En ce qui concerne les méthodes de reconstruction, nous faisons la distinction suivant qu'elles soient basées sur un modèle ou non. La première contribution de cette thèse est liée à la reconstruction paramétrique du signal de diffusion dans une base de fonctions continues. Cette contribution fait suite à une méthode proposée récemment, appelée transformée de Fourier sphérique, et y apporte une modification pour une reconstruction continue. Nous réduisons de façon significative la dimension de la base, tout en décrivant aussi bien le signal de diffusion. Nous donnons également l'expression de l'opérateur de régularisation de Laplace en fonction des coefficients dans cette base, afin de limiter l'impact du bruit sur la reconstruction. La seconde contribution est également liée à la reconstruction du signal de diffusion, et à la fonction de distribution d'orientation, dans un contexte d'application clinique. Nous proposons une méthode de reconstruction en temps réel basée sur le filtre de Kalman pour la probabilité marginale de diffusion angulaire. Nous développons un algorithme pour détecter les mouvements du patient, de façon précise et avec une grande sensibilité, et ce sans surcoût, comparé aux systèmes utilisant une camera et des algorithmes de vision robotique. Les deux dernières contributions présentées dans cette thèse sont liées aux techniques d'acquisition en IRM de diffusion, en particulier pour l'élaboration de schémas d'acquisition sur une ou plusieurs sphères dans l'espace de Fourier. Nous présentons d'abord une méthode géométrique pour placer des points dans l'espace de Fourier sur plusieurs sphères, en optimisant la couverture angulaire sur chacune des sphères, mais également de façon globale. Puis nous cherchons à établir un lien entre le schéma d'acquisition et la base de fonctions utilisée pour la reconstruction, et nous proposons en particulier une méthode pour élaborer un protocole d'acquisition qui permette de minimiser le nombre de conditionnement, pour la reconstruction dans la base des harmoniques sphériques, et dans la base de Fourier sphérique modifiée, proposée dans cette thèse. En conclusion de cette étude sur l'acquisition, nous pensons que l'élaboration du schéma d'échantillonnage doit être motivée à la fois pour répondre aux contraintes physiques du scanner, et par le choix de la base dans laquelle le signal sera reconstruit. Ces nouveaux schémas d'échantillonnage sont disponibles au téléchargement sur mon site internet.

The rapid-acquisition of q-space data from 1H2O undergoing restricted diffusion in suspensions of red blood cells (RBCs) is made possible using a recently implemented pulse sequence, where the phase cycling of the radio-frequency pulses is reduced by using unbalanced pairs of bipolar magnetic field-gradient pulses. The q-space plots obtained with this pulse sequence show a shift in the position of the first diffraction minimum when compared to data from classical pulsed field gradient stimulated echo experiments. Diffusion simulations were used to investigate the effect of the additional delay introduced by the bipolar gradient pulses on the form of the q-space plots. RBCs of normal discocyte shape align with an external magnetic field, and the angular dependence of q-space spectra from suspensions of RBCs was examined using a linear combination of gradients applied along the y- and z-axes. The resulting q-space plots showed a gradual disappearance of the first diffraction minimum as the angle at which the gradients were applied was changed from 0° (along the z-axis) to~40°, beyond which the q-space plots showed no diffraction features. These experimental results were also evident in Monte-Carlo random walk simulations of diffusion in RBCs with field-gradients applied at varying angles with respect to cells aligned with B0.

Questa Tesi aspira a mostrare un codice a livello di pacchetto, che abbia performance molto vicine a quello ottimo, per progetti di comunicazioni Satellitari. L’altro scopo di questa Tesi è quello di capire se rimane ancora molto più difficile maneggiare direttamente gli errori piuttosto che le erasures. Le applicazioni per comunicazioni satellitari ora come ora usano tutte packet erasure coding per codificare e decodificare l’informazione. La struttura dell’erasure decoding è molto semplice, perché abbiamo solamente bisogno di un Cyclic Redundancy Check (CRC) per realizzarla. Il problema nasce quando abbiamo pacchetti di dimensioni medie o piccole (per esempio più piccole di 100 bits) perché in queste situazioni il costo del CRC risulta essere troppo dispendioso. La soluzione la possiamo trovare utilizzando il Vector Symbol Decoding (VSD) per raggiungere le stesse performance degli erasure codes, ma senza la necessità di usare il CRC. Per prima cosa viene fatta una breve introduzione su come è nata e su come si è evoluta la codifica a livello di pacchetto. In seguito è stato introdotto il canale q-ary Symmetric Channel (qSC), con sia la derivazione della sua capacità che quella del suo Random Coding Bound (RCB). VSD è stato poi proposto con la speranza di superare in prestazioni il Verification Based Decoding (VBD) su il canale qSC. Infine, le effettive performance del VSD sono state stimate via simulazioni numeriche. I possibili miglioramenti delle performance, per quanto riguarda il VBD sono state discusse, come anche le possibili applicazioni future. Inoltre abbiamo anche risposto alla domande se è ancora così tanto più difficile maneggiare gli errori piuttosto che le erasure.

The rapid-acquisition of q-space data from 1H2O undergoing restricted diffusion in suspensions of red blood cells (RBCs) is made possible using a recently implemented pulse sequence, where the phase cycling of the radio-frequency pulses is reduced by using unbalanced pairs of bipolar magnetic field-gradient pulses. The q-space plots obtained with this pulse sequence show a shift in the position of the first diffraction minimum when compared to data from classical pulsed field gradient stimulated echo experiments. Diffusion simulations were used to investigate the effect of the additional delay introduced by the bipolar gradient pulses on the form of the q-space plots. RBCs of normal discocyte shape align with an external magnetic field, and the angular dependence of q-space spectra from suspensions of RBCs was examined using a linear combination of gradients applied along the y- and z-axes. The resulting q-space plots showed a gradual disappearance of the first diffraction minimum as the angle at which the gradients were applied was changed from 0° (along the z-axis) to~40°, beyond which the q-space plots showed no diffraction features. These experimental results were also evident in Monte-Carlo random walk simulations of diffusion in RBCs with field-gradients applied at varying angles with respect to cells aligned with B0.

In the first chapter, a Q-switched fiber laser that is capable of generating transform-limited pulses based on single-frequency fiber laser seeded ring cavity is demonstrated. The output pulse width can be tuned from hundreds of nanoseconds to several microseconds. This Q-switched ring cavity fiber laser can operate over the whole C-band. In addition, a theoretical model is developed to numerically study the pulse characteristics, and the numerical results are in good agreements with the experimental results. In the next chapter, a Raman fiber laser is developed for generating signal at 1480 nm. Initial experimental results has demonstrated generating of Raman laser at 1175 nm, 1240 nm, 1315 nm, and 1395 nm wavelength. Finally, a free space fiber amplifier is studied both theoretically and experimentally. The experimental work has demonstrated signal coupling efficiency up to 90% in the NP highly Er/Yb co-doped phosphate fiber.

Extraction of accurate quantitative parameters to characterize water diffusion in complex porous media like brain tissue in neuroimaging is a challenging inverse problem, that depends on medium\'s structural and geometrical factors [1,3]. If the role of membranes is generally invoked, probe collisions with the insoluble cytoskeleton network and water hydrodynamic interactions with dissolved macromolecules and cytoskeleton occur as well [2]. The latter two interactions have been shown to determine the phenomenological “anomalous diffusion” of probes in the cytoplasm [4,5,6,7,8]. Additionally, the diffusion of small solutes in cytoplasm could be slowed by the local micro-viscosity of the aqueous phase, a phenomenon generally not taken into account in simulations. We suggest that the Grey and White Matter contrast in Diffusion Decay Imaging (DDI) could be caused by differences in cytoskeleton structures, composed respectively of actin and tubulin that could act by the elimination of one possible water diffusion pathlength by the volume occupied by the network phase. This could explain why anomalous DDI signal has been shown to be independent of membrane integrity [9]. Cytoplasm is able to rapidly shift from a sol (aqueous solutions embedded with insolubles particles) to a gel state (a colloidal solutions with a structured semi-solid and an aqueous fluid phase) or to a viscous solution when the insoluble particles become soluble. Does water have the ability of being a sensor of its local “self-viscosity” ? What is the length of the water diffusion\'s path compared to cells size ? Compared to this path length, how many cellular structures should be probed by water\'s translational diffusion ? We try to respond to these questions by investigating Diffusion Decay Imaging models and their effects on the hypothese-free q-space diffusion propagator shape [3], containing all informations concerning viscosity-slowed gaussian diffusion, structural informations [3] and restricted diffusion [1].

