This diploma thesis deals with the design of a precise multi-channel pulse generator for the synchronization of laser-induced breakdown spectroscopy (LIBS) instrumentation. The thesis reflects current state-of-the-art solutions in communication and trends in the field of laser spectroscopy. Moreover, the device is based on the modern implementation of embedded devices using C and C++ programming languages using the Ethernet for communication. The communication with the control software is provided via the SDD protocol; designed at CEITEC BUT. Finally, the data of the SDD protocol are sent by the hypertext transfer protocol (HTTP) protocol in native file format (JSON). This thesis brings the synchronization of the 10 devices by precision timing.

No description available.

Submitted by PPG Engenharia El?trica (engenharia.pg.eletrica@pucrs.br) on 2018-04-23T19:26:15Z No. of bitstreams: 1 KAYOL_SOARES_MAYER_DIS.pdf: 19305237 bytes, checksum: 0750bd8f9d1820565918888661bc5acb (MD5) / Approved for entry into archive by Sheila Dias (sheila.dias@pucrs.br) on 2018-05-04T12:47:14Z (GMT) No. of bitstreams: 1 KAYOL_SOARES_MAYER_DIS.pdf: 19305237 bytes, checksum: 0750bd8f9d1820565918888661bc5acb (MD5) / Made available in DSpace on 2018-05-04T13:13:01Z (GMT). No. of bitstreams: 1 KAYOL_SOARES_MAYER_DIS.pdf: 19305237 bytes, checksum: 0750bd8f9d1820565918888661bc5acb (MD5) Previous issue date: 2018-03-28 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior - CAPES / Em sistemas de comunica??o digital sem fio, o sinal transmitido pode ser afetado por v?rias fontes de distor??o, sendo as mais significantes o ru?do gaussiano branco aditivo (AWGN), multipercurso e efeito Doppler. Em ambientes onde a resposta de impulso do canal de comunica??o ? vari?vel no tempo, como em comunica??es m?veis, a distor??o por multipercurso ? din?mica. Juntamente com o efeito Doppler, o multipercurso din?mico pode at? mesmo eventualmente interromper completamente o link de comunica??o sem fio. Para minimizar este problema, equalizadores de canais autodidatas s?o usados para mitigar os efeitos de multipercurso e Doppler. Neste contexto, esta disserta??o prop?e um novo equalizador de canal autodidata (blind), baseado no algoritmo de m?dulo constante modificado n?o linear (NMCMA) e no equalizador Soft Direct Decision (SDD) em uma arquitetura concorrente. Este novo equalizador concorrente NMCMA-SDD ? comparado com o estado da arte, o equalizador concorrente com algoritmo de m?dulo constante (CMA)-SDD, operando sob os chamados "Brazil channels A-E", proposto pela Uni?o Internacional das Telecomunica??es (UIT). O equalizador proposto apresenta resultados significativos em compara??o com o estado da arte, possibilitando a opera??o em links de comunica??o sem fio entre ve?culos a?reos n?o tripulados (UAVs), ve?culos terrestres e em outros cen?rios de comunica??o din?mica. / In wireless digital communication systems, the transmitted signal may be affected by several sources of distortion, the most significant being Additive Whit Gaussian Noise (AWGN), multipath and Dopplr effect. In environmets where the impulse response of the communication channel is time variant, as in mobile communications, the multipath distortion is dynamic. Together with the Doppler effect, the dynamic multipath may even completely interrupt the wireless communication link. In order to solve this issue, blind channel equalizers are used to mitigate the multipath and Doppler effects. In this context, this dissertation proposes a novel blind channel equalizer, based on the Nonlinear Modified Constant Modulus Algorithm (NMCMA) and on the Soft Direct Desicion (SDD) equalizers in a concurrent architecture. This novel NMCMA-SDD concurrent equalizer is compared with the state of the art, the Constant Modulus Algorithm (CMA)-SDD concurrent equalizer, over the so-called "Brazil channels A-E", proposed by the International Telecommunication Union (ITU). The proposed equalizer presents significant results when compared with the state of the art, making possible its operation in wireless communication links for Unmanned Aerial Vehicles (UAVs), terrestrial vehicles, and others dynamic communication scenarios.

