A.A. Milne - When We Were Very Young : cycle éclectique de chansons pour enfants, chanté par des enfants

Les fichiers qui accompagnent mon document on été réalisés avec les logiciels Microsoft Word et Finale.

Dynamic, convex, and robust optimization with Bayesian learning for response-guided dosing

Thesis (Ph.D.)--University of Washington, 2016-08

Toward Never-Ending Object Learning for Robots

Thesis (Ph.D.)--University of Washington, 2016-06

Improved Techniques for Video Compression and Communication

Thesis (Ph.D.)--University of Washington, 2016-06

Graph-based Semi-Supervised Learning in Acoustic Modeling for Automatic Speech Recognition

Thesis (Ph.D.)--University of Washington, 2016-06

Global changes in gene expression observed at the transition from growth to stationary phase in Listeria monocytogenes ScottA batch culture

Listeria monocytogenes is a food-borne Gram-positive bacterium that is responsible for a variety of infections (worldwide) annually. The organism is able to survive a variety of environmental conditions and stresses, however, the mechanisms by which L. monocytogenes adapts to environmental change are yet to be fully elucidated. An understanding of the mechanism(s) by which L. monocytogenes survives unfavourable environmental conditions will aid in developing new food processing methods to control the organism in foodstuffs. We have utilized a proteomic approach to investigate the response of L. monocytogenes batch cultures to the transition from exponential to stationary growth phase. Proteomic analysis showed that batch cultures of L. monocytogenes perceived stress and began preparations for stationary phase much earlier (approximately A(600) = 0.75, mid-exponential) than predicted by growth characteristics alone. Global analysis of the proteome revealed that the expression levels of more than 50% of all proteins observed changed significantly over a 7-9 h period during this transition phase. We have highlighted ten proteins in particular whose expression levels appear to be important in the early onset of the stationary phase. The significance of these findings in terms of functionality and the mechanistic picture are discussed.

Effect of nitrogen supply on leaf appearance, leaf growth, leaf nitrogen economy and photosynthetic capacity in maize (Zea mays L.)

Leaf area growth and nitrogen concentration per unit leaf area, N-a (g m(-2) N) are two options plants can use to adapt to nitrogen limitation. Previous work indicated that potato (Solanum tuberosum L.) adapts the size of leaves to maintain Na and photosynthetic capacity per unit leaf area. This paper reports on the effect of N limitation on leaf area production and photosynthetic capacity in maize, a C4 cereal. Maize was grown in two experiments in pots in glasshouses with three (0.84-6.0 g N pot(-1)) and five rates (0.5-6.0 g pot(-1)) of N. Leaf tip and ligule appearance were monitored and final individual leaf area was determined. Changes with leaf age in leaf area, leaf N content and light-saturated photosynthetic capacity, P a,, were measured on two leaves per plant in each experiment. The final area of the largest leaf and total plant leaf area differed by 16 and 29% from the lowest to highest N supply, but leaf appearance rate and the duration of leaf expansion were unaffected. The N concentration of expanding leaves (N-a or %N in dry matter) differed by at least a factor 2 from the lowest to highest N supply. A hyperbolic function described the relation between P-max and N-a. The results confirm the 'maize strategy': leaf N content, photosynthetic capacity, and ultimately radiation use efficiency is more sensitive to nitrogen limitation than are leaf area expansion and light interception. The generality of the findings is discussed and it is suggested that at canopy level species showing the 'potato strategy' can be recognized from little effect of nitrogen supply on radiation use efficiency, while the reverse is true for species showing the 'maize strategy' for adaptation to N limitation. (c) 2004 Elsevier B.V. All rights reserved.

A sociocultural analysis of learning to teach

This paper addresses the question of how teachers learn from experience during their pre-service course and early years of teaching. It outlines a theoretical framework that may help us better understand how teachers' professional identities emerge in practice. The framework adapts Vygotsky's Zone of Proximal Development, and Valsiner's Zone of Free Movement and Zone of Promoted Action, to the field of teacher education. The framework is used to analyse the pre-service and initial professional experiences of a novice secondary mathematics teacher in integrating computer and graphics calculator technologies into his classroom practice. (Contains 1 figure.) [For complete proceedings, see ED496848.]

