Um método para classificação de imagens de satélite usando Transformada Cosseno Discreta com detecção e remoção de nuvens e sombras

Apresenta-se nesta dissertação a proposta de um algoritmo supervisionado de classificação de imagens de sensoreamento remoto, composto de três etapas: remoção ou suavização de nuvens, segmentação e classificação.O método de remoção de nuvens usa filtragem homomórfica para tratar as obstruções causadas pela presença de nuvens suaves e o método Inpainting para remover ou suavizar a preseça de sombras e nuvens densas. Para as etapas de segmentação e classificação é proposto um método baseado na energia AC dos coeficientes da Transformada Cosseno Discreta (DCT). O modo de classificação adotado é do tipo supervisionado. Para avaliar o algioritmo foi usado um banco de 14 imagens captadas por vários sensores, das quais 12 possuem algum tipo de obstrução. Para avaliar a etapa de remoção ou suavização de nuvens e sombras são usados a razão sinal-ruído de pico (PSNR) e o coeficiente Kappa. Nessa fase, vários filtros passa-altas foram comparados para a escolha do mais eficiente. A segmentação das imagens é avaliada pelo método da coincidência entre bordas (EBC) e a classificação é avaliada pela medida da entropia relativa e do erro médio quadrático (MSE). Tão importante quanto as métricas, as imagens resultantes são apresentadas de forma a permitir a avaliação subjetiva por comparação visual. Os resultados mostram a eficiência do algoritmo proposto, principalmente quando comparado ao software Spring, distribuído pelo Instituto Nacional de Pesquisas Espaciais (INPE).

Aplicação de análise de séries temporais em perfis de poço

Esta tese descreve a aplicação de análise de séries temporais em perfis de poço. Através desta técnica é possível avaliar-se a repetição e a resolução vertical de perfis, e determinar-se o intervalo de amostragem e a velocidade de perfilagem ideais para diferentes perfis. A comparação entre três poços é também feita, baseada num mesmo tipo de perfil. Para tanto, na seqüência utilizada, procurou-se manter num mesmo domínio os dados cuja quantidade total de amostras (N) por perfil não ultrapassou 2048. Desses dados, foram inicialmente retirados o valor médio das amostras e o alinhamento polinomial algébrico eventualmente nelas embutido. Em seguida, foi efetuada a aplicação do ponderador cossenoidal, do filtro passa-alta, da janela Hanning, do cálculo da função coerência, do espectro de fase, da razão sinal-ruído e dos espectros de potência do sinal e do ruído, nesta ordem. Para a função coerência, fez-se necessário o cálculo dos níveis de confiança de 50%, 90% e 95%. O cálculo do primeiro nível teve por base a necessidade de se determinar a resolução vertical de alguns perfis, e dos demais, a fim de que fosse obtida uma informação referente à localização daqueles níveis para a coerência calculada. Em relação ao espectro de fase, seu cálculo surgiu da necessidade de se obter uma informação adicional a respeito dos perfis manipulados, ou seja, o conhecimento da ocorrência ou não de deslocamento relativo de profundidade entre a seção principal e a seção repetida. A razão sinal-ruído foi calculada no sentido de possibilitar a comparação, como elemento avaliador dos diversos tipos de perfis, com a coerência e o cálculo dos espectros de potência. Os espectros de potência do sinal e do ruído foram calculados para se ter mais um parâmetro de avaliação da seção repetida, já que em tese, os espectros de potência do sinal e do ruído da seção repetida devem ser iguais aos respectivos espectros da seção principal. Os dados utilizados na aplicação da metodologia proposta foram fornecidos pela PETROBRÁS e oriundos de quatro poços da Bacia Potiguar emersa. Por questões de sigilo empresarial, os poços foram identificados como poços A, B, C e D. A avaliação da repetição entre diferentes tipos de perfis indica que, para o poço A, o perfil micro-esférico (MSFL) tem repetição melhor do que o perfil de porosidade neutrônica (CNL), o qual tem, por sua vez, repetição melhor do que o perfil de raios gama normal (GR). Para os perfis do poço D, uma diminuição da velocidade de perfilagem de 550 m/h para 275 m/h é vantajosa apenas para o perfil de porosidade neutrônica. Já a velocidade de perfilagem de 920 m/h, utilizada, na obtenção dos perfis do poço C, é totalmente inadequada para os perfis de resistividade (MSFL, ILD e ILM). A diminuição do intervalo de amostragem de 0,20 m para 0,0508 m, nos perfis de raios gama e de porosidade neutrônica, e 0,0254 m para o perfil de densidade, apresenta bons resultados quando aplicada no poço D. O cálculo da resolução vertical indica, para o perfil de porosidade neutrônica, uma superioridade qualitativa em relação ao perfil de raios gama normal, ambos pertencentes ao poço A. Para o poço C, o perfil micro-esférico apresenta uma resolução vertical na mesma ordem de grandeza da resolução do perfil de raios gama do poço B, o que evidencia ainda mais a inconveniência da velocidade de perfilagem utilizada no poço C. Já para o poço D, o cálculo da resolução vertical indica uma superioridade qualitativa do perfil de densidade de alta resolução em relação ao perfil de raios gama de alta resolução. A comparação entre os poços A, B e D, levada a efeito através dos respectivos perfis de porosidade neutrônica normais, comprova que a presença de ruído aleatório, em geral, está diretamente ligada à porosidade da formação - uma maior porosidade indica uma presença maior de ruído e, por conseguinte, uma queda qualitativa no perfil obtido. A análise do espectro de fase de cada perfil indica um deslocamento em profundidade, existente entre as seções principal e repetida de todos os perfis do poço C. E isto pôde ser confirmado com a posterior superposição das seções.

