Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O tema deste trabalho está voltado para a representação a respeito da Língua Brasileira de Sinais Libras que realizam os sujeitos surdos, diante de aspectos do Exame Nacional para Certificação de Proficiência no uso e ensino da Libras Prolibras, em Roraima. Os aspectos referem-se aos dados estatísticos em Roraima ouvintes e surdos inscritos, presentes, habilitados e aprovados no exame. Relacionam-se também ao caráter do exame, que une teste de proficiência e certificação profissional. Além das etapas (prova prática e prova objetiva) e formato das avaliações. O principal objetivo é investigar a representação da língua de sinais para os sujeitos surdos em relação às questões do Prolibras. A partir desta questão primordial desenvolveram-se outros desdobramentos, como a discussão de identidade, cultura e surdez. Para a realização desta pesquisa, foram utilizadas como registro de dados, as técnicas de grupo focal e as entrevistas semi-estruturadas. Após registro de dados, a análise foi construída diante das percepções que os sujeitos surdos expuseram sobre três eixos temáticos que delimitaram este trabalho, são eles: 1) Libras; 2) Proficiência; e 3) Prolibras. As reflexões, sobre a impressão dos participantes surdos, resultaram num novo olhar e percepção a respeito da língua de sinais e desdobramentos do Prolibras: o entendimento das vivências surdas através da visão dos sujeitos que as realizam. / The theme of this work is focused on the representation about the Brazilian Sign Language - Libras - which perform the deaf, faced with aspects of the National Examination for Certification of Proficiency in the use and teaching of Pounds - Prolibras in Roraima. Aspects refer to the statistical data in Roraima - hearing and deaf subscribers, gifts, qualified and passed the examination. Are also related to the character of the examination, which unites proficiency testing and professional certification. In addition to the steps (practical test and objective test) and format of assessments. The main objective is to investigate the representation of sign language for deaf people in relation to the issues of Prolibras. From this key issue have developed other developments, as the discussion of identity, culture and deafness. For this research, were used as data logging, the techniques of focus group and semi-structured interviews. After recording data, the analysis was built on the perceptions of deaf people exposed on three themes that marked out this work, they are: 1) Libras, 2) proficiency, and 3) Prolibras. Reflections on the printing of the deaf participants resulted in a new perspective and perception about the sign language and splits the Prolibras: understanding the deaf experiences through the vision of subjects who perform it.

Orientadores: Edson Moschim, Luiz Henrique Bonani do Nascimento / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e de Computação / Made available in DSpace on 2018-08-21T20:14:55Z (GMT). No. of bitstreams: 1 Galdino_Lidia_D.pdf: 15165913 bytes, checksum: b0a79c7ac69b13cce3baf82b480c99fc (MD5) Previous issue date: 2012 / Resumo: A exigência crescente de se atingir maiores taxas de bits para satisfazer a demanda por largura de banda na internet tem motivado a investigação de dispositivos totalmente ópticos que viabilizem o desenvolvimento de redes ópticas transparentes (TON). O intuito dessas redes é manter o fluxo de dados sempre no domínio óptico, eliminando o gargalo no processamento das conversões óptico-elétricas e eletro-ópticas existentes nos atuais sistemas de comunicações ópticas. Neste trabalho foram investigadas, por meio de experimentos e simulações computacionais, técnicas de processamento óptico de sinais. Estas técnicas utilizam o fenômeno da mistura de quatro ondas (FWM) para multiplexar as informações de dois sinais em um único sinal. Em especial, foram investigadas duas novas técnicas para codificação e regeneração totalmente óptica do sinal não coerente de multiplexação por divisão de códigos ópticos no acesso (OCDMA). Estas técnicas convertem e regeneram no domínio totalmente óptico o sinal modulado em intensidade on-off keying (OOK) e o sinal modulado pelo chaveamento por desvio de fase diferencial (DPSK) no sinal OCDMA. Os resultados obtidos comprovaram que ambas as técnicas propostas funcionam corretamente. O sinal de saída OCDMA apresentou ganho na potência de regeneração de 1,69 dB em relação ao sinal de entrada OOK ou DPSK. A aplicação das técnicas de conversão e regeneração totalmente ópticas propostas, deu origem ao estudo da tecnologia híbrida de multiplexação por divisão de comprimentos de onda com a multiplexação por divisão de códigos ópticos (WDM/OCDM) na camada de transporte de uma rede TON. Heurísticas para atribuição de códigos ópticos e comprimentos de onda, bem como arquiteturas de conversão de códigos ópticos e comprimentos de onda foram propostas. Dos resultados, comprovou-se que a heurística Random-Random apresenta um melhor desempenho quando analisada a probabilidade de bloqueio. Ainda, a arquitetura de conversão esparsa parcial de códigos ópticos e comprimentos de onda (SPOCC/SPWC) melhora a utilização de recursos de rede, diminuindo o custo de implementação da rede / Abstract: The growing demand to achieve higher bit rates to satisfy the bandwidth in the internet has motivated the investigation of all optical devices that enable the development of transparent optical networks (TON). The idea is that the data stream always remains in the optical domain, eliminating the bottleneck in the processing of electro-optical conversion that existing in current optical communication systems. In this work new all-optical signal processing techniques are investigated through computer simulations and experiments. These techniques utilize the four wave mixing effect to multiplex information of two input signals into a single one. In particular, two new all-optical techniques for coding and regeneration incoherently optical code division multiple access (OCDMA) signals were investigated. These all optical techniques convert and regenerate the optical signal modulated in intensity, on-off keying (OOK) and / or the optical signal modulated in differential phase shift (DPSK) on OCDMA signal. The results showed that both techniques worked properly. The output OCDMA signal presented an improved in the optical power regeneration of 1,69 dB. The application of all optical conversion technique with simultaneous signals regeneration led to the study of hybrid wavelength division multiplexing with optical code division multiplexing (WDM/OCDM) technology at the transport layer in a TON network. Heuristics for assigning optical codes and wavelengths as well as optical codes and wavelengths conversion architectures were proposed. From the results, it was shown that Random-Random heuristic has a better performance when analyzing the blocking probability. Still, the sparse partial optical code and wavelength conversion (SPOCC/SPWC) architecture improves the utilization of network resources, reducing the deployment cost of network / Doutorado / Telecomunicações e Telemática / Doutor em Engenharia Elétrica

