Anorexia nervosa (AN) is a serious disorder with long-term consequences for those afflicted. No evidence-based care is available for adults with full or subthreshold AN. The thesis research investigated aspects of emotion and cognition relevant to the maintenance of AN that might inform psychological treatment. In addition, the effectiveness of a recent psychotherapy model of AN was investigated. Study I investigated alexithymia and emotional awareness and their associations with depression, anxiety, and perfectionism among patients with AN compared with a control group. The AN group exhibited the same level of emotional awareness as did the control group and the same level of alexithymia when controlling for depression and anxiety. Alexithymia and emotional awareness were not associated, despite representing an overlapping construct. The results of the present study indicate that those with AN can trust their emotional awareness. Study II explored implicit pro-thin and anti-fat attitudes (towards the self and others), striving for thinness (loosely corresponding to positive reinforcement), and avoidance of fatness (loosely corresponding to negative reinforcement). The AN and the control groups were found to have equally strong implicit pro-thin and striving for thinness attitudes. The AN group exhibited stronger implicit anti-fat and avoidance of fatness attitudes (loosely corresponding to negative reinforcement) than did the control group. There was no association between implicit and explicit measures. The results are in line with the over-evaluation of weight and shape as a core feature of eating disorders. Study III compared the effectiveness of Acceptance and Commitment Therapy (ACT) and treatment as usual (TAU) for adults with AN after day-care. Follow-up measures indicated no difference in improvement or deterioration between the two groups. The level of perfectionism was reduced in the ACT group relative to the TAU group. The study was compromised by a lower inflow of patients than anticipated and by a high drop-out rate, and thus fails to provide evidence of a difference between the two groups. The present thesis demonstrates that emotional awareness is intact in those with AN and that implicit attitudes concerning weight and shape reflect the explicit attitudes, although without association. The treatment study indicates that, when designing treatment, it is important to consider the ambivalence to treatment among those suffering from AN, which is reflected in the high drop-out rate in the present study.

Le transistor à effet tunnel bande à bande (TFET) est une architecture PIN à grille capable d’obtenir une pente sous le seuil inférieure à 60mV/dec à température ambiante, ce qui représente un avantage par rapport au MOSFET dans le cas d’applications basse consommation. L’objectif de cette thèse est d’étudier et de caractériser des TFETs fabriqués au CEA-LETI (sur substrats SOI avec les procédés standards CMOS), afin de comprendre et d’optimiser ces dispositifs. La première génération de TFETs a été réalisée en architecture planaire (FDSOI) et fournit une étude sur l’impact de l’hétérojonction canal source, de l’épaisseur du canal et de la température de recuit sur les performances. La seconde génération a été réalisée en architecture nanofil SiGe planaire, dont l’impact de la géométrie a été étudié en détail. Les mesures ont permis de valider l’injection par effet tunnel bande à bande, et les performances observées ont été comparées à la littérature et aux MOSFET. Par ailleurs, des caractérisations avancées ont également mené à une meilleure compréhension des caractéristiques de sortie courant-tension. Finalement, des mesures basse température nous avons confirmé la présence de défauts proches des jonctions (à l’origine des limitations de pente sous le seuil) et ainsi proposé des voies d’optimisation pour s’en affranchir. / Band to band tunneling field effect transistor (TFET) is a PIN-gated architecture able to reach sub 60mV/dec subthreshold slopes at room temperature, which is an advantage over MOSFET in low power applications. The objective of this thesis is to study and characterize TFETs fabricated in CEA-LETI using MOSFET SOI technology. The first generation of devices is realized on planar FDSOI technology, and studies the impact of source/channel heterojunction, channel thickness and annealing temperature on device performances. The second generation is planar SiGe nanowire architecture, with research focusing on the impact of the wire geometry. Through measurements we were able to prove the band to band tunneling injection, while the reported performances were compared with literature and with MOSFET. Furthermore, advanced characterizations led to a better understanding of the output characteristics. Through low temperature measurements we confirmed existence of defects close to the junctions (which cause slope degradation), as well as on which process steps to improve in the future.

