Thesis (M.Arch.) - Carleton University, 2005. / Includes bibliographical references (p. 90-91). Also available in electronic format on the Internet.

Thesis (M.Arch.) - Carleton University, 2007. / Includes bibliographical references (p. 161-164). Also available in electronic format on the Internet.

Thesis (MEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: The ever-increasing existence of electronic systems and devices within the residential environment, along with the human desire to simplify life and daily routine, is generating increased interest in the field of home automation (HA) and intelligent environments. A large variety of home automation solutions have been conceptualised or developed. However, many of these solutions are designed by experts and require professionals to install and/or operate them. Furthermore they lack the potential for seamless integration into an already functioning home environment. To bridge the gap between consumer and expert as well as allowing for integration into any existing home environment without physical alteration to the building, a modular home automation solution with seamless integration potential is proposed. The implemented system uses open source software and hardware allowing for development to continue within the existing, large open source community. It can be installed and configured without professional skills or physical alteration of the environment itself and due to its modular design, the system also allows users to add and remove functional components to and from the system providing them with a seamlessly customisable home automation solution. Conceptual design and practical implementations are covered in this document, along with recommendations for both continued research and potential avenues for expansion.

Made available in DSpace on 2014-06-11T19:26:55Z (GMT). No. of bitstreams: 0 Previous issue date: 2009-04-30Bitstream added on 2014-06-13T19:55:22Z : No. of bitstreams: 1 nunes_rso_me_sjrp.pdf: 433378 bytes, checksum: 58b9cb2328012880486c9f1011d4714a (MD5) Bitstreams deleted on 2014-12-19T18:32:50Z: nunes_rso_me_sjrp.pdf,Bitstream added on 2014-12-19T18:33:36Z : No. of bitstreams: 1 000592541.pdf: 548568 bytes, checksum: df983105ba2c1be9b281439d3ea72a48 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Dado um número inteiro positivo d, encontrar um corpo de grau d que tenha, em valor absoluto, o menor discriminante é um problema clássico e poucos resultados se tem até hoje no sentido de se resolver tal desafio. O principal interesse deste trabalho consiste em estudar o problema acima sobre os corpos de números Abelianos, particularmente aqueles de grau primo. Para tanto será preciso dominar algumas técnicas referentes ao cálculo do discriminante de corpos de números, em especial, dos corpos Abelianos. / Given a positive interger d, finding a field of degree d which has, to absolute value, the smallest discriminant it is a classical problem and few results has been got until at present time, to solve this challenge. The main purpose of this paper it is to study the problem above on Abelian numbers fields in special that ones of prime degree. However, it is necessary to know any techniques for calculating the numbers fiel discriminant, specially, to Abelian fields.

This is the main report for Mattias Larsson’s degree project for the Master’s programin System-on-Chip. The main part of this project has been to design a modularMPSoC-unit (Multi Processor System on Chip) for reuse in different typesof projects. A SoC includes and interconnects a processor and FPGA in one chip.MPSoC is an extension of SoC including more than one processor. Zynq Ultra-Scale MPSoC has been used in this design which comes in three different versionsCG, EG and EV. In this project the EG version has been used, this versionhas a quad-core ARM Cortex-A53 platform, two Cortex-R5 real-time processorand Mali-400 MP2 graphics processing unit.The modular MPSoC-unit is designed to be used in a project as the main computingunit. A modular unit has to be general enough to fit almost all demandsbut as small as possible. The area usage is always an important question butsince the modular MPSoC-unit will be designed once and have a certain size andshape its overall area coverage determines how usable it will be. It is thereforhighly important to keep the area as small as possible.This project has been focused on the design of a modular SoC-unit with focuson a schematic-level. This report raises and answers question as too which extentsit is worth using certain types of DC/DC regulators, on-chip functionalityetc. A big part of the design work has been dedicated towards the DC/DC regulationand generation of voltages to supply the circuits. The MPSoC requires anumber of voltage levels of which some supplies can be tied together while othersare analog and sensitive to digital switching noise on the supply. This reportinvestigates the possibilities to use a simple ferrite bead filter for sensitive loadssharing the same supply. Extra work has gone into trying to reduce the numberof regulators. In order to design a safe and general power supply an estimation ofthe power consumption has to be done. This report shows how to find, use and estimatethe power consumption for the main power consumer in the system. Withthis information a generic modular SoC-design approach can be found.