[pt] Este trabalho discute técnicas para simulação de redes elétricas com ênfase em operação em tempo real. As redes possuem linhas de transmissão como elementos a parâmetros distribuídos, enquanto redes RLC correspondem aos elementos concentrados, se bem que muitos dos resultados não se restringem só a estes tipos de elementos. A modelagem de linhas de transmissão é considerada primeiro, sob o ponto de vista da comparação entre dois modelos da literatura, caracterizando sua adequação para simulação em tempo real. Desta comparação resulta uma combinação de procedimentos como adequada à simulação. Na seqüência, é apresentado um novo algoritmo de integração numérica, baseado na descrição de estado de rede e em filtragem digital multitaxa. O novo algoritmo mostrou que pode reduzir o esforço computacional da simulação da rede, ao mesmo tempo que melhorou a qualidade das mesmas. O próximo objetivo relaciona-se com as oscilações numéricas, causadas pelo uso da Integração Trapezoidal em algumas configurações de rede. Um método de análise, baseado em teoria de rede e processamento digital de sinais é desenvolvido. A Partir dele são identificadas as configurações de redes, passíveis de gerações numéricas na sua contraparte digital. Métodos convencionais de eliminação das oscilações numéricas são analisados, incluindo no método o sistema multitaxa. Em todos os casos o sistema novo apresentou melhores resultados, sendo que seu uso apropriado, pode, por se, eliminar as oscilações numéricas. Alguns casos ilustrativos de redes de sistemas de potência estão incluídos, sendo desenvolvidos em linguagem MATLAB. Eles indicam que a combinação do modelo sugerido para linha de transmissão com o novo método de integração numérica multitaxa, permite a simulação da rede com reduzido esforço computacional e portanto aumentar a possibilidade de simulação de redes complexas em tempo real. / [en] This work discusses techniques for electrical network digital simulation, emphasizing real time operation. The networks contain transmission lines as distributed elements and RLC elements as lumped parameters, although many of the results are not restricted to these elements. Transmission line modeling is considered first, comparing two of the methods found in literature from the real time simulation point of view. The convenience of their use is subject of analysis, leading, later on, to suggest a blend of approaches for real time simulation. In the sequence of the work, it is introduced a new numerical integration algorithm, based on network state space description and multirate digital filtering. The new integration method reduces the simulation computational complexity while simultaneously improving its accuracy. Next, the focus is on numerical oscillations caused by the Trapezoidal Integration approach in some network configurations. It is presented a method to identify electrical network configurations that might generate numerical oscillations in their digital counterparts. The method is a result of both, network and digital signal theory. Conventional and new techniques are applied in order to eliminate the numerical oscillations. The new multirate approach improves convetional techniques results, while properly used eliminates by itself the numerical oscillations. Some power systems network illustrative cases are included, using the new integration method and the proposed digital transmission line model. It is shown that this combination produces improved performance, final accuracy and reduced computational effort, resulting appropriate for real time simulations.