Foram realizadas medidas do alinhamento atômico do Ta, W e Au depois da ionização da subcamada L3 desses elementos por impacto de elétrons, via distribuição angular das linhas de raios x L, empregando alvos finos. As medições foram realizadas na linha de feixe de 10100 keV do Microtron de São Paulo. Os alvos foram confeccionados no Laboratório de Filmes Finos, pela técnica de deposição física de vapor (alvo de Au), e no Laboratório de Materiais Magnéticos, pela técnica de sputtering (alvos de Ta e W), enquanto que a caracterização deles foi realizada no Laboratório de Materiais e Feixes Iônicos, pela técnica de espectrometria de retroespalhamento Rutherford. O Microtron de São Paulo, bem como os laboratórios onde os alvos foram confeccionados e caracterizados, estão no Instituto de Física da Universidade de São Paulo. Os raios x decorrentes da ionização dos alvos foram detetados simultaneamente por três Silicon Drift Detectors localizados a 31°, 90° e 125° com relação à direção do feixe incidente. A determinação da eficiência de pico dos detetores foi realizada com um modelo analítico testado separadamente. As intensidades das linhas do multipleto de raios x L foram ajustadas em um único procedimento empregando funções Voigt para descrevê-las. O parâmetro de alinhamento A20 foi estimado a partir do conjunto de intensidades das linhas Ll, L1 e L2. A fim de comparar os resultados experimentais com as previsões teóricas foi calculado o parâmetro de alinhamento com a aproximação de Born de ondas planas não relativística, empregando as aproximações de HartreeSlater e DiracHartreeSlater para descrever os potenciais atômicos. Os resultados experimentais confirmaram a emissão não isotrópica dos raios x decorrentes de vacâncias preenchidas na subcamada L3, sendo que a magnitude desta anisotropia varia com a energia do feixe de elétrons incidente. Além disso, foram observadas anisotropias de sinais opostos para as linhas L1 e L2. A dependência em energia do parâmetro de alinhamento A20 mostra uma queda rápida de A20, com duas inversões de sinal, uma em torno de 2 E/E(L3) e outra em torno de 8 E/E(L3), e um valor não nulo de alinhamento em torno de 10 E/E(L3), para os três elementos estudados. A comparação dos resultados experimentais com os cálculos realizados mostra um bom acordo ate cerca de 5 E/E(L3) para o feixe de elétrons incidente, a partir deste ponto o experimento evidencia uma segunda inversão no sinal de A20, que não é prevista por essa teoria. Este é, a nosso saber, o primeiro trabalho a quantificar o alinhamento atômico na ionização de átomos pesados por impacto de elétrons, via distribuição angular de raios x na faixa de 1 a 10 E/E(L3). / Measurements of the atomic alignment of Ta, W and Au after the ionization of the L3 subshell of these elements were performed by electron impact via the angular distribution of x-ray lines employing thin targets. The targets were made in the Laboratório de Filmes Finos, by the technique of physical vapor deposition (Au target), and in the Laboratório de Materiais Magnéticos, by the sputtering technique (Ta and W targets), while their characterization was performed in the Laboratório de Materiais e Feixes Iônicos, by Rutherford Backscattering Spectrometry. The Microtron of São Paulo, as well as the laboratories where the targets were made and characterized, are placed at Instituto de Física of the Universidade de São Paulo. The x-rays from the ionization of the targets were detected simultaneously by three Silicon Drift Detectors placed at 31°, 90° and 125° with respect to the direction of the incident beam. The determination of the full energy peak efficiency of the detectors was performed with an analytical model previously tested. In the data analysis the L x-ray multiplet lines were fitted in a single procedure employing Voigt functions to describe them, and the Ll, L1 and L2 lines were employed simultaneously in the estimation of the alignment parameter A20. In order to compare the experimental results with the theoretical predictions, the alignment parameter was calculated with the non-relativistic plane-wave Born approximation using HartreeSlater and DiracHartreeSlater atomic potentials. The experimental results confirmed the nonisotropic emission of x-rays from filled vacancies in the L3 subshell, with the magnitude of this anisotropy varying with the energy of the incident electron beam. In addition, anisotropies of opposite signs were observed for the L1 and L2 lines. The energy dependence of the alignment parameter A20 shows a sharp fall of A20, with two sign inversions, one around 2 E/E(L3) and another around 8 E/E(L3), and a non-zero alignment value around 10 E/E(L3), for the three studied elements. Comparison of the experimental results with the performed calculations shows a good agreement up to about 5 E/E(L3) for the incident electron beam, from which the experiment evidences a second inversion in the sign of A20, which is not predicted by this theory. This is, to our knowledge, the first work to quantify the atomic alignment in the ionization of heavy atoms by electron impact, via an angular distribution of x-rays in the interval from 1 to 10 E/E(L3).