abstract: The construction industry is becoming more aware of its impact on the environment. It has become more sensitive to how it operates and how it can reduce the carbon footprint of the construction process. This research identifies the source of and quantities of the carbon emissions created by an operating modular home fabrication plant in producing, transporting and installing modular structures. This study demonstrates how to measure the carbon footprint created in the production of a modular home. It quantifies and reports the results on a home, on a single module and on a per square foot basis. The primary conclusions of this study are: a) electricity was found to be the largest energy source used in this fabrication process; b) the modular fabrication process consumes a significant amount of electrical energy per month; c) production volume has a bearing on the carbon footprint of each home since the carbon footprint for each period is allocated to every home produced in that period; and d) transportation of fabricated modules and set-up add to the carbon footprint. Further, a carbon calculator was produced and is included with the study. The tool calculates the impact of energy consumption on the carbon footprint of a modular factory or a modular home. It may be expanded to other process driven fabrication entities. This research is valuable to developers and builders who wish to measure the carbon impact of a modular new home delivery system. The study also provides a methodology for modular home fabricators to measure the carbon footprint of their factories and factory production. / Dissertation/Thesis / Ph.D. Construction 2010

DNA assembly and rational design are cornerstones of synthetic biology. While many DNA assembly standards have been published in recent years, only the Modular Cloning standard, or MoClo, has the advantage of publicly available part libraries for use in plant, yeast, and mammalian systems. No multipart modular library has previously been developed for use in prokaryotes. Building upon the existing MoClo assembly framework, we developed a collection of DNA parts and optimized MoClo protocols for use in E. coli. We present this assembly standard and library along with part characterization, design strategies, potential applications, and troubleshooting. Developed as part of the Cross-disciplinary Integration of Design Automation Research (CIDAR) lab collection of tools, the CIDAR MoClo Library is publicly available and contains promoters, ribosomal binding sites, coding sequences, terminators, vectors, and a set of fluorescent control plasmids. Optimized protocols reduce reaction time and cost by >80% from previously published protocols. The CIDAR MoClo Library is the first bacterial DNA part library compatible with a multipart assembly standard. To demonstrate the utility of the CIDAR MoClo system in a traditional biology context, we used the library and previous expression data to create a series of dual expression plasmids. In this manner, we produced a dual expression plasmid capable of expressing equimolar amounts of two variants of rabbit aldolase, a His-tagged wildtype protein and a single-amino-acid substitution mutant deficient in binding actin. This expression plasmid will enable the production of dimer-of-dimer heterotetramers needed for structural determination of the actin-aldolase interaction by electron microscopy. To employ CIDAR MoClo in a synthetic biology context, we produced a bioelectronic pH-mediated genetic logic gate with DNA circuits built using MoClo and integrated with Raspberry Pi computers, Twitter, and 3D printed components. Logic gates are an increasingly common biological tool with applications in cellular memory and biological computation. MoClo facilitates rapid iteration of genetic designs, better enabling the development of cellular logic. The CIDAR MoClo Library and assembly standard enable rapid design-build-test cycles in E. coli making this system advantageous for use in many areas of synthetic biology as well as traditional biological research.

The main goal of this thesis is to construct a tensor product on the 2-category BFC-A of braided fusion categories containing a symmetric fusion category A. We achieve this by introducing the new notion of Z(A)-crossed braided categories. These are categories enriched over the Drinfeld centre Z(A) of the symmetric fusion category. We show that Z(A) admits an additional symmetric tensor structure, which makes it into a 2-fold monoidal category. ByTannaka duality, A= Rep(G) (or Rep(G; w)) for a finite group G (or finite super-group (G,w)). Under this identication Z(A) = VectG[G], the category of G-equivariant vector bundles over G, and we show that the symmetric tensor product corresponds to (a super version of) to the brewise tensor product. We use the additional symmetric tensor product on Z(A) to define the composition in Z(A)-crossed braided categories, whereas the usual tensor product is used for the monoidal structure. We further require this monoidal structure to be braided for the switch map that uses the braiding in Z(A). We show that the 2-category Z(A)-XBF is equivalent to both BFC=A and the 2-category of (super)-G-crossed braided categories. Using the former equivalence, the reduced tensor product on BFC-A is dened in terms of the enriched Cartesian product of Z(A)-enriched categories on Z(A)-XBF. The reduced tensor product obtained in this way has as unit Z(A). It induces a pairing between minimal modular extensions of categories having A as their Mueger centre.