Composite cascode stages have been used in operational amplifier designs to achieve ultra-high gain at very low power. The flexibility and simplicity of the stage makes it an appealing choice for low power op-amp designs. Op-amp design using the composite cascode stage is often made more difficult through the lack of a design process. A design process to aid in the selection of the MOSFET dimensions is provided in this thesis. This process includes a table-based method for selection of the widths and lengths of the MOSFETs used in the composite cascode stage. Equations are also derived for the gain, bandwidth, and noise of the composite cascode stage with each of the devices operating in the various regions of inversion.

No description available.

Logarithmic sensors are capable of capturing the wide dynamic range of intensities available in nature with minimum number of bits and post-processing required. A simple circuit able to perform logarithmic capture is one utilising a MOS device in weak inversion. However, the output of this pixel is crippled due to fixed pattern noise. Technique proposed to reduce this noise fail to produce high quality images on account of unaccounted high gain variations in the pixel. An electronic calibration technique is proposed which is capable of reducing both multiplicative as well as additive FPN. Contrast properties matching that of human eye are reported from these sensors. With reduced FPN, the pixel performance at low intensities becomes concerning. In these regions, the high leakage current of the CMOS process affects the logarithmic pixel. To reduce this current, two different techniques using a modified circuit and another with modified layout are tested. The layout technique is observed to reduce the leakage current. In addition, this layout can be used to linearise the output of logarithmic pixel in low light regions. The unique linear response at low light and logarithmic pixel at high light is further investigated. A new model based on the device physics is derived to represent this response. The fixed pattern noise profile is also investigated. An intelligent iterative scheme is proposed and verified to extract the photocurrent flowing in the pixel and correct the fixed pattern noise utilising the new model. Future research ideas leading to better designs of logarithmic pixels and post-processing of these signals are proposed at the end of the thesis.

Le secteur des communications sans fil a connu un essor considérable, soutenu par l’évolution des "smartphones" et par le développement des réseaux de capteurs sans fil et de l’Internet des Objets (connu en anglais sous le nom ‘IoT’ pour Internet of Things). Les applications actuelles visent l’autonomie énergétique des objets communicants et nécessitent la conception de circuits intégrés pouvant assurer à la fois un fonctionnement à hautes performances et à moindre coût. L’une des principales fonctions des systèmes de communications radiofréquences (RF) est la génération de fréquence, assurée par l’oscillateur. De nombreux efforts de conception sont ainsi nécessaires afin d’assurer les performances requises par les nouvelles applications sans fil. Nos travaux de recherche ont pour objectif de proposer une méthode de conception d’oscillateurs agiles à faible consommation au sein des systèmes d’émission-réception RF. Le travail s’est focalisé sur l’étude et l’optimisation des éléments constitutifs d’un oscillateur LC passif en technologie CMOS et sur la proposition d’une méthode de conception. La méthode proposée peut être exploitée pour différentes structures d’oscillateurs afin d’optimiser leurs performances essentiellement en termes de consommation de puissance et de bruit de phase. Cette méthode a été appliquée pour implémenter plusieurs VCOs en technologie CMOS. Une série de mesure sous pointes a permis de valider leur fonctionnement. La suite de ce travail de thèse est consacrée à la proposition d’une nouvelle topologie d’oscillateur LC reconfigurable à base d’inductance active dédiée aux applications multistandards faible coût / The rapid growth of the Internet of Things (IoT) applications and the wireless sensor networks boosts the need for low cost and low power radiofrequency (RF) transceivers. The voltage-controlled oscillator (VCO) is an essential building block of several RF transceivers. Design tradeoffs have been very stringent in terms of power consumption, phase-noise, area and tuning range. In this context, the aim of this work is to propose a design method, aiming to optimize the VCO design and to improve its performances essentially in terms of power consumption and phase noise.The first part of this thesis sets a study of the elements of passive LC oscillators in CMOS technology. The second part presents a complete design method, aiming to optimize the LC-VCO performance regarding the phase noise and power consumption. The evaluation of the proposed method is carried out with some test-cases in full CMOS technology. Many RF LC-VCOs have been implemented and measured. The final part of this thesis presents a new tunable VCO suitable for multi-standards applications. The frequency tuning of the VCO is ensured using an active inductor based on CMOS inverters. The desired bandwidth can be selected while achieving low surface area and low power consumption.