Foram realizadas medidas do alinhamento atômico do Ta, W e Au depois da ionização da subcamada L3 desses elementos por impacto de elétrons, via distribuição angular das linhas de raios x L, empregando alvos finos. As medições foram realizadas na linha de feixe de 10100 keV do Microtron de São Paulo. Os alvos foram confeccionados no Laboratório de Filmes Finos, pela técnica de deposição física de vapor (alvo de Au), e no Laboratório de Materiais Magnéticos, pela técnica de sputtering (alvos de Ta e W), enquanto que a caracterização deles foi realizada no Laboratório de Materiais e Feixes Iônicos, pela técnica de espectrometria de retroespalhamento Rutherford. O Microtron de São Paulo, bem como os laboratórios onde os alvos foram confeccionados e caracterizados, estão no Instituto de Física da Universidade de São Paulo. Os raios x decorrentes da ionização dos alvos foram detetados simultaneamente por três Silicon Drift Detectors localizados a 31°, 90° e 125° com relação à direção do feixe incidente. A determinação da eficiência de pico dos detetores foi realizada com um modelo analítico testado separadamente. As intensidades das linhas do multipleto de raios x L foram ajustadas em um único procedimento empregando funções Voigt para descrevê-las. O parâmetro de alinhamento A20 foi estimado a partir do conjunto de intensidades das linhas Ll, L1 e L2. A fim de comparar os resultados experimentais com as previsões teóricas foi calculado o parâmetro de alinhamento com a aproximação de Born de ondas planas não relativística, empregando as aproximações de HartreeSlater e DiracHartreeSlater para descrever os potenciais atômicos. Os resultados experimentais confirmaram a emissão não isotrópica dos raios x decorrentes de vacâncias preenchidas na subcamada L3, sendo que a magnitude desta anisotropia varia com a energia do feixe de elétrons incidente. Além disso, foram observadas anisotropias de sinais opostos para as linhas L1 e L2. A dependência em energia do parâmetro de alinhamento A20 mostra uma queda rápida de A20, com duas inversões de sinal, uma em torno de 2 E/E(L3) e outra em torno de 8 E/E(L3), e um valor não nulo de alinhamento em torno de 10 E/E(L3), para os três elementos estudados. A comparação dos resultados experimentais com os cálculos realizados mostra um bom acordo ate cerca de 5 E/E(L3) para o feixe de elétrons incidente, a partir deste ponto o experimento evidencia uma segunda inversão no sinal de A20, que não é prevista por essa teoria. Este é, a nosso saber, o primeiro trabalho a quantificar o alinhamento atômico na ionização de átomos pesados por impacto de elétrons, via distribuição angular de raios x na faixa de 1 a 10 E/E(L3). / Measurements of the atomic alignment of Ta, W and Au after the ionization of the L3 subshell of these elements were performed by electron impact via the angular distribution of x-ray lines employing thin targets. The targets were made in the Laboratório de Filmes Finos, by the technique of physical vapor deposition (Au target), and in the Laboratório de Materiais Magnéticos, by the sputtering technique (Ta and W targets), while their characterization was performed in the Laboratório de Materiais e Feixes Iônicos, by Rutherford Backscattering Spectrometry. The Microtron of São Paulo, as well as the laboratories where the targets were made and characterized, are placed at Instituto de Física of the Universidade de São Paulo. The x-rays from the ionization of the targets were detected simultaneously by three Silicon Drift Detectors placed at 31°, 90° and 125° with respect to the direction of the incident beam. The determination of the full energy peak efficiency of the detectors was performed with an analytical model previously tested. In the data analysis the L x-ray multiplet lines were fitted in a single procedure employing Voigt functions to describe them, and the Ll, L1 and L2 lines were employed simultaneously in the estimation of the alignment parameter A20. In order to compare the experimental results with the theoretical predictions, the alignment parameter was calculated with the non-relativistic plane-wave Born approximation using HartreeSlater and DiracHartreeSlater atomic potentials. The experimental results confirmed the nonisotropic emission of x-rays from filled vacancies in the L3 subshell, with the magnitude of this anisotropy varying with the energy of the incident electron