Extraction of accurate quantitative parameters to characterize water diffusion in complex porous media like brain tissue in neuroimaging is a challenging inverse problem, that depends on medium\''s structural and geometrical factors [1,3]. If the role of membranes is generally invoked, probe collisions with the insoluble cytoskeleton network and water hydrodynamic interactions with dissolved macromolecules and cytoskeleton occur as well [2]. The latter two interactions have been shown to determine the phenomenological “anomalous diffusion” of probes in the cytoplasm [4,5,6,7,8]. Additionally, the diffusion of small solutes in cytoplasm could be slowed by the local micro-viscosity of the aqueous phase, a phenomenon generally not taken into account in simulations. We suggest that the Grey and White Matter contrast in Diffusion Decay Imaging (DDI) could be caused by differences in cytoskeleton structures, composed respectively of actin and tubulin that could act by the elimination of one possible water diffusion pathlength by the volume occupied by the network phase. This could explain why anomalous DDI signal has been shown to be independent of membrane integrity [9]. Cytoplasm is able to rapidly shift from a sol (aqueous solutions embedded with insolubles particles) to a gel state (a colloidal solutions with a structured semi-solid and an aqueous fluid phase) or to a viscous solution when the insoluble particles become soluble. Does water have the ability of being a sensor of its local “self-viscosity” ? What is the length of the water diffusion\''s path compared to cells size ? Compared to this path length, how many cellular structures should be probed by water\''s translational diffusion ? We try to respond to these questions by investigating Diffusion Decay Imaging models and their effects on the hypothese-free q-space diffusion propagator shape [3], containing all informations concerning viscosity-slowed gaussian diffusion, structural informations [3] and restricted diffusion [1].

Doctor of Philosophy
Physics
Christopher M. Sorensen
We present light scattering studies of irregularly shaped particles which significantly affect the climate. We built and calibrated our apparatus which was able to measure all six independent scattering matrix elements. Our apparatus detects light from 0.32° to 157° simultaneously. We studied all six scattering matrix elements of irregularly shaped Arizona Road Dust which behave differently than those of spheres. We strongly focused on the most important scattering matrix element – the phase function, scattered intensity vs. the scattering angle, which we applied Q-space analysis to. Q-space analysis involves plotting the scattering intensity vs. the magnitude of the scattering wave vector q or qR with R the radius of a particle, on a double logarithmic scale. We measured and studied the phase functions of Al₂O₃ abrasives; compared the scattering from the abrasives with the scattering of spheres. To generalize the study, we collected a large amount of experimental and theoretical data from our group and others and applied Q-space analysis. They all displayed a common scattering pattern. The power law exponents showed a quasi-universal functionality with the internal coupling parameter ρ'. In situ studies of the soot fractal aggregates produced from a burner were also conducted. A power law exponent -1.85 is seen to imply the aggregates have fractal dimension of D[subscript f]=1.85. The overall work presented shows Q-space analysis uncovers patterns common to all particles: a q-independent forward scattering regime is followed by a Guinier regime, a power law regime, and sometimes an enhanced back scattering regime. The description of the patterns applies to spheres as well, except the power law regime has more than a single power law. These simple patterns give a unified description for all particle shapes. Moreover, the power law exponents have a quasi-universal functionality with ρ' for non-fractal aggregates. The absolute value of the exponents start from 4 when ρ' is small. As ρ' increases, the exponents decrease until the trend levels off at ρ'≳10 where the exponents reach a constant 1.75±0.25. All the non-fractal particles fall on the same trend regardless of the detail of their structure.

El presente trabajo está dedicado al análisis de una clase particular de operadores (lineales y continuos) entre espacios de Banach de funciones. El objetivo es avanzar en la teoría de los llamados operadores factorizables a la p-potencia analizando todos los aspectos de la dualidad. Esta clase de operadores ha demostrado ser de utilidad tanto en la teoría de factorización de operadores sobre espacios de Banach de funciones (teoría de Maurey-Rosenthal) como en el Análisis Armónico (dominios óptimos de la transformada de Fourier y operadores de convolución). A ¿n de desarrollar esta teoría de dualidad y sus aplicaciones, se de¿ne y estudia una nueva clase de operadores con propiedades de extensión que involucran al operador y a su adjunto. Ésta es la familia de operadores factorizables a la (p,q)- potencia, 1 · p,q Ç 1, y pueden caracterizarse mediante un esquema de factorización a través del espacio de p-potencias del dominio y el dual del espacio de q-potencias del dual del codominio. También se obtiene una equivalencia mediante un diagrama de factorización a través de espacios L p (m) y L q (n) 0 , donde m y n son medidas vectoriales adecuadas y ésta será nuestra principal herramienta. Para esta construcción resultan necesarios algunos resultados preliminares relativos a las p-potencias de los espacios de Banach de funciones que intervienen y que también se estudian. Con estos útiles se dan algunos resultados para caracterizar el rango óptimo ¿el menor espacio de Banach de funciones en el que puede tomar valores el operador¿ para operadores que van de un espacio de Banach a un espacio de Banach de funciones. Además, se desarrolla y presenta formalmente la idea de factorización óptima de un operador que optimiza una factorización previa, en términos del diagrama que debe satisfacer un operador factorizable a su (p,q)-potencia. Todos estos resultados extienden los actuales cálculos del dominio óptimo mediante medidas vectoriales para operadores sobre espacios de Banach de funciones. Dichos cálculos han dado resultados relevantes en diversas áreas del análisis matemático mediante una descripción del mayor espacio de Banach de funciones al cual, operadores relevantes ¿como la transformada de Fourier o el operador de Hardy¿ se pueden extender. La teoría se aplica para encontrar nuevos resultados en determinados campos: como la teoría de interpolación de operadores entre espacios de Banach de funciones, los operadores de núcleo y en particular, la transformada de Laplace.
Galdames Bravo, OE. (2013). Duality theory for p-th power factorable operators and kernel operators [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/31523
TESIS

Molecular diffusion is an inherent feature of all fluid systems. The processes and interactions that characterize these systems are in some way dependent upon the mobility of the component molecules. Pulsed field-gradient spin-echo nuclear magnetic resonance (PGSE NMR) is a powerful tool for the study of molecular diffusion; for heterogeneous systems, such as those typically found in biology, this technique is unsurpassed in the diversity of systems that yield to its probing. The aim of the work presented in this thesis was to use an integrated NMR-based approach, in conjunction with computer modeling, for the study of molecular diffusion in compartmentalized and multicomponent biological systems. Erythrocyte suspensions provided an ideal experimental system for the study of compartmentalized diffusion in cells. Water exchanges rapidly between the intra- and extracellular regions and, as the major constituent of the cell, provides a strong and predominant proton NMR signal. In addition, the cells are known to align in the strong static magnetic field of the spectrometer. As a consequence of these two properties, the signal intensity from a suitably designed series of PGSE NMR experiments exhibits a series of maxima and minima when graphed as a function of the magnitude of the spatial wave number vector q. The apparently periodic phenomenon is mathematically analogous to optical diffraction and interference and is referred to here as diffusion-coherence. It is the characterization of this phenomenon, with the aid of computer-based models, which was the focus of a major section of the work described herein. Two quite distinct molecular systems formed the basis of the work in which I investigated diffusion in multicomponent systems. Both systems involved molecules that undergo self-association such that at equilibrium a population distribution of different oligomeric species is present. The first of these was tropoelastin, the monomeric subunit of elastin, which under certain conditions aggregates to form a coacervate. The second system was N-methylacetamide (NMA) which also undergoes extensive self-association. NMA oligomers have previously been studied as peptide analogues due to the presence in the monomer of a peptide linkage. In this work the aim was to use PGSE NMR diffusion measurements, in a manner that is in many ways analogous to analytical ultracentrifugation, to obtain estimates of hydrodynamic and thermodynamic parameters. Computer modeling was also used extensively in this section of work for the interpretation of the experimental data.

The displacement operator is related to the displaced parity operator through a two dimensional Fourier transform. Both operators are important operators in phase space and the trace of both with respect to the density operator gives the Wigner functions (displaced parity operator) and Weyl functions (displacement operator). The generalisation of the parity-displacement operator relationship considered here is called the bi-fractional displacement operator, O(α, β; θα, θβ). Additionally, the bi-fractional displacement operators lead to the novel concept of bi-fractional coherent states. The generalisation from Fourier transform to fractional Fourier transform can be applied to other phase space functions. The case of the Wigner-Weyl function is considered and a generalisation is given, which is called the bi-fractional Wigner functions, H(α, β; θα, θβ). Furthermore, the Q−function and P−function are also generalised to give the bi-fractional Q−functions and bi-fractional P−functions respectively. The generalisation is likewise applied to the Moyal star product and Berezin formalism for products of non-commutating operators. These are called the bi-fractional Moyal star product and bi-fractional Berezin formalism. Finally, analysis, applications and implications of these bi-fractional transforms to the Heisenberg uncertainty principle, photon statistics and future applications are discussed.