Des anomalies macro et micro-structurales des réseaux cérébraux impliqués dans la régulation émotionnelle ont été observées chez des adolescents et des adultes présentant un trouble dépressif majeur. Cependant, on ignore si ces anomalies se développent au fur et à mesure de la maladie dépressive ou si elles sont présentes avant. La dépression sub-syndromique de l’adolescent étant associée à un risque élevé, mais non systématique, de développer ultérieurement un trouble dépressif majeur, l'étude des corrélats neuro-anatomiques qui lui sont associées pourrait apporter des informations sur la physiopathologie de la dépression. Nous avons utilisé les données cliniques et d’Imagerie par Résonance Magnétique pondérée en T1 et en diffusion de l’étude européenne IMAGEN portant sur 2131 adolescents recrutés en population générale à 14 ans, puis réévalués à 16 ans, afin de comparer les données d’adolescents présentant une dépression sub-syndromique à celles d’adolescents non déprimés. Nous avons mis en évidence des changements structuraux chez des adolescents présentant une dépression sub-syndromique dans des régions impliquées dans la dépression. La relation entre dépression sub-syndromique à 14 ans et dépression clinique à 16 ans était en partie expliquée par un plus petit volume de cortex préfrontal médian chez les filles et par de plus faibles valeurs de fraction d’anisotropie dans les faisceaux connectant le corps calleux au cortex cingulaire antérieur chez les deux sexes. A l’adolescence, des changements cérébraux dans des régions impliquées dans la régulation émotionnelle semblent être associés à un risque accru de transition vers des formes syndromiques de dépression. / Neuroimaging findings have been reported in emotional regions in both adults and adolescents with depression but it still remains unknown whether such brain alterations can be detected before depression onset or reflect disease progression. Although subthreshold-depression in adolescence is a condition at risk for Major Depressive Disorder, not all youths with subthreshold depression will develop full-syndrome depression. Thus, studying brain correlates of subthreshold-depression in adolescence may inform on the pathophysiology of depression. We used clinical and, T1 weighted and diffusion magnetic resonance imaging data from the IMAGEN study, an European and population-based cohort of 2131 adolescents recruited from secondary schools at age 14 and followed-up at age 16. Regional gray and white matter morphometry and white matter microstructure were compared between adolescents with subthreshold-depression and healthy control adolescents. Macro and micro structural brain changes were found in adolescents with subthreshold-depression in regions involved in Major Depressive Disorder. The relation between subthreshold-depression at baseline and clinical depression at follow-up was mediated by lower medial-prefrontal gray matter volume in girls and by lower fractional anisotropy in tracts projecting from the corpus callosum to the anterior cingulate cortex in both sexes. The findings suggest that subthreshold-depression in early adolescence is associated with structural volumetric and connectivity changes in emotion-regulation circuits, and that some of these changes might denote high risk for later clinical depression.

