Dans ce mémoire, je démontre que la distribution de probabilités de l'état quantique Greenberger-Horne-Zeilinger (GHZ) sous l'action locale de mesures de von Neumann indépendantes sur chaque qubit suit une distribution qui est une combinaison convexe de deux distributions. Les coefficients de la combinaison sont reliés aux parties équatoriales des mesures et les distributions associées à ces coefficients sont reliées aux parties réelles des mesures. Une application possible du résultat est qu'il permet de scinder en deux la simulation de l'état GHZ. Simuler, en pire cas ou en moyenne, un état quantique comme GHZ avec des ressources aléatoires, partagées ou privées, et des ressources classiques de communication, ou même des ressources fantaisistes comme les boîtes non locales, est un problème important en complexité de la communication quantique. On peut penser à ce problème de simulation comme un problème où plusieurs personnes obtiennent chacune une mesure de von Neumann à appliquer sur le sous-système de l'état GHZ qu'il partage avec les autres personnes. Chaque personne ne connaît que les données décrivant sa mesure et d'aucune façon une personne ne connaît les données décrivant la mesure d'une autre personne. Chaque personne obtient un résultat aléatoire classique. La distribution conjointe de ces résultats aléatoires classiques suit la distribution de probabilités trouvée dans ce mémoire. Le but est de simuler classiquement la distribution de probabilités de l'état GHZ. Mon résultat indique une marche à suivre qui consiste d'abord à simuler les parties équatoriales des mesures pour pouvoir ensuite savoir laquelle des distributions associées aux parties réelles des mesures il faut simuler. D'autres chercheurs ont trouvé comment simuler les parties équatoriales des mesures de von Neumann avec de la communication classique dans le cas de 3 personnes, mais la simulation des parties réelles résiste encore et toujours. / In this Master's thesis, I show that the probability distribution of the Greenberger-Horne-Zeilinger quantum state (GHZ) under local action of independent von Neumann measurements follows a convex distribution of two distributions.The coefficients of the combination are related to the equatorial parts of the measurements, and the distributions associated with those coefficients are associated with the real parts of the measurements. One possible application of my result is that it allows one to split into two pieces the simulation of the GHZ state. Simulating, in worst case or in average, a quantum state like the GHZ state with random resources, shared or private, as well as with classical communication resources or even odd resources like nonlocal boxes is a very important in the theory of quantum communication complexity. We can think of this simulation problem as a problem in which many people get the description of a von Neumann measurement. Each party does not know the description of any other measurements belonging to the other parties. Each party after having applied his measurement on the subsystem of the state that he shares with the others gets a classical outcome. The joint distribution of the outcomes of every parties follows the distribution studied in this thesis in the case of the GHZ state. My result indicates that in order to simulate the distribution, we can first simulate the equatorial parts of the measurements in order to know which distribution associated to the real parts of the measurements to simulate. Other researchers have found how to simulate the equatorial parts of the von Neumann measurements with classical resources in the case of 3 parties, but it is still unknown how to simulate the real parts.