The displacement operator is related to the displaced parity operator through a two dimensional Fourier transform. Both operators are important operators in phase space and the trace of both with respect to the density operator gives the Wigner functions (displaced parity operator) and Weyl functions (displacement operator). The generalisation of the parity-displacement operator relationship considered here is called the bi-fractional displacement operator, O(α, β; θα, θβ). Additionally, the bi-fractional displacement operators lead to the novel concept of bi-fractional coherent states. The generalisation from Fourier transform to fractional Fourier transform can be applied to other phase space functions. The case of the Wigner-Weyl function is considered and a generalisation is given, which is called the bi-fractional Wigner functions, H(α, β; θα, θβ). Furthermore, the Q−function and P−function are also generalised to give the bi-fractional Q−functions and bi-fractional P−functions respectively. The generalisation is likewise applied to the Moyal star product and Berezin formalism for products of non-commutating operators. These are called the bi-fractional Moyal star product and bi-fractional Berezin formalism. Finally, analysis, applications and implications of these bi-fractional transforms to the Heisenberg uncertainty principle, photon statistics and future applications are discussed.

En géométrie algébrique, les invariants de Gromov—Witten sont des invariants énumératifs qui comptent le nombre de courbes complexes dans une variété projective lisse qui vérifient des conditions d’incidence. En 2001, A. Givental et Y.P. Lee ont défini de nouveaux invariants, dits de Gromov—Witten K-théoriques, en remplaçant les définitions cohomologiques dans la construction des invariants de Gromov—Witten par leurs analogues K-théoriques. Une question essentielle est de comprendre comment sont reliées ces deux théories. En 2013, Iritani- Givental-Milanov-Tonita démontrent que les invariants K-théoriques peuvent être encodés dans une fonction qui vérifie des équations aux q-différences. En général, ces équations fonctionnelles vérifient une propriété appelée “confluence”, selon laquelle on peut dégénérer ces équations pour obtenir une équationdifférentielle. Dans cette thèse, on propose de comparer les deux théories de Gromov— Witten à l’aide de la confluence des équations aux q-différences. On montre que, dans le cas des espaces projectifs complexes, que ce principe s’adapte et que les invariants Kthéoriques peuvent être dégénérés pour obtenir leurs analogues cohomologiques. Plus précisément, on montre que la confluence de la petite fonction J de Givental K-théorique permet de retrouver son analogue cohomologique après une transformation par le caractère de Chern
In algebraic geometry, Gromov— Witten invariants are enumerative invariants that count the number of complex curves in a smooth projective variety satisfying some incidence conditions. In 2001, A. Givental and Y.P. Lee defined new invariants, called Ktheoretical Gromov—Witten invariants. These invariants are obtained by replacing cohomological objects used in the definition of the usual Gromov—Witten invariants by their Ktheoretical analogues. Then, an essential question is to understand how these two theories are related. In 2013, Iritani-Givental- Milanov-Tonita show that K-theoretical Gromov—Witten invariants can be embedded in a function which satisfies a q-difference equation. In general, these functional equations verify a property called “confluence”, which guarantees that we can degenerate these equations to obtain a differential equation. In this thesis, we propose to compare our two Gromov—Witten theories through the confluence of q-difference equations. We show that, in the case of complex projective spaces, this property can be adapted to degenerate Ktheoretical invariants into their cohomological analogues. More precisely, we show that theconfluence of Givental’s small K-theoretical Jfunction produces its cohomological analogue after applying the Chern character

Electromagnetic wave absorbing properties of SiC-based and alumina ceramic woven fabrics have been investigated. Electrical conductivities of SiC-based ceramic woven fabrics were modified by heat treatment in air resulting in their oxidation. Surface properties of alumina woven fabrics were altered by gold-sputtering resulting in a high conductivity layer on the surface of the wovens. Electromagnetic wave interactions of single layer and double layered combinations of these ceramic woven fabrics were determined in 17-40 GHz frequency range using &ldquo
free-space&rdquo
method. Electromagnetic wave absorption potential of ceramic woven fabrics with different chemical compositions and woven types were correlated with their material properties by X-ray diffraction, scanning electron microscopy and electrical conductivity measurements. Effects of modifications and varying woven fabric arrangements in combinations on the electromagnetic wave absorption potential of the ceramic woven fabrics have been discussed. Various double layer combinations of SiC-based and alumina woven fabrics revealed a promising potential in terms of both reduced reflection and transmission resulting in more than ~95% absorption in millimeter wavelength range, which makes them powerful candidate materials for electromagnetic wave absorption applications.

The main focus of my dissertation is the Modified Stochastic Sine-Gordon Equation: utt = 2uxx − ut − sin(|u|^(&gamma)) + b(u, du/dt)dW/dt where &gamma > 0 is the parameter of the power of non-linearity, &delta &ge 0 is the magnitude of non-linearity, &alpha> 0 be the damping parameter, and &sigma the diffusion intensity, on one dimensional domain. We analyze the properties of the solution of the SPDE by the eigenfunctions approach allowing us to truncate the infinite-dimensional stochastic system (i.e the SDEs of Fourier coefficients related to the SPDE), to control its energy, existence, uniqueness, continuity and stability. The analysis relies on the investigation of expected Lyapunov functional of the energy in terms of all system-parameters. We simulate the model with respect to all system-parameters to visualize our conclusions.

Le sujet de cette thèse est dédié au développement d'une technique, l'IRM de diffusion, qui permet d'obtenir des images de propriétés micro-structurelles (inférieures à la taille du pixel obtenu par IRM) des milieux biologique, et à son application à l'étude du cerveau. La capacité de cette technique à révéler des modifications précoces de changement micro-structuraux qui sont associés à des modifications métaboliques énergétiques complexes en a fait une méthode de référence pour détecter précocement certaines pathologies focales telles les Accidents Vasculaires Cérébraux Ischémiques. Le lecteur pourra trouver dans cette thèse une introduction complète aux phénomènes physiques de la diffusion brownienne dans les milieux biologiques, de l'IRM et de la RMN de diffusion ainsi qu'une synthèse originale sur les origine biologiques et biophysiques des modifications de Coefficient de Diffusion Apparent observés dans l'ischémie. Afin d'étendre le domaine de l'IRM de diffusion des phénomènes ischémiques focaux (étudiés expérimentalement sur l'homme et l'animal) aux pathologies non focales, l'étude de la déviation à la diffusion gaussienne de l'eau dans les tissus biologiques a été étudiée du point de vue bibliographique et expérimental. Les méthodologies pratiques permettant de traiter et de préparer les images de diffusion pour l'étude des diffusion non gaussienne ainsi que les corrections d'artefacts nécessaires y sont soigneusement décrites. Ceci a pu donner lieu à une étude des modèles de diffusion non gaussienne en tant que problématique inverse et à des applications dans la détection de la Maladie d'Alzheimer, caractérisée par des lésions peu focales et microscopiques. Enfin, nous avons mis au point trois approches originales de développement technologiques de séquences d'IRM de diffusion ainsi que les traitements d'images associés nécessaires à leur exploitation. Le premier développement inclut les mesures de diffusion non gaussienne avec variation des temps de diffusion appliquée à l'imagerie à 4. 7 et 7 T. Le second a consisté à mettre en place et tester à 3 T des séquences combinées de diffusion et de transfert d'aimantation permettant d'apporter des indications additionnelles sur la nature de l'eau sondée par l'IRM de diffusion. La troisième approche a consisté à développer à 3 T une méthodologie d'IRM fonctionnelle du tenseur de diffusion destinée à expérimenter les postulats biologiques résumés dans la première partie de cette thèse concernant le rôle potentiel de l'eau et de sa biologie dans les phénomènes d'activation cérébrales fonctionnelles. Au fil des validations, les hypothèses sur les micro-structures des milieux biologiques sont testées et affinées par différentes approches in vivo, ex vivo et in silico, afin d'appliquer les avancées physiques récentes en l'IRM de diffusion à la détection médicale des pathologies focales et non focales et interpréter celle-ci
This PhD thesis is dedicated to a technique, diffusion MRI, which allow to obtain images of micro-structural properties (inferior to the MRI voxel size) of biological media, and to the application of this technique to study brain. Because of its ability to reveal early micro-structural changes (associated with complex energetic metabolism changes), diffusion MRI is become a reference method to detect focal diseases like ischemic stroke. The reader can find in this thesis a complete introduction to the physical phenomenon related to brownian motion in biological media and those related to diffusion NMR and MRI, and an original synthesis of the biological and biophysical determinisms of the changes of apparent diffusion coefficients observed in stroke animal models. To extend the field of the technique from stroke focal phenomenon (studied experimentally in man an rodents) to non focal pathologies, the study of the deviation of diffusion from Gaussian behaviour has been studied theoretically and experimentally. Pratical methodologies allowing the preparation of diffusion images for non-gaussian diffusion imaging, and artefacts corrections are described here. This work has lead to a study of non-gaussian diffusion MRI signal treated as an inverse problem and to applications for Alzheimer's disease detection, characterized by non-focal and microscopic lesions. Finally, we have developed three original approaches for technological developments of MRI sequences (with the associated image treatment necessary to use them). The first is the development of non-gaussian diffusion together with variation of diffusion time applied to imaging at 4. 7 and 7 T. The second concern the development of magnetization transfer and diffusion imaging that give additional information about water probed by MRI. The latter approach is the development of fonctionnal diffusion MRI at 3 T in DTI mode dedicated to apply the biological hypothesis resumed in the first part of this thesis, concerning the particular role of water in brain activation. With a progression for the experimental validations, hypothesis concerning micro-structures of biological media are tested and validated with different approaches (in vivo, ex vivo, in silico), to apply the recent discoveries concerning the physic of diffusion MRI in order to detect focal and non-focal pathologies and to interpret them