beam. In addition, anisotropies of opposite signs were observed for the L1 and L2 lines. The energy dependence of the alignment parameter A20 shows a sharp fall of A20, with two sign inversions, one around 2 E/E(L3) and another around 8 E/E(L3), and a non-zero alignment value around 10 E/E(L3), for the three studied elements. Comparison of the experimental results with the performed calculations shows a good agreement up to about 5 E/E(L3) for the incident electron beam, from which the experiment evidences a second inversion in the sign of A20, which is not predicted by this theory. This is, to our knowledge, the first work to quantify the atomic alignment in the ionization of heavy atoms by electron impact, via an angular distribution of x-rays in the interval from 1 to 10 E/E(L3).

Modern state-of-the-art object detection models are based on convolutional neural networks and can be divided into single-shot detectors and two-stage detectors. Two-stage detectors exhibit impressive detection performance but their complex pipelines make them slow. Single-shot detectors are not as accurate as two-stage detectors, but are faster and can be used for real-time object detection. Despite the fact that single-shot detectors are faster, a large number of calculations are still required to produce a prediction that not many embedded devices are capable of doing in a reasonable time. Therefore, it is natural to ask if single-shot detectors could become faster even. Pruning is a technique to reduce the size of neural networks. The main idea behind network pruning is that some model parameters are redundant and do not contribute to the final output. By removing those redundant parameters, fewer computations are needed to produce predictions, which may lead to a faster inference and since the parameters are redundant, the model accuracy should not be affected. This thesis investigates two approaches for pruning the SSD-MobileNet- V2 single-shot detector. The first approach prunes the single-shot detector by a large portion and retrains the remaining parameters only once. In the other approach, a smaller portion is pruned, but pruning and retraining are done in an iterative fashion, where pruning and retraining constitute one iteration. Beyond comparing two pruning approaches, the thesis also studies the tradeoff between model accuracy and inference speed that pruning induces. The results from the experiments suggest that the iterative pruning approach preserves the accuracy of the original model better than the other approach where pruning and finetuning are performed once. For all four pruning levels that the two approaches are compared iterative pruning yields more accurate results. In addition, an inference evaluation indicates that iterative pruning is a good compression method for SSD-MobileNet-V2, finding models that both are faster and more accurate than the original model. The thesis findings could be used to guide future pruning research on SSD-MobileNet- V2, but also on other single-shot detectors such as RetinaNet and the YOLO models. / Moderna modeller för objektsdetektering bygger på konvolutionella neurala nätverk och kan delas in i ensteg- och tvåstegsdetektorer. Tvåstegsdetektorer uppvisar imponerande detektionsprestanda, men deras komplexa pipelines gör dem långsamma. Enstegsdetektorer uppvisar oftast inte lika bra detektionsprestanda som tvåstegsdetektorer, men de är snabbare och kan användas för objektdetektering i realtid. Trots att enstegsdetektorer är snabbare krävs det fortfarande ett stort antal beräkningar för att få fram en prediktering, vilket inte många inbyggda enheter kan göra på rimlig tid. Därför är det naturligt att fråga sig om enstegsdetektorer kan bli ännu snabbare. Nätverksbeskärning är en teknik för att minska storleken på neurala nätverk. Huvudtanken bakom nätverksbeskärning är att vissa modellparametrar är överflödiga och inte bidrar till det slutliga resultatet. Genom att ta bort dessa överflödiga parametrar krävs färre beräkningar för att producera en prediktering, vilket kan leda till att nätverket blir snabbare och eftersom parametrarna är överflödiga bör modellens detektionsprestanda inte påverkas. I den här masteruppsatsen undersöks två metoder för att beskära enstegsdetektorn SSD-MobileNet-V2. Det första tillvägagångssättet går ut på att en stor del av detektorn vikter beskärs och att de återstående