[pt] Trata-se de um sistema capaz de sintetizar sinais musicais através da geração e do processamento de espectros de freqüência que podem ocupar qualquer região da faixa de audição humana. Para obtenção de componentes espectrais passíveis de controle em amplitude e em fase, descreve-se um gerador de série de Fourier concebido a partir de multiplicadores de freqüência que utilizam redes de fase presa. Com este gerador é possível ainda a produção de espectros cujas componentes não se relacionam harmonicamente. O processamento espectral consta da imposição de um caráter dinâmico às componentes: cada harmônico do gerador de série de Fourier é modulado em amplitude por curvas que podem assumir uma multiplicidade de forma, de modo a se estabelecerem possibilidades de elaboração musical. Além disso, dispõe-se de um sistema gerador de tempos para comando dos instantes de atuação das curvas de amplitude. Finalmente, discute-se a maneira de deslocamento do espectro dentro da faixa audível. / [en] It is presented a system capable of synthesize musical signals by generating and processing frequency spectra which can be placed anywhere in the áudio frequency range. For obtaining spectral components which are amplitude and phase controlable, a Fourier series generator was conveived through frequency multipliers which use phase- locked loop techniques. With this generator, there exists too the possibility of generate spectra whose upper partials are not harmonically related. The spectral processing is carried out by imposing a dynamic character to the partials: each of them is amplitude modulated by curves capable of assuming a plurality of forms, such that possibilities of musical work can be stablished. Further, a time generator system for control of the spectrum is described. Finallly , the means for shifting the spectrum within the audio frequency range is discussed.

Falling asleep while operating a moving vehicle is a contributing factor to the statistics of road related accidents. It has been estimated that 20% of all accidents where a vehicle has been involved are due to sleepiness behind the wheel. To prevent accidents and to save lives are of uttermost importance. In this thesis, given the world’s largest dataset of driver participants, two methods of evaluating driver sleepiness have been evaluated. The first method was based on the creation of epochs from lane departures and KSS, whilst the second method was based solely on the creation of epochs based on KSS. From the epochs, a number of features were extracted from both physiological signals and the car’s controller area network. The most important features were selected via a feature selection step, using sequential forward floating selection. The selected features were trained and evaluated on linear SVM, Gaussian SVM, KNN, random forest and adaboost. The random forest classifier was chosen in all cases when classifying previously unseen data.The results shows that method 1 was prone to overfit. Method 2 proved to be considerably better, and did not suffer from overfitting. The test results regarding method 2 were as follows; sensitivity = 80.3%, specificity = 96.3% and accuracy = 93.5%.The most prominent features overall were found in the EEG and EOG domain together with the sleep/wake predictor feature. However indications have been made that complexities might contribute to the detection of sleepiness as well, especially the Higuchi’s fractal dimension.

No description available.

A stochastic leaky integrate-and-fire neuron model was implemented in this study to simulate the spiking activity of the electrosensory "P-unit" receptor neurons of the weakly electric fish Apteronotus leptorhynchus. In the context of sensory coding, these cells have been previously shown to respond in experiment to natural random narrowband signals with either a linear or nonlinear coding scheme, depending on the intrinsic firing rate of the cell in the absence of external stimulation. It was hypothesised in this study that this duality is due to the relation of the stimulus to the neuron's excitation threshold. This hypothesis was validated with the model by lowering the threshold of the neuron or increasing its intrinsic noise, or randomness, either of which made the relation between firing rate and input strength more linear. Furthermore, synchronous P-unit firing to a common input also plays a role in decoding the stimulus at deeper levels of the neural pathways. Synchronisation and desynchronisation between multiple model responses for different types of natural communication signals were shown to agree with experimental observations. A novel result of resonance-induced synchrony enhancement of P-units to certain communication frequencies was also found.

No description available.