Cette thèse est dédiée à améliorer la compréhension neuro-scientifique à l'aide d'imagerie par résonance magnétique de diffusion (IRMd). Nous nous concentrons sur la modélisation du signal de diffusion et l'estimation par IRMd des biomarqueurs liés à la microstructure, appelé «Microstructure Imaging». Cette thèse est organisée en trois parties. Dans partie I nous commençons par la base de l'IRMd et un aperçu de l'anisotropie en diffusion. Puis nous examinons la plupart des modèles de microstructure utilisant PGSE, en mettant l'accent sur leurs hypothèses et limites, suivi par une validation par l'histologie de la moelle épinière de leur estimation. La partie II présente nos contributions à l'imagerie en 3D et à l’estimation de microstructure. Nous proposons une régularisation laplacienne de la base fonctionnelle MAP, ce qui nous permet d'estimer de façon robuste les indices d'espace q liés au tissu. Nous appliquons cette approche aux données du Human Connectome Project, où nous l'utilisons comme prétraitement pour d'autres modèles de microstructure. Enfin, nous comparons les biomarqueurs dans une étude ex-vivo de rats Alzheimer à différents âges. La partie III présente nos contributions au représentation de l’espace qt - variant sur l'espace q 3D et le temps de diffusion. Nous présentons une approche initiale qui se concentre sur l'estimation du diamètre de l'axone depuis l'espace qt. Nous terminons avec notre approche finale, où nous proposons une nouvelle base fonctionnelle régularisée pour représenter de façon robuste le signal qt, appelé qt-IRMd. Ce qui permet l'estimation des indices d’espace q dépendants du temps, quantifiant la dépendance temporelle du signal IRMd
This thesis is dedicated to furthering neuroscientific understanding of the human brain using diffusion-sensitized Magnetic Resonance Imaging (dMRI). Within dMRI, we focus on the estimation and interpretation of microstructure-related markers, often referred to as ``Microstructure Imaging''. This thesis is organized in three parts. Part I focuses on understanding the state-of-the-art in Microstructure Imaging. We start with the basic of diffusion MRI and a brief overview of diffusion anisotropy. We then review and compare most state-of-the-art microstructure models in PGSE-based Microstructure Imaging, emphasizing model assumptions and limitations, as well as validating them using spinal cord data with registered ground truth histology. In Part II we present our contributions to 3D q-space imaging and microstructure recovery. We propose closed-form Laplacian regularization for the recent MAP functional basis, allowing robust estimation of tissue-related q-space indices. We also apply this approach to Human Connectome Project data, where we use it as a preprocessing for other microstructure models. Finally, we compare tissue biomarkers in a ex-vivo study of Alzheimer rats at different ages. In Part III, we present our contributions to representing the qt-space - varying over 3D q-space and diffusion time. We present an initial approach that focuses on 3D axon diameter estimation from the qt-space. We end with our final approach, where we propose a novel, regularized functional basis to represent the qt-signal, which we call qt-dMRI. Our approach allows for the estimation of time-dependent q-space indices, which quantify the time-dependence of the diffusion signal

Oxidative stress occurs when cells are unable to cope with the levels of various reactive oxygen species (ROS) that arise as part of regular cellular metabolism or in response to ionising radiation (H2O2, O2-, OH-). The most well studied ROS is H2O2, due to its dual role as a mediator of oxidative stress and a signalling molecule for many cellular pathways. Cells possess a number of different mechanisms to combat ROS, in order to prevent their levels from becoming toxic. In this thesis, we studied three different aspects of the antioxidant defence in Saccharomyces cerevisiae. In the first part, we explored the role of erythroascorbic acid – the yeast analogue of ascorbic acid (vitamin C) – and attempted to determine its role as an antioxidant in yeast. Our results were inconclusive, though there were indications that the presence of erythroascorbic acid may have a protective effect on the mitochondrial inner membrane potential (ΔΨ), protecting it from depolarisation. The second part focused on elucidating the mitochondrial targeting of the main H2O2 sensor Gpx3 and, more specifically, whether the Yap1-binding proteins, Ybp1 and Ybp2, have an effect on the import of Gpx3 in yeast mitochondria. Our results show a slight effect of Ybp1 (but not Ybp2) on the import of Gpx3, indicating that Ybp1 may act as a chaperone for the more efficient targeting of Gpx3 from the cytosol to the outer mitochondrial membrane and, as a result, its eventual translocation into the IMS. The final part of this thesis focused on elucidating the import of Trx1 and Trr1 in the mitochondrial IMS, as well as their function in this particular subcompartment. The discovery of two members of the thioredoxin system in the IMS is important, due to the absence of a known reducing mechanism in this oxidising compartment. Our results determined that several well-known import factors are dispensable for the import of either Trx1 or Trr1, indicating that they follow a yet unknown pathway for their translocation into the IMS. Importantly, we showed that Trx1 is reduced (and thus, active) in the IMS and that it can interact in vitro with both components of the MIA machinery (Mia40 and Erv1), while in organello experiments showed that Trx1 most probably interacts with a large number of Mia40 substrates.

U ovoj tezi dokazane su teoreme o nepokretnoj tački koje predstavljaju jednoznačna i višeznačna uopštenja Banahovog prin­cipa kontrakcije u verovatnosnim metričkim i fazi metričkim pros­torima. Dokazana je teorema koja predstavlja uopštenje teoreme o nepokretnoj tački za verovatnosnu ^-kontrakciju / :  S —* S,gde je  ( S ,  J7, T ) kompletan Mengerov prostor. Uveden je pojam jake (6n)-kontrakcije i dokazana je teorema koja predstavlja uopštenje teoreme Sehgala i Bharuche-Reid kada je preslikavanje / :  S —> S jaka (6n)-kontrakcija. Teorema Caristija, koja predstavlja jedan od najvažnijih rezultata za teoriju nepokretne tačke i nelinearnu analizu uopštena je u kompletnom Mengerovom prostoru  (S ,  F , T ), gde je t-norma  T H -tipa. Kako Mengerovi prostori pripadaju klasi kvazi-uniformnih prostora dokazana je teorema o nepokretnoj tački tri preslikavanja u jednoj specijalnoj klasi kvazi-uniformnih prostora. Dokazana je teorema o nepokretnoj tački koja predstavlja verovatnosno uopštenje Nadlerove g-kontrakcije za tri preslikavanja kao i uopštenje Hiksovog principa kontrakcije za tri preslikavanja. Teorijakontraktora, koju je uveo M. Altman, odnosi se na rešavanje nelin­earnih operatorskih jednačina u Banahovim prostorima. U tezi su dokazane teoreme koje obezbeđuju postojanje i jedinstvenost rešenja za nelinearne operatorske jednačine sa jednoznačnim i višeznačnim operatorom u nearhimedovskim Mengerovim verovatnosnim normi­ ranim prostorima.
In this thesis fixed point theorems which present singleval­ued and multivalued generalization of Banach contraction principlein probabilistic metric and fuzzy metric spaces are proved. Thetheorem which presents generalization of fixed point theorem forprobabilistic g-contraction / :  S —* S is proved, where ( S ,  J7, T ) iscomplete Menger space. A notion of strong (£>n)-contraction is in­troduced and the theorem which presents a generalization of Sehgaland Bharucha-Raid theorem when the mapping / :  S  —»  S is strong(6n)-contraction is proved. Caristi’s theorem, which presents one ofthe most imortant results for the fixed point theorem and nonlinearanalysis is generalized in complete Menger space  (S, J-, T ), wheret-norm  T is of  H -type. As Menger’s spaces belong to the class ofquasi-uniformizable spaces, the fixed point theorem for three map­pings in one special class of quasi-uniformizable spaces is proved.The fixed point theorem which presents a probabilistic generaliza­tion of Nadler g-contraction for three mappings is proved as wellas the generalization of Hicks’s contraction principle for three map­pings. The theory of contractor, which was introduced by M. A lt­man refers to solving nonlinear operator equations in Banach spaces.This thesis proves the theorems which provide the existence anduniqueness of the solutions for nonlinear operator equations withsinglevalued and multivalued operators in nonarhimedian Menger’sprobabilistic normed spaces