Au cours des dernières années, différents travaux de recherche ont été consacrés à l’étude et au développement des solutions de type réseau de capteurs sans fil. Ces travaux sont une réponse à l’augmentation du nombre d’objets connectés dans le monde avec le développement de l’internet des objets. La consommation d’énergie dans les réseaux de capteurs représente un des domaines les plus étudiés. Les communications dans les réseaux de capteurs représentent une part importante de leur consommation. Afin de réduire la consommation des communications dans les réseaux de capteurs, différents niveaux d’optimisation sont possibles. Dans ce contexte, différents travaux de recherches visant à réduire la consommation des émetteurs-récepteurs, grâce à des architectures innovantes, ont été menés à l’IM2NP. Beaucoup de ces travaux ont été consacrés aux radios impulsionnelles Ultra-Large Bande (ULB). En outre, les protocoles d’accès aux canaux de communications dans les réseaux de capteurs sont également importants quant à l’optimisation de leurs consommations. Les travaux de recherches proposés dans ce manuscrit, basés sur les travaux sur les radios impulsionnelles ULB réalisés au sein de l’IM2NP, proposent une optimisation de la consommation des réseaux de capteurs sur deux niveaux. Tout d’abord, un protocole asynchrone à base de radio de réveil d’accès au canal de communication adapté aux communications ULB est proposé ainsi que son étude énergétique. Suite à l'étude du protocole asynchrone proposé, la conception d’un récepteur de réveil semi-passif et d’un générateur d’impulsion ULB sous-GHz est abordée, et leurs performances en termes d’efficacité énergétique discutées. / In recent years, research has been devoted to the study and development of Wireless Sensor Network (WSN). These research are a response to the increase of the connected objet number in the world with development of smartphones and Internet of things (IoT). Energy consumption in sensor networks is one of the most studied areas. In fact, the optimization of the consumption of the elements making up the sensor networks allows a reduction in the costs associated with their installation, operation and maintenance. Communications in sensor networks represent an important part of their power consumption. In order to reduce the consumption of communications in sensor networks, different levels of optimization are possible. In this context, various IM2NP research projects aimed at reducing transceiver consumption through innovative design were carried out. Much of this work has been devoted to Impulsionnal Radio Ultra-Wide Band (IR-UWB). Moreover, MAC protocols used to manage the communication channel access in wireless sensor networks are also important in optimizing their consumption. The research work proposed in this manuscript, based on the work on impulse radios UWB carried out within the IM2NP, propose an optimization of the consumption of sensor networks on two levels. Firstly, an asynchronous MAC protocol based on radio wake-up radio suitable for UWB communications is proposed as well as its energy study. Following the study of the proposed asynchronous protocol, the design of a low power semi-passive wake-up receiver (WuRx) and a UWB sub-GHz impulse generator is presented and their performance in terms of energy efficiency discussed.

Dans ce mémoire, je démontre que la distribution de probabilités de l'état quantique Greenberger-Horne-Zeilinger (GHZ) sous l'action locale de mesures de von Neumann indépendantes sur chaque qubit suit une distribution qui est une combinaison convexe de deux distributions. Les coefficients de la combinaison sont reliés aux parties équatoriales des mesures et les distributions associées à ces coefficients sont reliées aux parties réelles des mesures. Une application possible du résultat est qu'il permet de scinder en deux la simulation de l'état GHZ. Simuler, en pire cas ou en moyenne, un état quantique comme GHZ avec des ressources aléatoires, partagées ou privées, et des ressources classiques de communication, ou même des ressources fantaisistes comme les boîtes non locales, est un problème important en complexité de la communication quantique. On peut penser à ce problème de simulation comme un problème où plusieurs personnes obtiennent chacune une mesure de von Neumann à appliquer sur le sous-système de l'état GHZ qu'il partage avec les autres personnes. Chaque personne ne connaît que les données décrivant sa mesure et d'aucune façon une personne ne connaît les données décrivant la mesure d'une autre personne. Chaque personne obtient un résultat aléatoire classique. La distribution conjointe de ces résultats aléatoires classiques suit la distribution de probabilités trouvée dans ce mémoire. Le but est de simuler classiquement la distribution de probabilités de l'état GHZ. Mon résultat indique une marche à suivre qui consiste d'abord à simuler les parties équatoriales des mesures pour pouvoir ensuite savoir laquelle des distributions associées aux parties réelles des mesures il faut simuler. D'autres chercheurs ont trouvé comment simuler les parties équatoriales des mesures de von Neumann avec de la communication classique dans le cas de 3 personnes, mais la simulation des parties réelles résiste encore et toujours. / In this Master's thesis, I show that the probability distribution of the Greenberger-Horne-Zeilinger quantum state (GHZ) under local action of independent von Neumann measurements follows a convex distribution of two distributions.The coefficients of the combination are related to the equatorial parts of the measurements, and the distributions associated with those coefficients are associated with the real parts of the measurements. One possible application of my result is that it allows one to split into two pieces the simulation of the GHZ state. Simulating, in worst case or in average, a quantum state like the GHZ state with random resources, shared or private, as well as with classical communication resources or even odd resources like nonlocal boxes is a very important in the theory of quantum communication complexity. We can think of this simulation problem as a problem in which many people get the description of a von Neumann measurement. Each party does not know the description of any other measurements belonging to the other parties. Each party after having applied his measurement on the subsystem of the state that he shares with the others gets a classical outcome. The joint distribution of the outcomes of every parties follows the distribution studied in this thesis in the case of the GHZ state. My result indicates that in order to simulate the distribution, we can first simulate the equatorial parts of the measurements in order to know which distribution associated to the real parts of the measurements to simulate. Other researchers have found how to simulate the equatorial parts of the von Neumann measurements with classical resources in the case of 3 parties, but it is still unknown how to simulate the real parts.

