published_or_final_version / Language Studies / Master / Master of Arts

The power density of a switching converter is dependent on the size of the power circuit components. Converters are operated in the hundreds of kHz to achieve high power density since the size of the converter reactive components decrease as frequency increases. Most present day low power (<200W) DC-DC converters operate at switching frequencies up to 500kHz. Some research has been conducted on converters that can operate above 500kHz up to 4MHz. In the near future, most DC-DC switching converters for communications and computers will operate at switching frequencies of 1-10MHz in order to achieve greater power density and improved transient response. To meet the next generation requirements of these applications, four new ideas are proposed in this thesis. The first contribution is a new current source gate drive circuit for power MOSFETs. The circuit provides a nearly constant gate current to reduce switching transition times and therefore switching loss in power MOSFETs. In addition, it can recover a portion of the gate energy normally dissipated in a conventional driver. Demonstrated loss reduction of 24.8% at 10V/5A load are presented in comparison to a conventional voltage source driver for a boost converter switching at 1MHz. The second contribution is a new high efficiency 1MHz synchronous buck voltage regulator using an improved current source driver. The proposed circuit achieves short rise and fall times to reduce switching loss in the buck HS MOSFET. It also recovers a portion of the SR gate energy, enabling a loss reduction of 24% at 1.3V/30A load in comparison to a conventional driver. In the third contribution, a new switching loss model is proposed for synchronous buck voltage regulators. The model uses simple closed form equations to calculate the rise and fall times and piecewise linear approximations of the HS MOSFET voltage and current waveforms to allow quick and accurate calculation of switching loss. The final contribution is a new variable frequency digital control method for resonant converters, which is suitable for future applications switching at 10MHz. The proposed method uses frequency dithering to reduce the clock frequency demands of the digital controller. / Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2008-02-28 10:56:06.732

No description available.

In the present paper we discuss the problem of estimating model likelihoods from the MCMC output for a general mixture and switching model. Estimation is based on the method of bridge sampling (Meng and Wong, 1996), where the MCMC sample is combined with an iid sample from an importance density. The importance density is constructed in an unsupervised manner from the MCMC output using a mixture of complete data posteriors. Whereas the importance sampling estimator as well as the reciprocal importance sampling estimator are sensitive to the tail behaviour of the importance density, we demonstrate that the bridge sampling estimator is far more robust in this concern. Our case studies range from computing marginal likelihoods for a mixture of multivariate normal distributions, testing for the inhomogeneity of a discrete time Poisson process, to testing for the presence of Markov switching and order selection in the MSAR model. (author's abstract) / Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