En complément des analyses de concentration et de spéciation, les analyses des isotopes stables du mercure par MC-ICP-MS sont à l’heure actuelle un outil majeur pour tracer les sources et les transformations de cet élément dans l’environnement. Cette thèse propose le développement de techniques de pré-concentrations couplées au MC-ICP-MS afin de réaliser des analyses isotopiques de mercure au niveau du ng.L-1 (Ultra-trace). La stratégie de pré-concentration en ligne va générer des signaux transitoires courts dont la gestion en MC-ICP-MS représente un réel défi. Afin d’y répondre, une méthode de traitement de données adaptée à ce type de signaux ainsi qu’une méthode de correction du phénomène de dérive isotopique ont été développées. Dans le but de déterminer des compositions isotopiques en mercure total à de faibles concentrations, le couplage d’une génération de vapeur froide, d’une double amalgamation sur or (CVG-DGA) et d’un MC-ICP-MS a été développé. Cette nouvelle technique donne des précisions externes de l’ordre de 0.20 à 0.30‰ (2SD) pour des concentrations en mercure de 5 ng.L-1 en solution. Cette thèse présente également une nouvelle méthode d’analyse isotopique par espèce chimique (CSIA) réalisée à l’aide du couplage d’une chromatographie en phase gazeuse équipée d’un injecteur PTV (GC-PTV) et d’un MC-ICP-MS. Celle-ci permet l’analyse isotopique de plusieurs espèces chimiques avec des précisions externes de l’ordre de 0.30 à 0.40‰ (2SD) pour des concentrations en mercure de 150 ng.L-1 dans des échantillons biologiques.Les développements analytiques de cette thèse ont permis de mesurer directement et de façon automatisée des compositions isotopiques de mercure à des niveaux d’ultra-trace (jusqu’à 5ng.L-1) tout en conservant des précisions compatibles avec les principales questions environnementales posées. Ceci va permettre d’analyser des compartiments environnementaux contenant de faibles quantités de mercure (eaux naturelles, planctons,…) et ouvre ainsi de nouvelles perspectives pour une meilleure compréhension du cycle bio-géochimique du mercure. / In addition to the quantitative and speciation analysis, the analysis of mercury stable isotopes by MC-ICP-MS are now a tool of choice to track sources and pathways of this element in the environment. This PhD thesis presents the development of hyphenation between pre-concentration techniques and MC-ICP-MS to measure isotopic composition of mercury at ng.L-1 levels (Ultra-trace). The on-line pre-concentration strategy will create short transient signals which represent a real challenge for MC-ICP-MS. In order to solve it, a data treatment strategy for this particular signals and a correction method for the isotopic drift were developed.The hyphenation between a cold vapor generation, a dual gold amalgamation (CVG-DGA) and a MC-ICP-MS was developed to determine total mercury isotopic composition. This new technique gives external precisions ranging from 0.20 to 0.30‰ (2SD) for Hg concentration in solution of 5 ng.L-1. This PhD thesis also reports a new method to perform mercury compound specific isotopic analysis (CSIA) using a gas chromatography fitted with a PTV injector (GC-PTV) coupled to the MC-ICP-MS. This allows the isotopic analysis of several species with external precisions ranging from 0.30 to 0.40‰ (2SD) for mercury concentration down to 150 ng.L-1 in biological samples.The analytical developments proposed in this PhD thesis allows to automatically and directly measure mercury isotopic compositions at ultra-trace levels (down to 5 ng.L-1) while keeping precision compatible with main environmental questions. This will allow to analyze environmental compartment containing very low amount of mercury (natural waters, planktons,…) and then open new perspectives for a better understanding of the bio-geochimical cycle of mercury.

<p style="text-align: justify">Cilj ove doktorske disertacije je da predstavi model i računarski sistem koji re&scaron;ava problem osoba sa invaliditetom koja nisu u mogućnosti da koriste ruke ili funkciju govora u ostvarivanju komunikacije. Disertacija se bavi problematikom mogućnosti primene ekonomski pristupačnih asistivnih tehnologija u domenu primene servisa elektronske uprave. Od asistivnih tehnologija disertacija istražuje mogućnosti primene neinvazivne BCI tehnologije u poređenju sa sistemima baziranih na HD kamerama. Razvijen je računarski sistem koji omogućava integraciju u operativni sistem i upotrebu računara za unos komandi upotrebom detekcije moždanih talasa.</p> / <p>The main objective of this doctoral thesis is to present the model and a computer system that solves the communication problem of people with disabilities (people who cannot use their hands or the function of speech communication). The dissertation researches the possibility to apply economic affordable assistive technologies in the field of application of e-government services. Thesis explores the possibilities of application of non-invasive BCI technology compared to systems based on HD<br />cameras. Has been developed a computer system that allows the integration into the<br />operating system that allow to enter commands by the detection of brain waves.</p>