Made available in DSpace on 2016-03-15T19:37:45Z (GMT). No. of bitstreams: 1 Heidi Kaori Sato Pertile.pdf: 2322047 bytes, checksum: aaee9a054b94e9557394faaf3251bec9 (MD5) Previous issue date: 2013-01-24 / In this work we present a study on the generation of pulse train in an Erbium doped fiber laser in the hybrid mode-locking regime operating with short pulses at high repetition rates. The short pulses are generated by passive mode-locking technique using carbon nanotubes as saturable absorbers. High repetition rates are generated by active mode-locking technique using a phase modulator. We built cavities with three different mode-locking regimes: active, passive and, finally, hybrid, to compare results. In active and hybrid cavities we used an electro-optical modulator. In passive and hybrid cavities we used a homemade film of a polymer containing carbon nanotubes with diameter of 1 nm. With the cavity operating in the hybrid regime we obtained pulse durations of 1.77 ps with repetition rate of 10 GHz. / Neste trabalho apresentamos um estudo sobre a geração de trem de pulsos em laser à fibra dopada com Érbio operando em regime de acoplamento híbrido de modos, com pulsos curtos a altas taxas de repetição. Os pulsos curtos são obtidos pela técnica de acoplamento passivo de modos utilizando absorvedores saturáveis de carbono. As altas taxas de repetição são obtidas pela técnica de acoplamento ativo de modos através de um modulador. Construímos três cavidades distintas: ativa, passiva e finalmente a híbrida para comparação de resultados. Nas cavidades ativa e híbrida foi utilizado um modulador eletro-óptico de fase e, nas cavidades passiva e híbrida foi utilizado um filme de um polímero (NOA 73TM) contendo nanotubos de carbono com diâmetro de 1 nm por nós fabricado. Com a cavidade em regime híbrido de modos, foi obtida uma duração de pulso de 1,77 ps com uma taxa de repetição de 10 GHz.

In this master´s thesis an issue of broadband planar antennas was analyzed. Firstly, the basic elements that affect bandwidth, such as the influence of the dieletric substrate, suitable shape or feeding network, were described. Furthermore technics that can be used to widen the band of the planar antennas were described. These technics were applied to a chosen type of a planar antenna. This antenna was modeled and simulated for desired dielectric substrate in CST STUDIO SUITE 2010 software. Feeding network with power dividers was designed for the antenna array. In order to obtain a higher gain, the antenna was implemented into the 2x2 element array. The results of the simulation are evaluated in the conclusion.

The Master´s thesis deals with the theoretical design, practical numerical analyses and optimalization of disc-like antenna and disc-like slot antenna. The antennas have relative omnidirectional radiation characteristics in the horizontal plane, respectively in the vertical plane. Antennas exploit the substrate integrated waveguide technology. The antennas themselves are designed for the purpose of making the antenna array that will be work as an antenna unit for access point. The antennas are manufactured and experimentally verified. The program CST Microwave Studio is used for simulation in the thesis.

The thesis is aimed to design textile antennas that can operate in the ISM band 5.8 GHz. Attention is turned to practical exploitation of these antennas under regular conditions. Variations of the bandwidth of the antenna due to washing and waterproofing will be observed. Finally, the impact of the proximity of living tissue on the parameters of antennas will be studied.