This research investigates the role of product design on technology switching in the context of a capital-intensive product. I focus on switching rather than on new sales because switching is the primary means of changing market share in nearly mature markets. Further, the dominant logic ??? is that, because of switching costs and the related consequences, incumbents have a strong advantage when upgrading or replacing equipment. However, the literature on lead users suggests that those users at the cutting-edge are willing to meet the costs of changing technology because they have the capabilities needed to leverage significant advantages from the new technology. The extant literature on switching focuses primarily on consumers in highly competitive markets. There is little understanding of the antecedents of switching in business markets, especially in markets for capital-intensive technology-based products. This research investigates the influence of product design on switching behavior for capital-intensive high technology products, where buyers are faced with numerous implications and significant costs at each step of the process. The switching behavior for capital-intensive products has not been studied previously; because of this deficiency, we do not know the consequences for theory, that is, how different theoretical assumptions will contribute to the final decision to switch, or for managerial practice, that is, the kind of strategies managers should follow to retain existing buyers under such conditions. Previous literature did not explore explicitly the concept of product design as an influence on switching, because satisfaction and switching cost were widely used as determinants of switching decisions in competitive markets. This gap in knowledge is due to the difficulty in identifying a method that would allow one to differentiate among the products??? performance and how the difference would impact consumers??? objectives. It is also difficult for researchers to define the characteristics of high technology products that make certain products more attractive on the market than others, without substantial assistance from experts in particular products. These conditions create a barrier to investigating switching behavior for high technology products. This research is positioned in the overlapping area between product design and switching behavior. The linkage between these two bodies of literature has never been explored. The research answers two important questions: (1) what are the antecedents of technology switching in a context where there are considerable costs?, and (2) does product design encourage technology switching behavior? Dynamic capabilities theory is used to explain this research, because the decision to switch an old technology for a new one in rapidly changing technology markets is about renewing resources and capabilities to maintain competitive advantages. This research is conducted in the context of the Magnetic Resonance Imaging (MRI) industry as a case study. Considerable switching has occurred in this industry over the last decade, resulting in this industry offering a good opportunity to investigate the reasons why. The market is divided into different segments based on the region and the health care system. I selected the university hospitals segment, MRI research centers, to conduct this research study, because it is feasible to track the technology switching process for this segment over time and because this segment???s market is nearly mature. Data were collected from multiple sources including personal interviews, online surveys, annual conference database, product technical reports, and patent data. In this study, the independent variable is product design and other variables related to switching costs and marketing strategies. The dependent variable is switching behavior, which has two values: (1) ???switched,??? defined as purchasing a new technology from a different supplier, and (2) ???not switched,??? defined as repurchasing from the same supplier. After collecting surveys from decision makers who purchased MRI technology, I use logistic regression analysis to test the hypothesis that the product design has a direct impact on the switching decision of capital-intensive products. Research findings have shown that buyers are willing to switch to a different technology in spite of high associated costs, particularly when they are faced with a product that restricts their capabilities. Product design represents the most influential factor underpinning switching, because it provides more capabilities that motivate switching. Notwithstanding the fact that moving to a new supplier imposes significant challenges, including technology and relationship incompatibility, findings confirm that this distinction in product capabilities has induced some MRI buyers to move to a new supplier in order to maintain a competitive market position. The findings also confirm that support during the transition process can be achieved through marketing strategies. The findings of this research clarify our understanding of the switching behavior of capital-intensive products where successful product design is expected to play a significant role. This behavior is expected to be different from the behavior identified in previous research, because the previous research was conducted using mainly competitive markets with frequently purchased products. For lead users faced with products that restrict their capabilities, switching is an expected option despite high switching costs. Those early switchers, having capitalized on the real value of the new product, serve to encourage other users to pursue the same behavior later. The outcomes from this MRI study ??? as one example of a high technology device ??? could be applied to the different industries that share the same characteristics in terms of high rates of technological change and high switching costs, for example, military devices, aircrafts, and advanced medical and industrial devices.