Este trabalho apresenta o estudo de transistores de tunelamento controlados por efeito de campo, denominados TFETs. Foram realizadas análises com base em explicação teórica, simulação numérica e medidas experimentais para demonstrar a viabilidade do uso desta tecnologia como alternativa para permitir o contínuo escalamento de dispositivos. A motivação para o uso de transistores com corrente principal resultante do tunelamento de banda para banda consiste na proposta de superar o limite físico de inclinação de sublimiar da tecnologia CMOS convencional de 60 mV/década sob temperatura ambiente. Afinal, esta limitação impede a redução na tensão de alimentação de circuitos e, consequentemente, apresenta crescentes problemas quanto à dissipação de potência. Com este objetivo, foram realizadas simulações numéricas de diversas geometrias alternativas visando atenuar as características indesejáveis dos TFETs, como a corrente ambipolar e a relativamente baixa relação ION/IOFF. Inicialmente são definidos os modelos necessários para representar adequadamente os fenômenos relevantes sob variação de temperatura e é definida uma estrutura capaz de minimizar os efeitos da ambipolaridade. Posteriormente, medidas experimentais foram utilizadas para calibrar as simulações e estudar o efeito da temperatura e do dimensionamento no funcionamento de dispositivos desta tecnologia. Comparando resultados práticos e simulados, nota-se como uma redução no comprimento de porta, com a consequente inserção de uma subposição (underlap) em relação à junção canal/dreno, e uma diminuição na temperatura permitem a obtenção de valores promissores de inclinação de sublimiar e de relação ION/IOFF, compatível com a proposta de futuras aplicações digitais e analógicas. / This works presents the study of tunneling field effect transistors, namely TFETs. Analyses were performed based on theoretical explanations, numerical simulations and experimental data in order to show this technology suitability as an alternative for the continuous devices scaling. The basic idea of making use of band-to-band tunneling as the main current component comes from the possibility of reaching sub-60 mV/decade subthreshold slopes at room temperature, differently from conventional CMOS devices. After all, this physical limitation causes relevant power dissipation issues, since it requires relatively high power supply voltages. Bearing this objective, numerical simulations of several alternative geometries were performed in order to tackle TFETs disadvantages, as the undesirable ambipolar currents and the low ION/IOFF ratio. At first, it was necessary to choose the most appropriate models to take into consideration the relevant phenomena under temperature variation and to define the physical structure in order to minimize ambipolar effects. After these analyses, experimental data were used to calibrate simulation parameters and to study how temperature and physical dimensions affect the performance of devices based on this technology. Comparing experimental and simulated results, it was possible to notice that when the structure is designed with gate underlap related to channel/drain junction and the temperature decreases, the obtained values for subthreshold slope and ION/IOFF ratio may be used as an important reference of this technology as a promising alternative for both digital and analog applications.

Este trabalho apresenta o estudo de transistores de tunelamento controlados por efeito de campo, denominados TFETs. Foram realizadas análises com base em explicação teórica, simulação numérica e medidas experimentais para demonstrar a viabilidade do uso desta tecnologia como alternativa para permitir o contínuo escalamento de dispositivos. A motivação para o uso de transistores com corrente principal resultante do tunelamento de banda para banda consiste na proposta de superar o limite físico de inclinação de sublimiar da tecnologia CMOS convencional de 60 mV/década sob temperatura ambiente. Afinal, esta limitação impede a redução na tensão de alimentação de circuitos e, consequentemente, apresenta crescentes problemas quanto à dissipação de potência. Com este objetivo, foram realizadas simulações numéricas de diversas geometrias alternativas visando atenuar as características indesejáveis dos TFETs, como a corrente ambipolar e a relativamente baixa relação ION/IOFF. Inicialmente são definidos os modelos necessários para representar adequadamente os fenômenos relevantes sob variação de temperatura e é definida uma estrutura capaz de minimizar os efeitos da ambipolaridade. Posteriormente, medidas experimentais foram utilizadas para calibrar as simulações e estudar o efeito da temperatura e do dimensionamento no funcionamento de dispositivos desta tecnologia. Comparando resultados práticos e simulados, nota-se como uma redução no comprimento de porta, com a consequente inserção de uma subposição (underlap) em relação à junção canal/dreno, e uma diminuição na temperatura permitem a obtenção de valores promissores de inclinação de sublimiar e de relação ION/IOFF, compatível com a proposta de futuras aplicações digitais e analógicas. / This works presents the study of tunneling field effect transistors, namely TFETs. Analyses were performed based on theoretical explanations, numerical simulations and experimental data in order to show this technology suitability as an alternative for the continuous devices scaling. The basic idea of making use of band-to-band tunneling as the main current component comes from the possibility of reaching sub-60 mV/decade subthreshold slopes at room temperature, differently from conventional CMOS devices. After all, this physical limitation causes relevant power dissipation issues, since it requires relatively high power supply voltages. Bearing this objective, numerical simulations of several alternative geometries were performed in order to tackle TFETs disadvantages, as the undesirable ambipolar currents and the low ION/IOFF ratio. At first, it was necessary to choose the most appropriate models to take into consideration the relevant phenomena under temperature variation and to define the physical structure in order to minimize ambipolar effects. After these analyses, experimental data were used to calibrate simulation parameters and to study how temperature and physical dimensions affect the performance of devices based on this technology. Comparing experimental and simulated results, it was possible to notice that when the structure is designed with gate underlap related to channel/drain junction and the temperature decreases, the obtained values for subthreshold slope and ION/IOFF ratio may be used as an important reference of this technology as a promising alternative for both digital and analog applications.