Cette thèse est consacrée à l'élaboration de nouvelles méthodes d'acquisition et de traitement de données en IRM de diffusion (IRMd) afin de caractériser la diffusion des molécules d'eau dans les fibres de matière blanche à l'échelle d'un voxel. Plus particulièrement, nous travaillons sur un moyen de reconstruction précis de l'Ensemble Average Propagator (EAP), qui représente la fonction de probabilité de diffusion des molécules d'eau. Plusieurs modèles de diffusion tels que le tenseur de diffusion ou la fonction de distribution d'orientation sont très utilisés dans la communauté de l'IRMd afin de quantifier la diffusion des molécules d'eau dans le cerveau. Ces modèles sont des représentations partielles de l'EAP et ont été développés en raison du petit nombre de mesures nécessaires à leurs estimations. Cependant, il est important de pouvoir reconstruire précisément l'EAP afin d'acquérir une meilleure compréhension des mécanismes du cerveau et d'améliorer le diagnostique des troubles neurologiques. Une estimation correcte de l'EAP nécessite l'acquisition de nombreuses images de diffusion sensibilisées à des orientations différentes dans le q-space. Ceci rend son estimation trop longue pour être utilisée dans la plupart des scanners cliniques. Dans cette thèse, nous utilisons des techniques de reconstruction parcimonieuses et en particulier la technique connue sous le nom de Compressive Sensing (CS) afin d'accélérer le calcul de l'EAP. Les multiples aspects de la théorie du CS et de son application à l'IRMd sont présentés dans cette thèse.

This thesis is divided into the following three parts. A categorical reconstruction of crystals and quantum groups at q = 0. The quantum co-ordinate algebra A_q(𝔤) associated to a KacMoody Lie algebra g forms a Hopf algebra whose comodules are direct sums of finite dimensional irreducible U_q(𝔤) modules. In Part I we investigate whether an analogous result is true when q = 0. We classify crystal bases as coalgebras over a comonadic functor on the category of pointed sets and encode the monoidal structure of crystals into a bicomonadic structure. In doing this we prove that there is no coalgebra in the category of pointed sets whose comodules are equivalent to crystal bases. We then construct a bialgebra over ℤ whose based comodules are equivalent to crystals, which we conjecture is linked to Lusztig's quantum group at v = ∞. A Tannakian Reconstruction Theorem for IndBanach Spaces. Classically, Tannaka-Krein duality allows us to reconstruct a (co)algebra from its category of representation. In Part II we present an approach that allows us to generalise this theory to the setting of Banach spaces. This leads to several interesting applications in the directions of analytic quantum groups, bounded cohomology and Galois descent. A large portion of Part II is dedicated to such examples. On analytic analogues of quantum groups. In Part III we present a new construction of analytic analogues of quantum groups over non-Archimedean fields and construct braided monoidal categories of their representations. We do this by constructing analytic Nichols algebras and use Majid's double-bosonisation construction to glue them together. We then go on to study the rigidity of these analytic quantum groups as algebra deformations of completed enveloping algebras through bounded cohomology. This provides the first steps towards a p-adic Drinfel'd-Kohno Theorem, which should relate this work to Furusho's p-adic Drinfel'd associators. Finally, we adapt these constructions to working over Archimedean fields.

De nombreux phénomènes naturels sont trop complexes pour être pleinement reconnus par un seul instrument de mesure ou par une seule modalité. Par conséquent, le domaine de recherche de la multi-modalité a émergé pour mieux identifier les caractéristiques riches du phénomène naturel de la multi-propriété naturelle, en analysant conjointement les données collectées à partir d’uniques modalités, qui sont en quelque sorte complémentaires. Dans notre étude, le phénomène d’intérêt multi-propriétés est l’activité du cerveau humain et nous nous intéressons à mieux la localiser au moyen de ses propriétés électromagnétiques, mesurables de manière non invasive. En neurophysiologie, l’électroencéphalographie (EEG) et la magnétoencéphalographie (MEG) constituent un moyen courant de mesurer les propriétés électriques et magnétiques de l’activité cérébrale. Notre application dans le monde réel, à savoir le problème de reconstruction de source EEG / MEG, est un problème fondamental en neurosciences, allant des sciences cognitives à la neuropathologie en passant par la planification chirurgicale. Considérant que le problème de reconstruction de source EEG /MEG peut être reformulé en un système d’équations linéaires sous-déterminé, la solution (l’activité estimée de la source cérébrale) doit être suffisamment parcimonieuse pour pouvoir être récupérée de manière unique. La quantité de parcimonie est déterminée par les conditions dites de récupération. Cependant, dans les problèmes de grande dimension, les conditions de récupération conventionnelles sont extrêmement strictes. En regroupant les colonnes cohérentes d’un dictionnaire, on pourrait obtenir une structure plus incohérente. Cette stratégie a été proposée en tant que cadre d’identification de structure de bloc, ce qui aboutit à la segmentation automatique de l’espace source du cerveau, sans utiliser aucune information sur l’activité des sources du cerveau et les signaux EEG / MEG. En dépit du dictionnaire structuré en blocs moins cohérent qui en a résulté, la condition de récupération conventionnelle n’est plus en mesure de calculer la caractérisation de la cohérence. Afin de relever le défi mentionné, le cadre général des conditions de récupération exactes par bloc-parcimonie, comprenant trois conditions théoriques et une condition dépendante de l’algorithme, a été proposé. Enfin, nous avons étudié la multi-modalité EEG et MEG et montré qu’en combinant les deux modalités, des régions cérébrales plus raffinées sont apparues
Three main challenges have been addressed in this thesis, in three chapters.First challenge is about the ineffectiveness of some classic methods in high-dimensional problems. This challenge is partially addressed through the idea of clustering the coherent parts of a dictionary based on the proposed characterisation, in order to create more incoherent atomic entities in the dictionary, which is proposed as a block structure identification framework. The more incoherent atomic entities, the more improvement in the exact recovery conditions. In addition, we applied the mentioned clustering idea to real-world EEG/MEG leadfields to segment the brain source space, without using any information about the brain sources activity and EEG/MEG signals. Second challenge raises when classic recovery conditions cannot be established for the new concept of constraint, i.e., block-sparsity. Therefore, as the second research orientation, we developed a general framework for block-sparse exact recovery conditions, i.e., four theoretical and one algorithmic-dependent conditions, which ensure the uniqueness of the block-sparse solution of corresponding weighted mixed-norm optimisation problem in an underdetermined system of linear equations. The mentioned generality of the framework is in terms of the properties of the underdetermined system of linear equations, extracted dictionary characterisations, optimisation problems, and ultimately the recovery conditions. Finally, the combination of different information of a same phenomenon is the subject of the third challenge, which is addressed in the last part of dissertation with application to brain source space segmentation. More precisely, we showed that by combining the EEG and MEG leadfields and gaining the electromagnetic properties of the head, more refined brain regions appeared

L'optomécanique est la science des interactions entre la lumière et les mouvements mécaniques. Ce rapport de thèse décrit des expériences réalisées avec des microdisques fabriqué dans différents résonateurs semi-conducteurs III-V: l'Arséniure de Gallium (GaAs), l'Arséniure d'Aluminium Gallium (AlGaAs) et l'Arséniure d'Indium Phosphide (InGaP). Ces matériaux sont compatibles avec les fonctionnalités de l’optoélectronique et procurent un couplage optomécanique géant. Pour améliorer les performances des résonateurs en GaAs, nous avons développé des méthodes de traitement de surface permettant de réduire la dissipation optique par un facteur dix et ainsi d'atteindre un facteur de qualité de six millions. En plus de ces études sur le GaAs, nous avons réalisés une étude comparative des interactions optomecaniques dans des microdisques d'InGaP et d'AlGaAs, et nous avons mis en évidences leurs résonances optomécaniques. Finalement, nous avons réalisé des mesures de force avec des résonateurs en GaAs, démontrant un nouveau principe de détection basé sur notre étude de leur la trajectoire dans l'espace de phase et leur bruit de phase
Optomechanics studies the interaction between light and mechanical motion. This PhD thesis reports on optomechanical experiments carried with miniature disk resonators fabricated out of distinct III-V semiconductors: Gallium Arsenide (GaAs), Aluminium Gallium Arsenide (AlGaAs) and Indium Gallium Phosphide (InGaP). These materials are compliant with optoelectronics functionalities and provide giant optomechanical coupling. In order to boost performances of GaAs resonators, we implemented surface control techniques and obtained a ten-fold reduction of optical dissipation, attaining a Q of six million. On top of GaAs, we performed a comparative investigation of optomechanical interactions in InGaP and AlGaAs disk resonators, and demonstrated their operation as optomechanical oscillators. Finally, we carried out optomechanical force sensing experiments with GaAs resonators, analyzing a new sensing principle in light of the phase space trajectory and phase noise of the corresponding oscillators

This thesis is twofold. In the first part, a proof of the interpolation inequality along geodesics in p-Wasserstein spaces is given and a new curvature condition on abstract metric measure spaces is defined. In the second part of the thesis a proof of the identification of the q-heat equation with the gradient flow of the Renyi (3-p)-Renyi entropy functional in the p-Wasserstein space is given. For that, a further study of the q-heat flow is presented including a condition for its mass preservation.