Cette thèse démontre le caractère fondamental du multilinguisme de Nabokov dans son écriture en anglais : après avoir été un auteur prolifique en russe, l’écrivain a opéré une transition linguistique et n’a plus composé ses romans qu’en langue anglaise. Cette dernière ne peut nullement être réduite à une langue seconde puisque Nabokov a su l’écrire avant le russe et la parlait depuis l’enfance. Il a par ailleurs bénéficié d’une éducation polyglotte puisqu’il maîtrisait parfaitement le français. Cette thèse étudie le style de Nabokov en soulignant sa créativité linguistique et en examinant le rôle fondateur que la traduction a joué dans le positionnement de ses différentes langues les unes par rapport aux autres. L’identité bilingue de Nabokov a pour conséquence que son écriture est déviante, mais aussi défiante : elle est marquée d’étrangeté (le caractère de ce qui est étrange) et d’étrangéité (ce qui est étranger). Dans ses romans, la langue anglaise voit surgir en son sein des mots étrangers qui, typographiquement et sémantiquement, déstabilisent la lecture : le phénomène de l’alternance codique (ou code-switching) est au cœur de cette thèse et interroge le rapport entre l’auteur tyrannique (Couturier) et son lecteur. La langue anglaise elle-même perd de sa familiarité et devient étrangère : le bilinguisme de Nabokov lui confère un statut d’étranger dans la langue, ce qui lui permet de voir la violence du langage (Lecercle) et de la faire apparaître, notamment grâce à des calembours et des néologismes. Nabokov est ce que Deleuze appelle un grand écrivain à l’aide d’une formule proustienne : ses livres « sont écrits dans une sorte de langue étrangère ». / This thesis shows how central Nabokov’s multilingualism was to his prose in English: after authoring several works in Russian, the writer changed languages and then only wrote his novels in English. It was not merely a second language for him because Nabokov could write English before Russian and he had spoken it since childhood. Besides, he enjoyed a polyglot education and was fluent in French. This thesis studies Nabokov’s style in English by focusing on his linguistic creativity and by examining the founding role that translation played in how his several tongues were going to position themselves in relation to one another. Nabokov’s bilingual identity means that his writing both challenges and deviates from the norm: it is dappled with strangeness and foreignness. In his novels, foreign words irrupt in the midst of the English prose and, typographically and semantically, they destabilize the reading experience: code-switching is at the heart of this thesis and questions the relation between “the narrator’s tyranny” (Couturier) and his reader. English itself loses its familiarity and becomes a foreign language: Nabokov’s bilingualism means he is like a foreigner in this tongue; therefore he reveals the violence of language (Lecercle) and puts it into play through puns and neologisms. Nabokov is what Deleuze calls, thanks to a Proustian expression, a great writer: his books “are written in a kind of foreign language”.

Hybrid renewable energy system was introduced to improve the individual renewable energy power system’s productivity and operation-ability. This circumstance has led towards an extensive technological study and analysis on the hybrid renewable energy system. The extensive technological study is conducted using many different approaches, but in this research the linear programming, artificial intelligence and smart grid approaches are studied. This thesis proposed a complete hardware system development, implementation and construction of real-time DC Hybrid Renewable Energy System for solar-wind-battery energy source integrated with grid network support. The proposed real-time DC HRES hardware system adopts the hybrid renewable energy system concept which is composed of solar photovoltaic, wind energy system, battery energy storage system and grid network support. The real-time DC HRES hardware system research work is divided into three stages. Stage 1 involves modelling and simulation of the proposed system using MATLAB Simulink/Stateflow software. During this stage, system’s methodological design and development is emphasised. The obtained results are considered as fundamental finding to design, develop, integrate, implement and construct the real-time DC HRES hardware system. Stage II is designing and developing the electronic circuits for the real-time DC HRES hardware system using PROTEUS software. Real time simulation is performed on the electronic circuits to study and analyse the circuit’s behaviour. This stage also involves embedded software application development for the microcontroller PIC16F877A. Thus, continuous dynamic decision-making algorithm is developed and incorporated into microcontroller PIC16F877A. Next, electronic circuits and continuous dynamic decision-making algorithm are integrated with the microcontroller PIC16F877A as a real-time DC HRES hardware system to perform real time simulation. The real-time DC HRES hardware system simulation results are studied, analysed and compared with the results obtained in Stage 1. Any indifference between the obtained results in Stage 1 and Stage 2 are analysed and necessary changes are made. Stage 3 involves integrating, implementation and construction of real-time DC HRES. The continuous dynamic decision-making algorithm is also incorporated into the real microcontroller PCI16F877A development board. Real-time DC HRES’s experimental results have successfully demonstrated the system’s ability to perform supervision, coordination, management and control of all the available energy sources with lease dependency on the grid network. The obtained results demonstrated the energy management and optimisation of the available energy sources as primary power source deliver.