Avec le contexte actuel des défis sécuritaires, la détection de petits objets devient un enjeu majeur pour lutter contre les actes malveillants. Mais les évolutions des technologies en bandes millimétriques et le potentiel de ces bandes, notamment autour de 60 GHz peuvent faciliter la conception de systèmes de plus en plus performants, permettant de répondre à ces enjeux. Cette thèse s’inscrit dans ce contexte pour proposer un système de détection ultra-large bande (ULB) millimétrique pour des applications de courte portée. Après un état de l'art détaillé des fondamentaux de la détection, nous présentons une validation de l'estimation de la surface équivalente radar (SER) par simulation avec le logiciel HFSS et par mesures en chambre anéchoïque. Ces deux études nous ont permis d'identifier les éléments essentiels à l'analyse système et les paramètres critiques de la détection d'objets canoniques cylindres et plaques. Une fois l'analyse système cernée, nous avons proposé une approche de détection multi-bande basée sur le radar monostatique pour améliorer la couverture de détection des systèmes, mais aussi et surtout pour pallier la limitation de la détection des objets en fonction de leur orientation par rapport à l'axe de viser de l'antenne. Nous avons également proposé une architecture (émetteur-récepteur) simple pouvant être associée au principe de détection. Le dimensionnement du système nous a permis d'identifier les amplificateurs faible bruit (LNAs) comme éléments critiques du récepteur et ainsi d'établir leurs spécifications pour atteindre les performances visées. Une comparaison des technologies SG13S de IHP et D007IH de OMMIC est menée dans la dernière partie de ce travail. Le choix de la technologie est justifié et la conception des LNAs sous ADS Keysight est détaillée. Une simulation de l'ensemble du système basée sur les performances des LNAs conçus et sur les simulations de SER est présentée pour illustrer la mise en œuvre de la détection. Enfin, les performances de la détection sont évaluées pour des cibles cylindriques et les apports du système proposé sont illustrés en comparaison avec une détection classique mono-bande / With the current context of security challenges, small objects detection is becoming a major issue in the fight against malicious acts. However, developments in millimeter-band technologies and the potential of these bands, particularly around 60 GHz, can facilitate the design of increasingly efficient systems to meet these challenges. This thesis is part of this context to propose an ultra-wideband (UWB) millimeter-wave detection system for short-range applications.After a detailed state-of-the-art of the fundamentals of detection, we present a validation of the radar cross-section (RCS) estimation by simulations with HFSS software and by measurements in anechoic chamber. These two studies allowed us to identify the essential elements for system analysis and the critical parameters for the detection of canonical objects such as cylinders and plates.Once the system analysis was identified, we proposed a multi-band detection approach based on monostatic radar to improve the detection coverage of the systems, but also and especially to overcome the limitation of objects detection according to their orientation in relation to the antenna boresight. We also proposed a simple architecture (transceiver) that could be associated with the detection principle. The system design allowed us to identify low noise amplifiers (LNAs) as critical elements of the receiver and thus establish their specifications to achieve the targeted performances.A comparison of IHP's SG13S and OMMIC's D007IH technologies is conducted in the last part of this work. The choice of technology is justified and the design of LNAs under ADS Keysight is detailed. A simulation of the entire system based on the performance of the designed LNAs and RCS simulations is presented to illustrate the implementation of detection. Finally, detection performances are evaluated for cylindrical targets and the contributions of the proposed system are illustrated in comparison with conventional single-band detection

[pt] A perda de percurso (PL) é um parâmetro essencial em modelos de propagação e crucial na determinação da área de cobertura de sistemas móveis. Os métodos de aprendizado de máquina (ML) tornaram-se ferramentas promissoras para a previsão de propagação de rádio. No entanto, ainda existem alguns desafios para sua implantação completa, relacionados à seleção das entradas mais significativas do modelo, à compreensão de suas contribuições para as previsões do modelo e à avaliação adicional da capacidade de generalização para amostras desconhecidas. Esta tese tem como objetivo projetar modelos de PL baseados em ML otimizados para diferentes aplicações das tecnologias 5G e além. Essas aplicações abrangem links de ondas milimétricas (mmWave) para ambientes indoor e outdoor na faixa de frequência de 26,5 a 40 GHz, cobertura de macrocélulas no espectro sub-6 GHz e comunicações veiculares usando campanhas de medições desenvolvidas em CETUC, Rio de Janeiro, Brazil. Vários algoritmos de ML são explorados, como redes neurais artificiais (ANN), regressão de vetor de suporte (SVR), floresta aleatória (RF) e aumento de árvore de gradiente (GTB). Além disso, estendemos dois modelos empíricos para mmWave com previsão de PL melhorada. Propomos uma metodologia para seleção robusta de modelos de ML e uma metodologia para selecionar os preditores mais adequados para as máquinas consideradas com base na melhoria de desempenho e na interpretabilidade do modelo. Além disso, para o canal veículo-veículo (V2V), uma técnica de rede neural convolucional (CNN) também é proposta usando uma abordagem de aprendizado por transferência para lidar com conjuntos de dados pequenos. Os testes de generalização propostos mostram a capacidade dos modelos de ML de aprender o padrão entre as entradas do modelo e a PL, mesmo em ambientes e cenários mais desafiadores de amostras desconhecidas. / [en] Path loss (PL) is an essential parameter in propagation models and critical in determining mobile systems’ coverage area. Machine learning (ML) methods have become promising tools for radio propagation prediction. However, there are still some challenges for its full deployment, concerning to selection of the most significant model s inputs, understanding their contributions to the model s predictions, and a further evaluation of the generalization capacity for unknown samples. This thesis aims to design optimized ML-based PL models for different applications for the 5G and beyond technologies. These applications encompass millimeter wave (mmWave) links for indoor and outdoor environments in the frequency band from 26.5 to 40 GHz, macrocell coverage in the sub-6 GHz spectrum, and vehicular communications using measurements campaign carried out by the Laboratory of Radio-propagation, CETUC, in Rio de Janeiro, Brazil. Several ML algorithms are exploited, such as artificial neural network (ANN), support vector regression (SVR), random forest (RF), and gradient tree boosting (GTB). Furthermore, we have extended two empirical models for mmWave with improved PL prediction. We proposes a methodology for robust ML model selection and a methodology to select the most suitable predictors for the machines considered based on performance improvement and the model’s interpretability. In adittion, for the vehicle-to-vehicle (V2V) channel, a convolutional neural network (CNN) technique is also proposed using a transfer learning approach to deal with small datasets. The generalization tests proposed shows the ability of the ML models to learn the pattern between the model’s inputs and PL, even in more challenging environments and scenarios of unknown samples.