The purpose of this study was to examine how academic deans perceived the characteristics of facility built environment and its impact on learning in higher education. Q methodology was used as the means to explore the subjective opinions of academic deans within the State of Florida regarding the facility built environment’s impact on learning in higher education. For this Q study, the concourse statements were the result of communications taken from the subject literature and participant responses to this study’s online concourse questionnaire. The resulting 32 item Q sample was sorted online by 43 academic deans, associate and assistant deans. In completing the survey, the participants ranked statements representative of the characteristics of facility built environment according to their own beliefs and subjective opinions. From the resulting data and subsequent analysis, three distinct factors emerged that represented the collective opinions of this study’s participants. The emergent factors for this study were named Traditionalist – Focused on Functionality and Universal Rationality; Modernist – Technology Conscious Seeking Innovation and Flexibility; and Abstractionist – Contextual and Expressive.

This thesis tackles issues of particular interest regarding analysis and design of passive components at the mm-wave and Terahertz (THz) bands. Innovative analysis techniques and modeling of complex structures, design procedures, and practical implementation of advanced passive devices are presented. The first part of the thesis is dedicated to THz passive components. These days, THz technology suffers from the lack of suitable waveguiding structures since both, metals and dielectric, are lossy at THz frequencies. This implies that neither conventional closed metallic structures used at microwave frequencies, nor dielectric waveguides used in the optical regime, are adequate solutions. Among a variety of new proposals, the Single Wire Waveguide (SWW) stands out due to its low attenuation and dispersion. However, this surface waveguide presents difficult excitation and strong radiation on bends. A Dielectric-Coated Single Wire Waveguide (DCSWW) can be used to alleviate these problems, but advantages of the SWW are lost and new problems arise. Until now, literature has not given proper solution to radiation on bends and, on the other hand, rigorous characterization of these waveguides lacks these days. This thesis provides, for the first time, a complete modal analysis of both waveguides, appropriated for THz frequencies. This analysis is later applied to solve the problem of radiation on bends. Several structures and design procedures to alleviate radiation losses are presented and experimentally validated. The second part of the thesis is dedicated to mm-wave passive components. These days, when implementing passive components to operate at such small, millimetric wavelengths, to ensure proper metallic contact and alignment between parts results challenging. In addition, dielectric absorption becomes significant at mm-wave frequencies. Consequently, conventional hollow metallic waveguides and planar transmission lines present high attenuation so that new topologies are being considered. Gap Waveguides (GWs), based on a periodic structure introducing an Electromagnetic Bandgap effect, result very suitable since they do not require metallic contacts and avoid dielectric losses. However, although GWs have great potential, several issues prevent GW technology from becoming consolidated and universally used. On the one hand, the topological complexity of GWs difficulties the design process since full-wave simulations are time-costly and there is a lack of appropriate analysis methods and suitable synthesis procedures. On the other hand, benefits of using GWs instead of conventional structures are required to be more clearly evidenced with high-performance GW components and proper comparatives with conventional structures. This thesis introduces several efficient analysis methods, models, and synthesis techniques that will allow engineers without significant background in GWs to straightforwardly implement GW devices. In addition, several high-performance narrow-band filters operating at Ka-band and V-band, as well as a rigorous comparative with rectangular waveguide topology, are presented.
Esta tesis aborda problemas actuales en el análisis y diseño de componentes pasivos en las bandas de onda milimétrica y Terahercios (THz). Se presentan nuevas técnicas de análisis y modelado de estructuras complejas, procedimientos de diseño, e implementación práctica de dispositivos pasivos avanzados. La primera parte de la tesis se dedica a componentes pasivos de THz. Actualmente no se disponen de guías de onda adecuadas a THz debido a que ambos, metales y dieléctricos, introducen grandes pérdidas. En consecuencia, no es adecuado escalar las estructuras metálicas cerradas usadas en microondas, ni las guías dieléctricas usadas a frecuencias ópticas. Entre un gran número de recientes propuestas, la Single Wire Waveguide (SWW) destaca por su baja atenuación y casi nula dispersión. No obstante, como guía superficial, la SWW presenta difícil excitación y radiación en curvas. El uso de un recubrimiento dieléctrico, creando la Dielecric-Coated Single Wire Waveguide (DCSWW), alivia estos inconvenientes, pero las ventajas anteriores se pierden y nuevos problemas aparecen. Hasta la fecha, no se han encontrado soluciones adecuadas para la radiación en curvas de la SWW. Además, se echa en falta una caracterización rigurosa de ambas guías. Esta tesis presenta, por primera vez, un análisis modal completo de SWW y DCSWW, adecuado a la banda de THz. Este análisis es aplicado posteriormente para evitar el problema de la radiación en curvas. Se presentan y validan experimentalmente diversas estructuras y procedimientos de diseño. La segunda parte de la tesis abarca componentes pasivos de ondas milimétricas. Actualmente, estos componentes sufren una importante degradación de su respuesta debido a que resulta difícil asegurar contacto metálico y alineamiento adecuados para la operación a longitudes de onda tan pequeñas. Además, la absorción dieléctrica incrementa notablemente a estas frecuencias. En consecuencia, tanto guías metálicas huecas como líneas de transmisión planares convencionales presentan gran atenuación, siendo necesario considerar topologías alternativas. Las Gap Waveguides (GWs), basadas en una estructura periódica que introduce un efecto de Electromagnetic Bandgap, resultan muy adecuadas puesto que no requieren contacto entre partes metálicas y evitan las pérdidas en dieléctricos. No obstante, a pesar del potencial de las GWs, varias barreras impiden la consolidación y uso universal de esta tecnología. Por una parte, la compleja topología de las GWs dificulta el proceso de diseño dado que las simulaciones de onda completa consumen mucho tiempo y no existen actualmente métodos de análisis y diseño apropiados. Por otra parte, es necesario evidenciar el beneficio de usar GWs mediante dispositivos GW de altas prestaciones y comparativas adecuadas con estructuras convencionales. Esta tesis presenta diversos métodos de análisis eficientes, modelos, y técnicas de diseño que permitirán la síntesis de dispositivos GW sin necesidad de un conocimiento profundo de esta tecnología. Asimismo, se presentan varios filtros de banda estrecha operando en las bandas Ka y V con altas prestaciones, así como una comparativa rigurosa con la guía rectangular.
Aquesta tesi aborda problemes actuals en relació a l'anàlisi i disseny de components passius en les bandes d'ona mil·limètrica i Terahercis. Es presenten noves tècniques d'anàlisi i modelatge d'estructures complexes, procediments de disseny, i implementació pràctica de dispositius passius avançats. La primera part de la tesi es focalitza en components passius de THz. Actualment no es disposen de guies d'ona adequades a THz causa que tots dos, metalls i dielèctrics, introdueixen grans pèrdues. En conseqüència, no és adequat escalar les estructures metál·liques tancades usades en microones, ni les guies dielèctriques usades a freqüències òptiques. Entre un gran nombre de propostes recents, la Single Wire Waveguide (SWW) destaca per la seua baixa atenuació i quasi nul·la dispersió. No obstant això, com a guia superficial, la SWW presenta difícil excitació i radiació en corbes. L'ús d'un recobriment dielèctric, creant la Dielecric-Coated Single Wire Waveguide (DCSWW), alleuja aquests inconvenients, però els avantatges anteriors es perden i nous problemes apareixen. Fins a la data, no s'han trobat solucions adequades per a la radiació en corbes de la SWW. A més, es troba a faltar una caracterització rigorosa d'ambdues guies. Aquesta tesi presenta, per primera vegada, un anàlisi modal complet de SWW i DCSWW, adequat a la banda de THz. Aquest anàlisi és aplicat posteriorment per evitar el problema de la radiació en corbes. Es presenten i validen experimentalment diverses estructures i procediments de disseny. La segona part de la tesi es centra en components passius d'ones mil·limètriques. Actualment, aquests components pateixen una important degradació de la seua resposta a causa de que resulta difícil assegurar contacte metàl·lic i alineament adequats per a l'operació a longituds d'ona tan menudes. A més, l'absorció dielèctrica incrementa notablement a aquestes freqüències. En conseqüència, tant guies metàl·liques buides com línies de transmissió planars convencionals presenten gran atenuació, sent necessari considerar topologies alternatives. Les Gap Waveguides (GWs), basades en una estructura periòdica que introdueix un efecte de Electromagnetic Bandgap, resulten molt adequades ja que no requereixen contacte entre parts metàl·liques i eviten les pèrdues en dielèctrics. No obstant, tot i el potencial de les GWs, diverses barreres impedixen la consolidació i ús universal d'aquesta tecnologia. D'una banda, la complexa topologia de les GWs dificulta el procés de disseny atés que les simulacions d'ona completa consumeixen molt de temps i no existeixen actualment mètodes d'anàlisi i disseny apropiats. D'altra banda, és necessari evidenciar el benefici d'utilitzar GWs mitjançant dispositius GW d'altes prestacions i comparatives adequades amb estructures convencionals. Aquesta tesi presenta diversos mètodes d'anàlisi eficients, models, i tècniques de disseny que permetran la síntesi de dispositius GW sense necessitat d'un coneixement profund d'aquesta tecnologia. Així mateix, es presenten diversos filtres de banda estreta operant en les bandes Ka i V amb altes prestacions, així com una comparativa rigorosa amb la guia rectangular.
Berenguer Verdú, AJ. (2017). Analysis and design of efficient passive components for the millimeter-wave and THz bands [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/84004
TESIS