Le déploiement à grande échelle de l’internet des objets nécessite le développement de systèmes de radiocommunication plus économes en énergie, dont le circuit de génération de fréquences est connu pour être particulièrement énergivore. L’objectif de ce travail de thèse est donc d’une part de développer une synthèse de fréquences très faible consommation et d’autre part de démontrer les performances de la technologie FDSOI pour des applications analogiques et radiofréquences. Dans le premier chapitre sont présentées les spécifications du standard choisi -le BLE-, les spécificités de la technologie FDSOI et l'état de l’art des architectures de transmetteurs radiofréquences à faible consommation. Nous avons retenue de cette comparaison l'architecture à division par phases. Le deuxième chapitre présente les résultats de trois types de modélisation système de l’architecture ; 1 - le fonctionnement de ses composants et les points clés à respecter pour son implémentation, 2 - le comportement en bruit de phase pour la définition des spécifications, et 3 - l’impact de l’architecture sur la génération de raies spectrales parasites. Cette étude nous a permis de fixer le cahier des charges du VCRO développé au chapitre suivant. Le troisième chapitre est consacré à la conception, la réalisation et le test de 4 topologies de VCROs en technologie FDSOI 28nm et d'un circuit de test. Les premiers résultats de mesure sont encourageants mais nécessitent d’être complétés par des mesures avec PLL fractionnaire intégrée. En effet, la sensibilité des circuits à la tension d’alimentation (pushing de l’ordre de 5 GHz/V) a rendu les mesures du bruit de phase très délicates. / The large-scale deployment of IoT requires the development of more efficient energy radio systems, within which the frequency generation circuit is known to be particularly energy-consuming. The objective of this thesis is firstly to develop a very low consumption frequency synthesis and secondly to demonstrate the performance of the FDSOI technology for analog and RF applications.In the first chapter are the specifications of the chosen standard -the BLE-, the specifications of the FDSOI technology and state of the art of low power radio frequency synthesizers architecture. We have chosen from this comparison the Fractional Phase Divider architecture. The second chapter presents the results of three types of system simulations of the PLL; 1 - the operation of its components and the key points to be respected for its implementation, 2 - the phase noise behavior for the definition of specifications, and 3 - the impact of architecture on the generation of spurious. This study allowed us to set the specifications of VCROs developed in the next chapter. The third chapter is dedicated to the design, implementation and testing of four topologies of VCROs and a test circuit in FDSOI 28nm technology. The first measurement results are encouraging but they need to be complemented by an integrated fractional PLL measurement. Indeed, the sensitivity of the circuits to the supply voltage (pushing of about 5 GHz/V) made measurements of phase noise very delicate. The measured consumption is less than 0.8 mA and the surface of the circuits is of the order of 600 µm².In the fourth and final chapter we present the implementation at circuit-level of a phase synchronization PLL.

[pt] Esta dissertação investiga o fenômeno lingüístico conhecido na literatura em línguas em contato como code-switching, que consiste no uso alternado de dois ou mais códigos por falantes bilíngües ou multilíngües em uma mesma interação conversacional. O estudo desta prática discursiva em diversas línguas tem obtido interesse crescente nas últimas três décadas. Duas abordagens principais prevalecem: sócio-pragmática e sintático-gramatical. Enquanto a primeira se volta aos aspectos sócio-pragmáticos de produção individual, através da categorização, quantificação e análise das funções pragmáticas e motivações sócio-psicológicas subjacentes ao code-switching, a segunda busca formular restrições formais e padrões de ocorrência universais para este comportamento lingüístico. Este trabalho tem como objetivo apresentar uma sistematização e avaliação bibliográfica da literatura internacional em code-switching através da apresentação e caracterização das principais linhas de pesquisa, ressaltando as obras de referência e os caminhos apontados por cada abordagem. A necessidade da apresentação estruturada dos diversos aspectos do code- switching encontradas na literatura sobre línguas em contato se deve ao fato de que a pesquisa sobre este fenômeno é incipiente no Brasil e o acesso ao material é ainda restrito. Esta pesquisa, portanto, busca preencher este espaço e tornar a literatura em code-switching acessível à academia brasileira. / [en] This dissertation investigates the linguistic phenomenon referred to as codeswitching in languages in contact´s literature. Code-switching consists of the alternate use of two or more codes by bilinguals or multilinguals in the same verbal interaction. The study of this discursive practice in several languages has obtained increasing interest in the last three decades. Two main approaches prevail: socio-pragmatic and grammatical-syntactic. Whilst the first approach focuses on the socio-discursive aspects of individual production through the categorization, quantification and analysis of codeswitching pragmatic functions and socio-psychological motivations; the second one intends to formulate formal constraints and patterns of occurrence of this linguistic behavior. The present work aims at presenting a bibliographic systematization and evaluation of the international literature in code- switching through the presentation and characterization of its major research lines, emphasizing the reference works and the directions provided by each approach. The need for the structured presentation of the several aspects of code-switching found in the literature in languages in contact is due to the fact that research on this phenomenon is incipient in Brazil and the access to this material is still restricted. This research, therefore, seeks to fill this gap and make the literature in code-switching accessible to the Brazilian academy.