SAÚDE MENTAL EM SALA DE AULA: A PERCEPÇÃO DE PROFESSORES UNIVERSITÁRIOS

A presente dissertação de mestrado aborda a percepção de professores universitários a respeito da Saúde Mental no contexto da sala de aula. O objetivo geral do trabalho é compreender como o professor universitário percebe e lida com a sua saúde mental e a dos seus alunos em sala de aula. O autor utiliza as entrevistas de um total de 11 professores de primeiro e segundo semestres dos cursos de fisioterapia, jornalismo e sistemas de informação da Universidade Metodista de São Paulo - UMESP. Com base no método da Ground Theory, são realizadas as codificações aberta, axial e seletiva dos dados que por final são agrupados em nove categorias, sendo essas: percepção da saúde mental dos professores; percepção da saúde mental dos alunos; manejo da saúde mental do professor; manejo da saúde mental dos alunos; conceito de saúde; conceito de promoção da saúde; contextos promotores da saúde; contextos não promotores da saúde e manejo de alunos geradores de desconforto. O fenômeno central levantado a partir dos dados observado indica uma prática solitária, onde o professor considera apenas seus recursos psicológicos, seus valores e crenças pessoais como instrumentos para lidar com as adversidades emergentes em sala de aula. Observase também uma super valorização da aula expositiva e da cultura do dar aula, práticas que não estimulam uma participação mais democrática, autônoma e ativa por parte dos alunos. Nesse ambiente, o aluno surge como o maior problema à prática docente, na medida em que com sua imaturidade e conflitos inerentes a adolescência não permite que o professor transmita os conteúdos planejados sendo este motivo de frustrações e desmotivação no trabalho. A falta de um espaço institucional onde os professores possam trocar experiências, expor suas angústias e buscar apoio psicológico denuncia o distanciamento dos docentes das demais instâncias da instituição. A sala de aula também não é entendida pelos docentes como um espaço possível para a promoção da saúde, sendo esta compreendida mais como uma disciplina do que um conjunto de ações e práticas passíveis de serem aplicadas e incentivadas em sala de aula.(AU)

The development of hard real-time systems using a formal approach

Hard real-time systems are a class of computer control systems that must react to demands of their environment by providing `correct' and timely responses. Since these systems are increasingly being used in systems with safety implications, it is crucial that they are designed and developed to operate in a correct manner. This thesis is concerned with developing formal techniques that allow the specification, verification and design of hard real-time systems. Formal techniques for hard real-time systems must be capable of capturing the system's functional and performance requirements, and previous work has proposed a number of techniques which range from the mathematically intensive to those with some mathematical content. This thesis develops formal techniques that contain both an informal and a formal component because it is considered that the informality provides ease of understanding and the formality allows precise specification and verification. Specifically, the combination of Petri nets and temporal logic is considered for the specification and verification of hard real-time systems. Approaches that combine Petri nets and temporal logic by allowing a consistent translation between each formalism are examined. Previously, such techniques have been applied to the formal analysis of concurrent systems. This thesis adapts these techniques for use in the modelling, design and formal analysis of hard real-time systems. The techniques are applied to the problem of specifying a controller for a high-speed manufacturing system. It is shown that they can be used to prove liveness and safety properties, including qualitative aspects of system performance. The problem of verifying quantitative real-time properties is addressed by developing a further technique which combines the formalisms of timed Petri nets and real-time temporal logic. A unifying feature of these techniques is the common temporal description of the Petri net. A common problem with Petri net based techniques is the complexity problems associated with generating the reachability graph. This thesis addresses this problem by using concurrency sets to generate a partial reachability graph pertaining to a particular state. These sets also allows each state to be checked for the presence of inconsistencies and hazards. The problem of designing a controller for the high-speed manufacturing system is also considered. The approach adopted mvolves the use of a model-based controller: This type of controller uses the Petri net models developed, thus preservIng the properties already proven of the controller. It. also contains a model of the physical system which is synchronised to the real application to provide timely responses. The various way of forming the synchronization between these processes is considered and the resulting nets are analysed using concurrency sets.

Linking finite-element analysis and computer-aided drafting processes

The aim of this research was to investigate the integration of computer-aided drafting and finite-element analysis in a linked computer-aided design procedure and to develop the necessary software. The Be'zier surface patch for surface representation was used to bridge the gap between the rather separate fields of drafting and finite-element analysis because the surfaces are defined by analytical functions which allow systematic and controlled variation of the shape and provide continuous derivatives up to any required degree. The objectives of this research were achieved by establishing : (i) A package which interpretes the engineering drawings of plate and shell structures and prepares the Be'zier net necessary for surface representation. (ii) A general purpose stand-alone meshed-surface modelling package for surface representation of plates and shells using the Be'zier surface patch technique. (iii) A translator which adapts the geometric description of plate and shell structures as given by the meshed-surface modeller to the form needed by the finite-element analysis package. The translator was extended to suit fan impellers by taking advantage of their sectorial symmetry. The linking processes were carried out for simple test structures, simplified and actual fan impellers to verify the flexibility and usefulness of the linking technique adopted. Finite-element results for thin plate and shell structures showed excellent agreement with those obtained by other investigators while results for the simplified and actual fan impellers also showed good agreement with those obtained in an earlier investigation where finite-element analysis input data were manually prepared. Some extensions of this work have also been discussed.