Der Charakter der Schwellwertdynamik vieler physikalischer, chemischer und biologischer Systeme hat sich in neueren Experimenten als im wesentlichen nicht Markowsch herausgestellt. In diesem Fall sind die "Ubergangsraten von der Zeit und den Anfangsbedingungen abh"angig und es stellen sich komplexe Wahrscheinlichkeitsverteilungen f"ur die erste Durchgangszeit ein. In dieser Arbeit werden verschiedene Aspekte nicht Markowscher Schwellwertprobleme und deren Anwendung bei der Beschreibung der Dynamik von Neuronen untersucht. In dieser Arbeit entwickeln wir einen analytischen Zugang zu nicht Markowschen Problemen, dem die Theorie der Schwellwert"uberschreitung zu Grunde liegt. Im Ergebnis erhalten wir mehrere analytische N"aherungen f"ur die Wahrscheinlichkeitsverteilung der ersten Durchgangszeit f"ur Zufallsprozesse mit differenzierbaren Trajektorien. Die Qualit"at und der G"ultigkeitsbereich der N"aherungen werden von uns sorgf"altig untersucht. Die abgeleiteten N"aherungen decken dabei den gesamten Bereich zwischen fast Markowschen und stark nicht Markowschen Problemen ab. Diese analytischen N"aherungen werden in Kombination mit numerischen Methoden genutzt, um Spikemuster in resonanten und nicht-resonanten Neuronen zu untersuchen. Im Besonderen haben wir uns dabei f"ur die Entstehung spontaner, durch zellinternes Rauschen hervorgerufener, Spikemuster in stellaten (resonanten) und pyramidalen (nicht-resonanten) Zellen des entorhinalen Kortex in Ratten interessiert. Diese zwei Neuronentypen zeigten deutliche Unterschiede in den Spikemustern, die den jeweiligen Unterschieden in den unterschwelligen Dynamiken zuzuordnen sind. Des weiteren wurden negative Korrelationen in den Spikesequenzen f"ur beide Neuronentypen gefunden. Um diese negativen Korrelationen angemessen zu beschreiben, haben wir einen nicht erneuerbaren Schwellenmechanismus in das Resonate-and-Fire Modell integriert. / Recent experiments revealed the non-Markovian character of the escape dynamics in many physical, chemical and biological systems on time scales prior to relaxation. The escape rates in the non-Markovian case are time-dependent and the escape times are dictated by the initial conditions. Complex, multipeak distributions of the first passage time are characteristic for the non-Markovian case. In this thesis we investigate various aspects of the non-Markovian first passage time problem and in particular its application to the dynamics of neurons. We elaborate an analytical approach to the non-Markovian first passage time problem, which is based on the theory of level-crossings, and obtain several analytical approximations for the first passage time density of a random process with differentiable trajectories. We compare the quality of these approximations and ascertain their regions of validity. Our approximations are applicable and provide accurate results for different types of dynamics, ranging from almost Markovian to strongly non-Markovian cases. These analytical approximations in combination with numerical methods are applied to investigate the spike patterns observed in resonant and nonresonant neurons. In particular, we focus on spontaneous (driven by intrinsic noise) spike patterns obtained in stellate (resonant) and pyramidal (nonresonant) cells in the entorhinal cortex in rat. These two types of neurons exhibit striking different spike patterns attributed to the differences in their subthreshold dynamics. We show that the resonate-and-fire model with experimentally estimated parameter values can quantitatively reproduce the interspike interval distributions measured in resonant as well as in nonresonant cells. We also found negative interspike interval correlations in both types of neurons. To capture these negative correlations, we introduce a novel nonrenewal threshold mechanism in the resonate-and-fire model.