Perceptual bi/multi-stability—the phenomenon in which perceptual awareness switches between alternative interpretations of a stimulus—can be elicited by a large range of stimuli. The phenomenon is explored in vision, audition, touch, and even olfaction. The degree to which perceptual switching across visual and auditory bi/multi-stable paradigms depends on common or separate mechanisms remains unanswered. This main question was addressed in the current work by using four ambiguous tasks that give rise to bi/multi-stability and which are thought to involve rivalry at different levels of cognitive processing: auditory streaming and ambiguous-structure-from-motion (low- level tasks), and verbal transformations and ambiguous figures (high-level tasks). It was also investigated if individual differences in executive function (inhibitory control and set-shifting), creativity and personality traits have common relationships with perceptual switching in adults and children. A series of five experiments (four studies) were conducted. In Study 1 (two experiments), perceptual switching behaviour of adult participants was examined in the four perceptual tasks mentioned above. In Experiment 1, participants reported higher switching rates for the ambiguous figure and verbal transformations than for ambiguous motion and auditory streaming. However, in Experiment 2 participants had a higher switching rate in verbal transformations than in auditory streaming, while the switching rates in the two visual tasks did not differ significantly. The correlations between visual and auditory switching rates were similarly inconclusive: in Experiment 1, no cross-modal correlations emerged, while in Experiment 2 there were correlations between ambiguous figure and verbal transformations and between ambiguous motion and verbal transformation. Furthermore, inhibitory control, set-shifting, and creativity correlated with perceptual ii switching rates in some of the perceptual tasks, although not in a consistent manner. In Study 2, the development of perceptual switching was investigated in children in the same four tasks used in Study 1. Findings showed that the number of switches increased with age in all four perceptual tasks, indicating general maturational developments. Executive functions and creativity were not associated with the ongoing perceptual switching, which was similar to what was found in adults. In Study 3, a neuroscientific perturbation approach was used to investigate whether the superior parietal cortex is causally involved in both visual and auditory multistability as a top-down mechanism. Transcranial magnetic stimulation on the anterior and posterior superior parietal cortex did not increase or decrease the median phase durations in response to the ambiguous motion and auditory streaming. These regions were not causally involved in either visual or auditory multistability. Perceptual switching across modalities correlated nevertheless, indicating common perceptual mechanisms. In Study 4, the effects of attentional control and instructions were further investigated in ambiguous motion and auditory streaming. There were strong correlations between perceptual switching in the two tasks, confirming that there are common mechanisms. However, the effects of voluntary attention did not explain the commonalities found. Possibly the commonalities found reflect similar functionalities at more low-level sensorial mechanisms. In conclusion, perceptual switching in vision and audition share common mechanisms. These commonalities do not seem to be due to the same neural underpinning in parietal cortex. Moreover, attentional control does not explain the commonalities found, indicating a more low-level common mechanism or functionality. Perceptual switching across all ages is task-specific, more than modality specific. No central influence of inhibitory control and creativity was constantly associated with perceptual switching regardless of task/modality, supporting the distributed mechanisms hypothesis.