parametrarna endast finjusteras en gång. I det andra tillvägagångssättet beskärs en mindre del, men beskärning och finjustering sker på ett iterativt sätt, där beskärning och finjustering utgör en iteration. Förutom att jämföra två metoder för beskärning studeras i masteruppsatsen också den kompromiss mellan modellens detektionsprestanda och inferenshastighet som beskärningen medför. Resultaten från experimenten tyder på att den iterativa beskärningsmetoden bevarar den ursprungliga modellens detektionsprestanda bättre än den andra metoden där beskärning och finjustering utförs en gång. För alla fyra beskärningsnivåer som de två metoderna jämförs ger iterativ beskärning mer exakta resultat. Dessutom visar en hastighetsutvärdering att iterativ beskärning är en bra komprimeringsmetod för SSD-MobileNet-V2, eftersom modeller som både snabbare och mer exakta än den ursprungliga modellen går att hitta. Masteruppsatsens resultat kan användas för att vägleda framtida forskning om beskärning av SSD-MobileNet-V2, men även av andra enstegsdetektorer, t.ex. RetinaNet och YOLO-modellerna.

Cette thèse développe un système d'analyse en ligne compact, précis, rapide et autonome par spectrométrie de fluorescence X raies L, pour quantifier les transuraniens (uranium, plutonium et américium) présents dans les solutions et effluents des procédés du plutonium. Au cours de cette étude, la configuration, les paramètres de ce système et une méthode de quantification non destructive ont été définis puis la capacité du spectromètre à dissocier ces éléments dans de fortes concentrations relatives a été vérifiée. Des solutions contenant des éléments non radioactifs ont d'abord été étudiées afin de s'affranchir des effets dus à la désintégration des transuraniens et d'optimiser efficacement le spectromètre. Par simulations Monte-Carlo avec le code PENELOPE, une méthode de quantification a été mise au point et diverses configurations pour optimiser l'intensité d'excitation de l'échantillon et mesurer une raie de diffusion Compton de la source X ont été testées numériquement. Ces configurations ont été expérimentées avec un spectromètre prototype sur des solutions non radioactives pour valider et compléter la configuration la plus intéressante. Des solutions contenant des éléments en fortes concentrations relatives ont été analysées afin de déterminer une méthode de déconvolution des spectres avec le logiciel COLEGRAM et démontrer la faculté de cette méthode à quantifier de telles solutions. Quelques solutions d'uranium et de plutonium ont été analysées pour s'approcher au plus près des futures conditions expérimentales, caractériser et diminuer les effets dus à la désintégration des radioéléments. / This thesis deals with the development of a new compact, accurate, fast, without cooling liquid, fluorescence L X-ray spectrometer, with the aim to determinate online transuranic elements (uranium, plutonium and americium) in nuclear materials reprocessing. The objective was to define the configuration and the characteristics of this spectrometer and the method to quantify transuranic elements between 0.1 g/L and 20 g/L. To minimize as law as possible the manipulation of these elements, we have used an original approach, namely Monte-Carlo simulations and none radioactive surrogate elements. The study of these solutions allowed to eliminate the specific effects of transuranic elements decay (internal conversion) and to optimize the spectrometer. Monte-Carlo simulations with the PENELOPE code were used for two reasons. Firstly, we have developed an analytical method to correct the matrix effects. Then, we have selected three systems to produce a quasi-monochromatic X-rays beam from the X-ray generator, to optimize the intensity of the L X-ray fluorescence spectra, and to measure the Compton scatter peak. These systems were then tested with an adjustable spectrometer in order to select and optimize the best configuration. We have shown that, after peaks and left tails separation from spectra with COLEGRAM software, the spectrometer can be used to analyze solutions with various concentrations of thallium and bismuth. Several solutions of uranium or plutonium were analyzed to determine the effects of interferences from gamma rays and internal conversion with the fluorescence L X-rays. These effects have been corrected by subtracting a passive spectrum to an active one.