碩士
國立中央大學
資訊工程研究所
94
In reinforcement learning, there is no supervisor to critically judge the chosen action at each step. The learning is through a trial-and-error procedure interacting with a dynamic environment. Q-learning is one popular approach to reinforcement learning. It is widely applied to problems with discrete states and actions and usually implemented by a look-up table where each item corresponds to a combination of a state and an action. However, the look-up table plementation of Q-learning fails in problems with continuous state and action space because an exhaustive enumeration of all state-action pairs is impossible. In this thesis, an implementation of Q-learning for solving problems with continuous state and action space using SOM-based fuzzy systems is proposed. Simulations of training a robot to complete two different tasks are used to demonstrate the effectiveness of the proposed approach. Reinforcement learning usually is a slow process. In order to accelerate the learning procedure, a hybrid approach which integrates the advantages of the ideas of hierarchical learning and the progressive learning to decompose a complex task into simple elementary tasks is proposed.

abstract: Achieving human level intelligence is a long-term goal for many Artificial Intelligence (AI) researchers. Recent developments in combining deep learning and reinforcement learning helped us to move a step forward in achieving this goal. Reinforcement learning using a delayed reward mechanism is an approach to machine intelligence which studies decision making with control and how a decision making agent can learn to act optimally in an environment-unaware conditions. Q-learning is one of the model-free reinforcement directed learning strategies which uses temporal differences to estimate the performances of state-action pairs called Q values. A simple implementation of Q-learning algorithm can be done using a Q table memory to store and update the Q values. However, with an increase in state space data due to a complex environment, and with an increase in possible number of actions an agent can perform, Q table reaches its space limit and would be difficult to scale well. Q-learning with neural networks eliminates the use of Q table by approximating the Q function using neural networks. Autonomous agents need to develop cognitive properties and become self-adaptive to be deployable in any environment. Reinforcement learning with Q-learning have been very efficient in solving such problems. However, embedded systems like space rovers and autonomous robots rarely implement such techniques due to the constraints faced like processing power, chip area, convergence rate and cost of the chip. These problems present a need for a portable, low power, area efficient hardware accelerator to accelerate the process of such learning. This problem is targeted by implementing a hardware schematic architecture for Q-learning using Artificial Neural networks. This architecture exploits the massive parallelism provided by neural network with a dedicated fine grain parallelism provided by a Field Programmable Gate Array (FPGA) thereby processing the Q values at a high throughput. Mars exploration rovers currently use Xilinx-Space-grade FPGA devices for image processing, pyrotechnic operation control and obstacle avoidance. The hardware resource consumption for the architecture has been synthesized considering Xilinx Virtex7 FPGA as the target device.
Dissertation/Thesis
Masters Thesis Engineering 2016

碩士
國立臺灣大學
電信工程學研究所
100
Nowadays, direct-conversion radio frequency (RF) receivers become more appealing due to its cost advantage. However, orthogonal frequency division multiplexing (OFDM) systems with direct-conversion RF receivers are very sensitive to non-idealities at the front-end of receiver, such as I/Q imbalance and carrier frequency offset (CFO). These non-idealities at the receiver result in inter-carrier interference (ICI). Accurate estimates of the non-idealities and channel response are required in OFDM systems. This thesis studies the CFO estimation for OFDM systems with multiple transmit antennas. The thesis consists mainly of two parts. In the first part, we consider OFDM systems with 2 transmit antennas. The Alamouti code is employed at the transmitter and both ST-OFDM and SF-OFDM systems are considered. For these systems with 2 transmit antennas, we extend two known blind CFO estimation algorithms (which were derived for OFDM systems with one transmit antenna) to solve the CFO estimation problem in ST and SF-OFDM systems. IN the second part, we consider OFDM systems with 3 transmit antennas. Orthogonal space time block codes (OSTBC) codes are used to encode the single at the transmitter. For this system, we derive a new blind CFO estimation algorithm. The idea is to exploit the fact that the rate of OSTBC is only 3/4. We show how to exploit the redundant information in OSTBC for blind CFO estimation. Moreover, the problem of I/Q imbalance is also consider in the second part. At the end, we also provide numerical simulation to verify the performance of the proposed methods.

Yes
Coherent subspaces spanned by a finite number of coherent states are introduced, in a quantum system with Hilbert space that has odd prime dimension d. The set of all coherent subspaces is partitioned into equivalence classes, with d 2 subspaces in each class. The corresponding coherent projectors within an equivalence class, have the 'closure under displacements property' and also resolve the identity. Different equivalence classes provide different granularisation of the Hilbert space, and they form a partial order 'coarser' (and 'finer'). In the case of a two-dimensional coherent subspace spanned by two coherent states, the corresponding projector (of rank 2) is different than the sum of the two projectors to the subspaces related to each of the two coherent states. We quantify this with 'non-addditivity operators' which are a measure of quantum interference in phase space, and also of the non-commutativity of the projectors. Generalized Q and P functions of density matrices, which are based on coherent projectors in a given equivalence class, are introduced. Analogues of the Lorenz values and the Gini index (which are popular quantities in mathematical economics) are used here to quantify the inequality in the distribution of the Q function of a quantum state, within the granular structure of the Hilbert space. A comparison is made between Lorenz values and the Gini index for the cases of coarse and also fine granularisation of the Hilbert space. Lorenz values require an ordering of the d 2 values of the Q function of a density matrix, and this leads to the ranking permutation of a density matrix, and to comonotonic density matrices (which have the same ranking permutation). The Lorenz values are a superadditive function and the Gini index is a subadditive function (they are both additive quantities for comonotonic density matrices). Various examples demonstrate these ideas.

碩士
國立臺灣大學
電信工程學研究所
102
Orthogonal frequency division multiplexing (OFDM) has recently received considerable interest for its advantages in high-bit-rate transmissions over frequency-selective fading channels. Space-time block coding (STBC) has emerged as a powerful approach to exploit spatial diversity and to combat fading in multiple-input multiple-output (MIMO) wireless communication systems. However, OFDM systems with direct-conversion RF receivers are very sensitive to non-idealities at the front-end of receiver, such as I/Q imbalance and carrier frequency o&;#8629;set (CFO). These non-idealities at the receiver result in intercarrier interference (ICI). Accurate estimates of the non-idealities and channel response are required in OFDM systems. This thesis studies the blind CFO and I/Q imbalance estimation for OFDM systems with multiple transmit antennas. The thesis consists of three parts. In the first part, we review an existing CFO and I/Q imbalance estimation method based on rank criterion. This method though gives good performance but not for joint estimation, it cannot estimate the I/Q imbalance in the presense of CFO. In order to solve this issue, we propose an improved method. In the second part, we propose a new rank criteron and group method for joint CFO and independent I/Q imbalance estimation. The proposed method exploits the redundant information in OSTBC for joint blind CFO and I/Q imbalance estimation. In the third part, we extend the proposed method to the estimation of frequency dependent I/Q imbalance in the presence of CFO. We also provide numerical simulation to verify the performance of the proposed methods.