This thesis presents an assessment of a modular flap-type wave energy converter. Comparisons are made to an equivalent width rigid device. All quoted relative difference results here use the rigid device as the reference point. The variables that are evaluated are the power capture and surge and yaw foundation loads. The power capture is evaluated at both module and device level, while the foundation loads are assessed just at the device level. The investigation is carried out through testing of a 30th scale physical model in a wave tank. A key output from the work is the development of the physical model. The model consists of six flap modules, mounted on a common base structure. Each module contains a highly controllable and compact power take off system. The devices are tested in a range of conditions, primarily consisting of regular waves of different period and direction. The damping strategy employed is the simplest approach available, setting the achievable damping level on each module to be the same. For the modular device in head-on regular waves, the results show that the power capture increases significantly moving from the outer to the central modules. On average, the central pair of modules produce 68 % of the total mean power, the inner modules 25 % and the outer modules only 7 %. Between the devices, it is shown that the power captures in head-on waves are similar, with a mean relative difference of -3 %, with +/-5 % uncertainty. Thus, no statistically significant change in power capture is shown. In off-angle waves, the mean relative difference is –1 %, with +/-4 % uncertainty. However, for the highest wave direction that was tested in, 27.5 degrees, the modular device outperforms the rigid flap, by 10 %, with uncertainty of +/-1 %. The surge foundation loads are shown to be very similar for the two devices - in head-on waves, the mean relative difference is +2 %. Depending on the level of applied damping, however, significant differences in the yaw foundation loads are shown. Using damping where the power capture is maximised, the yaw loads increase by a mean of 10 %; using damping where the power to load ratio is instead maximised, the modular yaw loads are 26 % lower. Finally, the economics of the power production is estimated through division of the power capture with a cost metric, the foundation loads. While this does not provide a full techno-economic assessment, it effectively captures the interdependency of the power capture and foundation loads for the devices. The mean relative differences in the power per load ratios of the devices are found to be similar across the wave conditions. In the head-on waves, the differences are between –8 and –0.4 %, depending on damping strategy; in the off-angle waves, the differences are between –6 and +10 %. For both sets of wave conditions, the modular flap performs better when the damping is set to maximise the ratio of power capture to foundation loads. The work concludes that the modular and rigid devices produce power and experience foundation loads at similar levels in head-on waves. Given the high power capture efficiency, nearshore location, simple mode of operation and high survivability of the flap-type WEC, this suggests that the modular device is a viable stand-alone concept. The work also finds that in off-angle waves, some benefits can be achieved with an appropriately damped modular system, notably in improved power capture and reduced yaw foundation loads. These could reduce the sensitivity that flap-type devices have in off-angle waves and allow expansion of the width and hence capacity of machines. Further work should extend the wave conditions tested in, by using more irregular and directional waves, and investigate more damping strategies and geometries. Economic assessment should also be carried out.