Detta examensarbete undersöker möjligheten att tillämpa tidiga och ej så beräkningstunga CFD-simuleringar (’Computational Fluid Dynamics’) för optimering av gasväxlingskomponenter tidigt i produktutvecklingsprocessen. Examensarbetet har innefattat framtagande av generellt applicerbara inställningar och riktlinjer vid modellering av gasväxlingskomponenter med programvaran OpenFOAM, optimerade för att nyttja så låg beräkningskraft som möjligt men fortfarande erhålla tillräckligt noggranna och pålitliga resultat. Rapporten ger en teoretiskt genomlysning med motivering av simuleringsrelaterade val i OpenFOAM i syfte att validera och efter behov justera de inställningar och parameterval ansatta i det befintliga Excel skriptet – framtaget av avdelningen med ansvar för avgasefterbehandlingssystem – som automatiserar uppställningen av OpenFOAM:s simuleringsrelaterade filstruktur. Detta i syfte att erhålla och säkerställa pålitliga resultat vid användning av skriptet anpassat för simulering av gasväxlingskomponenter. Rapporten redovisar även bakomliggande analyser för olika inställningar och val i OpenFOAM i syfte att tjäna som välgrundade riktlinjer genom hela CFD-processen; såsom för skapande av geometriska modeller, generering av volymsnät, åtgärder för att säkerställa konvergens samt uppställning och tolkning av resultatet. En av examensarbetets leveranser är en metodbeskrivning utformad med praktiska riktlinjer för genomförandet av hela CFD-processen med avsedda programvaror samt med tillhörande praxis avseende simuleringsrelaterade val genom processen. Ett antal genomförda fallstudier i examensarbetet visar att OpenFOAM har potential att utföra initiala och indikativa simuleringar för att mer tidseffektivt möjliggöra en bredare idégenerering i konceptutvecklingsfasen. / This thesis investigates the possibility of applying early and not so computational heavy simulations using Computational Fluid Dynamics (CFD) for optimization of gas exchange components early in the product development process. The scope of the thesis includes the development of generally applicable settings and guidelines – documented in a methodology – when modelling gas exchange components with the software OpenFOAM, optimized to use as low computational power as possible but still obtaining sufficiently accurate and reliable data. The report consists of a theoretical study and motives for simulation-related choices in OpenFOAM in order to adjust and validate the settings and parameter choices applied in the existing Excel script – developed by the department responsible for exhaust aftertreatment systems – that automates the set-up of OpenFOAM's simulation-related file structure. This in order to obtain and ensure reliable results using the script adapted to the environment for gas exchange components. The report also presents underlying analyses for different settings and choices in order to serve as well-founded guidelines throughout the CFD-process; such as the creation of geometric models, the generation of mesh, measures to ensure convergence, and the set-up and interpretation of the results. One of the deliverables of the thesis is a method description designed with practical guidelines for the entire CFD-process with intended software’s and with associated best practices in regards of simulation-related choices throughout the process. A number of case studies – that has been simulated in this thesis – show that OpenFOAM has the potential to perform initial and indicative simulations enabling a more time-efficiently and broader ideation in the concept development phase.