Investigation of a cross-layer link adaptation algorithm for IEEE 802.11n networks

Link adaptation is a critical component of IEEE 802.11 systems, which adapts transmission rates to dynamic wireless channel conditions. In this paper we investigate a general cross-layer link adaptation algorithm which jointly considers the physical layer link quality and random channel access at the MAC layer. An analytic model is proposed for the link adaptation algorithm. The underlying wireless channel is modeled with a multiple state discrete time Markov chain. Compared with the pure link quality based link adaptation algorithm, the proposed cross-layer algorithm can achieve considerable performance gains of up to 20%.

Injectable hydrogels with high fixed charge density and swelling pressure for nucleus pulposus repair:biomimetic glycosaminoglycan analogues

The load-bearing biomechanical role of the intervertebral disc is governed by the composition and organization of its major macromolecular components, collagen and aggrecan. The major function of aggrecan is to maintain tissue hydration, and hence disc height, under the high loads imposed by muscle activity and body weight. Key to this role is the high negative fixed charge of its glycosaminoglycan side chains, which impart a high osmotic pressure to the tissue, thus regulating and maintaining tissue hydration and hence disc height under load. In degenerate discs, aggrecan degrades and is lost from the disc, particularly centrally from the nucleus pulposus. This loss of fixed charge results in reduced hydration and loss of disc height; such changes are closely associated with low back pain. The present authors developed biomimetic glycosaminoglycan analogues based on sulphonate-containing polymers. These biomimetics are deliverable via injection into the disc where they polymerize in situ, forming a non-degradable, nuclear "implant" aimed at restoring disc height to degenerate discs, thereby relieving back pain. In vitro, these glycosaminoglycan analogues possess appropriate fixed charge density, hydration and osmotic responsiveness, thereby displaying the capacity to restore disc height and function. Preliminary biomechanical tests using a degenerate explant model showed that the implant adapts to the space into which it is injected and restores stiffness. These hydrogels mimic the role taken by glycosaminoglycans in vivo and, unlike other hydrogels, provide an intrinsic swelling pressure, which can maintain disc hydration and height under the high and variable compressive loads encountered in vivo. © 2013 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

ACARS:adaptive context - aware rate selection for DSRC in vehicular networks

Throughput plays a vital role for data transfer in Vehicular Networks which is useful for both safety and non-safety applications. An algorithm that adapts to mobile environment by using Context information has been proposed in this paper. Since one of the problems of existing rate adaptation algorithm is underutilization of link capacity in Vehicular environments, we have demonstrated that in wireless and mobile environments, vehicles can adapt to high mobility link condition and still perform better due to regular vehicles that will be out of communication range due to range checking and then de-congest the network thereby making the system perform better since fewer vehicles will contend for network resources. In this paper, we have design, implement and analyze ACARS, a more robust algorithm with significant increase in throughput performance and energy efficiency in the mist of high mobility of vehicles.

Field programmable gate array based target detection and gesture recognition

The move from Standard Definition (SD) to High Definition (HD) represents a six times increases in data, which needs to be processed. With expanding resolutions and evolving compression, there is a need for high performance with flexible architectures to allow for quick upgrade ability. The technology advances in image display resolutions, advanced compression techniques, and video intelligence. Software implementation of these systems can attain accuracy with tradeoffs among processing performance (to achieve specified frame rates, working on large image data sets), power and cost constraints. There is a need for new architectures to be in pace with the fast innovations in video and imaging. It contains dedicated hardware implementation of the pixel and frame rate processes on Field Programmable Gate Array (FPGA) to achieve the real-time performance. ^ The following outlines the contributions of the dissertation. (1) We develop a target detection system by applying a novel running average mean threshold (RAMT) approach to globalize the threshold required for background subtraction. This approach adapts the threshold automatically to different environments (indoor and outdoor) and different targets (humans and vehicles). For low power consumption and better performance, we design the complete system on FPGA. (2) We introduce a safe distance factor and develop an algorithm for occlusion occurrence detection during target tracking. A novel mean-threshold is calculated by motion-position analysis. (3) A new strategy for gesture recognition is developed using Combinational Neural Networks (CNN) based on a tree structure. Analysis of the method is done on American Sign Language (ASL) gestures. We introduce novel point of interests approach to reduce the feature vector size and gradient threshold approach for accurate classification. (4) We design a gesture recognition system using a hardware/ software co-simulation neural network for high speed and low memory storage requirements provided by the FPGA. We develop an innovative maximum distant algorithm which uses only 0.39% of the image as the feature vector to train and test the system design. Database set gestures involved in different applications may vary. Therefore, it is highly essential to keep the feature vector as low as possible while maintaining the same accuracy and performance^