[pt] A perda de percurso (PL) é um parâmetro essencial em modelos de propagação e crucial na determinação da área de cobertura de sistemas móveis. Osmétodos de aprendizado de máquina (ML) tornaram-se ferramentas promissoras para a previsão de propagação de rádio. No entanto, ainda existem algunsdesafios para sua implantação completa, relacionados à seleção das entradasmais significativas do modelo, à compreensão de suas contribuições para asprevisões do modelo e à avaliação adicional da capacidade de generalizaçãopara amostras desconhecidas. Esta tese tem como objetivo projetar modelosde PL baseados em ML otimizados para diferentes aplicações das tecnologias5G e além. Essas aplicações abrangem links de ondas milimétricas (mmWave)para ambientes indoor e outdoor na faixa de frequência de 26,5 a 40 GHz,cobertura de macrocélulas no espectro sub-6 GHz e comunicações veicularesusando campanhas de medições desenvolvidas em CETUC, Rio de Janeiro,Brazil. Vários algoritmos de ML são explorados, como redes neurais artificiais(ANN), regressão de vetor de suporte (SVR), floresta aleatória (RF) e aumentode árvore de gradiente (GTB). Além disso, estendemos dois modelos empíricospara mmWave com previsão de PL melhorada. Propomos uma metodologiapara seleção robusta de modelos de ML e uma metodologia para selecionar ospreditores mais adequados para as máquinas consideradas com base na melhoria de desempenho e na interpretabilidade do modelo. Além disso, para o canalveículo-veículo (V2V), uma técnica de rede neural convolucional (CNN) também é proposta usando uma abordagem de aprendizado por transferência paralidar com conjuntos de dados pequenos. Os testes de generalização propostosmostram a capacidade dos modelos de ML de aprender o padrão entre as entradas do modelo e a PL, mesmo em ambientes e cenários mais desafiadoresde amostras desconhecidas. / [en] Path loss (PL) is an essential parameter in propagation models and critical in determining mobile systems coverage area. Machine learning (ML) methods have become promising tools for radio propagation prediction. However, there are still some challenges for its full deployment, concerning to selection of the most significant model s inputs, understanding their contributions to the model s predictions, and a further evaluation of the generalization capacity for unknown samples. This thesis aims to design optimized ML-based PL models for different applications for the 5G and beyond technologies. These applications encompass millimeter wave (mmWave) links for indoor and outdoor environments in the frequency band from 26.5 to 40 GHz, macrocell coverage in the sub-6 GHz spectrum, and vehicular communications using measurements campaign carried out by the Laboratory of Radio-propagation, CETUC, in Rio de Janeiro, Brazil. Several ML algorithms are exploited, such as artificial neural network (ANN), support vector regression (SVR), random forest (RF), and gradient tree boosting (GTB). Furthermore, we have extended two empirical models for mmWave with improved PL prediction. We proposes a methodology for robust ML model selection and a methodology to select the most suitable predictors for the machines considered based on performance improvement and the model s interpretability. In adittion, for the vehicle-to-vehicle (V2V) channel, a convolutional neural network (CNN) technique is also proposed using a transfer learning approach to deal with small datasets. The generalization tests proposed shows the ability of the ML models to learn the pattern between the model’s inputs and PL, even in more challenging environments and scenarios of unknown samples.