Dissertação de mestrado integrado em Engenharia Eletrónica Industrial e Computadores
A Qualidade de Energia Elétrica é um assunto cada vez mais importante devido ao crescente número de problemas e custos relacionados com a diminuição dos níveis de Qualidade de Energia Elétrica. Para compensar os atuais problemas de Qualidade de Energia Elétrica foram desenvolvidos os filtros ativos de potência, que possuem uma resposta temporal rápida, sendo capazes de se adaptar dinamicamente às condições do sistema que estão a compensar. Nesta Dissertação de Mestrado são apresentadas as principais topologias de filtros ativos de potência, sendo que a utilizada foi o Filtro Ativo Paralelo. Esta é considerada uma das mais importantes, por ser capaz de efetuar compensações a nível das correntes. Para a implementação deste filtro ativo de potência foi necessário definir o tipo de inversor, a teoria de controlo e a técnica de modulação a serem utilizadas. Tendo isto em conta, foi efetuado o levantamento do estado da arte relativo aos inversores utilizados em filtros ativos de potência, assim como das principais teorias de controlo e técnicas de modulação aplicáveis no âmbito desta Dissertação. Após isto, foi selecionado o Inversor Tipo Fonte de Corrente (Current-Source Inverter - CSI). A teoria de controlo e a técnica de modulação escolhidas foram, respetivamente, a Teoria p-q e a Modulação Vetorial (Space Vector Modulation - SVM). O sistema de controlo e os circuitos de potência foram simulados no software PSIM para que seja possível validar a sua correta implementação e funcionamento. Uma vez obtida essa validação, os circuitos e sistemas de controlo foram implementados em laboratório, para verificar o seu funcionamento. Finalmente, é de referir que com esta Dissertação pretende-se contribuir para a investigação da utilização de Inversores Tipo Fonte de Corrente em Filtros Ativos Paralelos.
Power Quality is a subject with ever increasing importance due to the rising numbers of problems and costs related with the reduction of the Power Quality levels. To compensate the Power Quality problems were developed active power filters that have a quick time response, and that are capable of adapting to the conditions of the system they are compensating. In this Master Thesis are presented the main topologies of active power filters, with the Shunt Active Power Filter being implemented. This topology is considered to be one of the most important active power filter topologies, because it is capable of performing current compensation. For the implementation of this active power filter, it was necessary to define the type of inverter, the control strategy and the modulation technique to be used. Regarding this, a research of the state of the art of inverters, control strategies and modulation techniques used in Shunt Active Filters was performed. After this, the Current-Source Inverter (CSI) was selected. The control strategy and the modulation technique implemented were the p-q Theory and the Space Vector Modulation (SVM), respectively. The control and power circuits were simulated using the software PSIM, so that their operation could be validated. Once this validation was obtained, these systems were implemented in laboratory, to assess their operation capabilities. Finally, it must be said that with this work it is intended to contribute to the research of the use of Current Source Inverters in Shunt Active Power Filters.

This work focuses on high-capacity spectral-efficient Earth-space microwave communication links, supporting the access to wider and less congested bandwidths and providing the required technological advances on radiowave propagation for the design and operation of system-technologies improving the spectral efficiency, proposing the operation of frequency-reuse schemes and polarization diversity techniques at Ka- and Q/V-bands. Because the suitable planning, successful deployment and effective, spectralefficient and interference-free, operation of all space-borne systems and services is necessarily related and directly dependent, although not exclusively, on the propagation channel conditions, this work provides a comprehensive and systematic approach intending the relevant propagation phenomena characterization and understanding, modelling and mitigation. It is proved that there is a bigger operational margin for the operation of the abovementioned system-technologies than what may be believed by considering the international recommended models. Major challenges are envisaged, specially in what concerns the ice-induced depolarization that is more persistent and takes longer than the corresponding rain-induced effects. Nevertheless, their mitigation is possible for which the depolarization shall be the controlling variable. The time lag of 15 minutes is identified as the best trade-off between affordable time and achievable gain in the framework of a time diversity scheme.
Este trabalho foca-se em links de comunicação Terra-espaço em microondas espectralmente eficientes e de elevada capacidade, suportando o acesso a larguras de banda maiores e menos congestionadas e ofereçendo os avanços tecnológicos necessários em termos de propagação de ondas de rádio para o projeto e operação de tecnologias de sistema que visam o uso otimizado do espectro, propondo a operação de esquemas de reuso de frequência e de técnicas de diversidade de polarização nas bandas Ka e Q/V. Porque o adequado planeamento, a implementação bem sucedida, bem como a efetiva operação, espectralmente eficiente e livre de interferências, de todos os sistemas e serviços espaciais está necessariamente relacionada e diretamente dependente, ainda que não exclusivamente, das condições do canal de propagação, este trabalho oferece uma abordagem completa e sistemática pretendendo a caracterização, compreensão, modelação e mitigação dos fenómenos de propagação relevantes. É provado existir uma maior margem operacional para a operação das tecnologias acima mencionadas do que aquela que se acredita existir considerando os modelos internacionais recomendados. São previstos maiores desafios, especialmente no que concerne à despolarização causada por gelo que é mais persistente e que dura mais tempo do que os correspondentes efeitos causados por chuva. De qualquer forma, a sua mitigação é possível e para a qual a despolarização deverá ser a variável de controlo. O intervalo de tempo de 15 minutos é identificado como o melhor compromisso entre tempo acessível e ganho alcançável no contexto de um esquema de diversidade temporal.
Programa Doutoral em Engenharia Eletrotécnica

Dissertação de mestrado em Engenharia Eletrónica Industrial e de Computadores
Atualmente, a utilização de cargas não lineares é cada vez mais frequente. Estas, originam problemas de qualidade de energia elétrica (QEE), resultando em várias consequências, como falhas em equipamentos eletrónicos, podendo causar prejuízos elevados em diversos setores. Os filtros ativos de potência, que também podem fazer interface com fontes de energia renovável, apresentam-se como uma boa solução para mitigar estes problemas. Nesta dissertação de mestrado apresentam-se as topologias de filtros ativos de potência, tendo sido simulada uma topologia de Filtro Ativo Paralelo do Tipo Fonte de Corrente. Nesta simulação, para além do inversor fonte de corrente (Current-Source Inverter - CSI), implementou-se um algoritmo baseado na Teoria p-q, bem como a técnica de modulação vetorial (Space-Vector Modulation - SVM). Estas técnicas foram as escolhidas para efetuar o controlo e a modulação do Filtro Ativo Paralelo do Tipo Fonte de Corrente. Nesta dissertação é ainda efetuado um estudo sobre o estado da arte dos conversores CC-CC utilizados em sistemas solares fotovoltaicos, e sobre os sistemas de controlo aplicáveis aos mesmos. Tendo em conta as características do sistema, foi selecionado o conversor CC-CC push-pull, sendo o mesmo simulado e implementado. Nesta dissertação é realizado o dimensionamento e a construção do transformador de alta-frequência deste conversor CC-CC. São ainda apresentadas as placas do andar de potência e o sistema de controlo, desenvolvido no DSP TMS320F28335 da Texas Instruments. A presente dissertação de mestrado apresenta também os testes realizados ao transformador de alta-frequência e ao conversor CC-CC push-pull a operar isoladamente. São ainda apresentados resultados experimentais para a operação de todas as placas desenvolvidas.
Nowadays, the use of no-linear loads in industry and by general consumers is increasingly common. This situation causes Power Quality problems, resulting in various consequences, such as flaws in electronic equipment, which may cause revenue losses in several sectors. Besides that, Power Quality problems contributes to the increase of efficiency losses in domestic and industrial equipment, and also in the electrical installations. The Active Power Filters, which can be incorporated with Renewable Energy solutions, are a good solution to mitigate these problems. In this Master Thesis are presented the main topologies of Active Power Filters, with Current-Source Shunt Active Power Filter being simulated. In this simulation, in addition to the Current-Source Inverter, an algorithm based on the p-q Theory and the Space Vector Modulation technique was implemented. These techniques were chosen to control and modulate the Current-Source Shunt Active Power Filter. In this work, it is also made a study about the state of the art of DC-DC Converters used in Photovoltaic Solar Systems, and the control systems that can be applied to them. Considering the characteristics of the system, the DC-DC Push Pull Converter has been selected, where it is simulated and implemented. In this Master Thesis the design and construction of the high-frequency transformer of this converter was developed. The power stage boards and the control system, implemented in the Texas Instruments microcontroller DSP TMS320F28335, are also developed and presented. This work also presents the tests performed for the developed high-frequency transformer and DC-DC Push Pull Converter. Experimental results of all developed circuits are presented.