Depuis plusieurs années, une nouvelle technologie de détecteurs à partir de matériaux semi-conducteurs CdTe/CdZnTe permet de proposer des appareils d’imagerie médicale bénéficiant de meilleures performances, notamment en imagerie gamma. Ces performances sont obtenues en combinant simultanément les technologies des détecteurs, leur géométrie, les algorithmes de reconstruction et les architectures de traitement des données. Des travaux ont précédemment montré que des améliorations étaient possibles en revoyant la collimation du système. En plus des bonnes performances intrinsèques des détecteurs à semi-conducteur, leur intérêt réside dans leur compacité, permettant de proposer des géométries plus flexibles. De nouvelles architectures ont alors vu le jour, notamment en imagerie cardiaque.Afin d’aller au-delà du compromis entre résolution spatiale, sensibilité et champ de vue imposé par la surface de détection disponible, cette thèse propose d’étudier les possibilités d’adaptation offertes par ces nouvelles architectures. Une adaptation de l’échantillonnage angulaire des têtes de détection indépendamment les unes des autres est envisagée, permettant des protocoles d’acquisition plus adaptés à la diversité des examens et des morphologies.Depuis plusieurs années, une nouvelle technologie de détecteurs à partir de matériaux semi-conducteurs CdTe/CdZnTe permet de proposer des appareils d’imagerie médicale bénéficiant de meilleures performances, notamment en imagerie SPECT. Ces performances sont obtenues en travaillant simultanément sur la localisation précise des mesures dans les détecteurs, la collimation, l’adéquation entre la géométrie des systèmes d’acquisition et les spécificités de l’examen, les algorithmes de reconstruction et les architectures de traitement des données. En plus des bonnes performances intrinsèques des détecteurs à semi-conducteur, leur intérêt réside dans leur compacité accrue par rapport aux scintillateurs et leurs photomultiplicateurs, permettant de proposer des géométries plus flexibles. De nouvelles architectures ont alors vu le jour, notamment en imagerie cardiaque.L’imagerie SPECT est contrainte par un compromis entre résolution spatiale, sensibilité et champ de vue imposé par le collimateur, et la surface de détection disponible. Afin d’aller au-delà de ce compromis, cette thèse propose d’étudier les possibilités d’adaptation en ligne offertes par ces nouvelles architectures flexibles. Un système inspiré d’un dispositif industriel déjà proposés, composé de plusieurs têtes de détection indépendantes est étudié. Une adaptation de l’échantillonnage angulaire des têtes de détection indépendamment les unes des autres est envisagée, permettant des protocoles d’acquisition plus adaptés à la diversité des examens et des morphologies.L’objectif de cette thèse est de mettre en place un processus d’acquisition / adaptation permettant de reconfigurer le parcours des têtes en fonction d’une estimation de l’objet à imager. Des solutions algorithmiques sont proposées afin d’implémenter la reconstruction en temps réel du volume à imager à partir de données rendues plus complexes par les degrés de liberté du système et par la précision du détecteur. La reconstruction obtenue en temps réel doit permettre de déterminer les zones du volume à imager les plus porteuses d’information afin d’y orienter les têtes. La simulation permet dans un premier temps l’étude théorique du concept de système modulaire, ainsi que la mise en place algorithmique du processus de reconstruction rapide, permettant une estimation exploitable de l’objet lors de l’acquisition. Quelques résultats expérimentaux permettront de valider la modélisation du système, ainsi que l’algorithme de reconstruction. Enfin, l’étude de différentes stratégies d’adaptation sera amorcée par simulation. / In the last decades, new CdTe/CdZnTe based detectors enable enhanced performances on medical imaging systems, especially in SPECT imaging. This improvements are achieved by developing simultaneously the precise localization of interaction in detectors, the collimator, matching between the system geometry and the specificities of the examination, reconstruction algorithms, and data processing architectures. In addition to their interesting intrinsic performances, CdZnTe detectors benefits from their compactness compared to previously used scintillators and photo-multiplicators. This compactness enables more flexible systems. New designs are thus proposed, especially in cardiac imaging.Performances of SPECT systems are limited by a trade-off between the spatial resolution, the sensitivity and the field of view imposed by the collimator and the available detection surface. In order to go further this compromise, this work focuses on possibilities for online adaptation offered by flexibles designs. Adaptation on a system composed of several independent detection heads already proposed by industrials is under study. We propose to adapt the angular sampling of heads, enabling acquisition protocols more accurate for different patient morphologies and examination protocols.The purpose of this work is to propose acquisition and adaptation protocols enabling the reconfiguration of the angular sampling of detection heads depending on an estimation of the imaged object. Algorithmic solutions are proposed in order to compute the reconstruction in real time from more complex data because of additional degrees of freedom and detector precision. The resulting estimation have to enable the identification of informative areas in order to focus detection heads on it. Simulations approves theoretically the use of our rapid algorithmic solution on a modular system carrying a reliable estimation of the object. Some experimental results validate the system model and the reconstruction algorithm. Then, some adaptation strategies are investigated using simulations.