Análise de variáveis eletromiográficas e cinemáticas em protocolo incremental de corrida na esteira

Muscle fatigue is described as a cause of injuries among the many related to the running practice. Therefore, the purpose of this study was to analyze the behavior of the amplitude (RMS) and median frequency (MF) of EMG signal of the iliocostalis (CI), rectus femoris (RF), vastus lateralis (VL), vastus medialis (VM) biceps femoris (long head) (BFCL), tibialis anterior (TA) and gastrocnemius (lateral) (LNG) from the right lower limb, and the behavior of the parameters of amplitude (AP) and frequency (PF) spent in different percentages of the maximum speed during incremental protocol of treadmill running. 10 volunteers participated in this study, athletes, male, aged between 18 and 30 years with no history of injury in lower limbs and similar anthropometry. The protocol consisted of a treadmill test with initial velocity of 10 km.h-¹ and increments of 1 km.h-¹ each three minutes until volitional exhaustion, without rest interval. Synchronized collections were made of electromyographic and kinematic data. The signals were obtained through an acquisition module of biological signals (Telemyo 900 - Noraxon - USA) and software (Myoresearch - Noraxon - USA) calibrated with a sampling frequency of 1000 Hz, gain 2000 times. The raw data were filtered with a 60Hz notch filter, high pass and low pass 20Hz to 500Hz. To capture the image was used a digital video recorder (model NV-GS320, PANASONIC brand), and for image scanning and kinematic data collection was used the software Peak Motus 9.0 (ViconPeak). To obtain the values of RMS and FM analyzed the last ten passes of each speed through a specific routine (Matlab). To obtain these variables AP (m) and FP (stride I min) were analyzed for the last ten past each speed, using specific software (Peak Motus 9.0). After verification of data normality (Shapiro-Wilk) and homogeneity of the data (Levene), the comparison ...(Complete abstract click electronic access below)

Desenvolvimento do cultivo 3D a partir de células primarias de neoplasias mamárias caninas: estudo da apoptose sobre efeito ou não da carboplatina

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Tomotherapy dose distribution verification using MAGIC-f polymer gel dosimetry

Purpose: This paper presents the application of MAGIC-f gel in a three-dimensional dose distribution measurement and its ability to accurately measure the dose distribution from a tomotherapy unit. Methods: A prostate intensity-modulated radiation therapy (IMRT) irradiation was simulated in the gel phantom and the treatment was delivered by a TomoTherapy equipment. Dose distribution was evaluated by the R2 distribution measured in magnetic resonance imaging. Results: A high similarity was found by overlapping of isodoses of the dose distribution measured with the gel and expected by the treatment planning system (TPS). Another analysis was done by comparing the relative absorbed dose profiles in the measured and in the expected dose distributions extracted along indicated lines of the volume and the results were also in agreement. The gamma index analysis was also applied to the data and a high pass rate was achieved (88.4% for analysis using 3%/3 mm and of 96.5% using 4%/4 mm). The real three-dimensional analysis compared the dose-volume histograms measured for the planning volumes and expected by the treatment planning, being the results also in good agreement by the overlapping of the curves. Conclusions: These results show that MAGIC-f gel is a promise for tridimensional dose distribution measurements. (C) 2012 American Association of Physicists in Medicine. [http://dx.doi.org/10.1118/1.4704496]

Mikroskopische Untersuchung der ferroelektrischen Elektronenemission von der freien Oberfläche von Triglyzinsulfat

In dieser Arbeit wurden erstmalig orts- und energieaufgelöste Untersuchungen der ferroelektrischen Elektronenemission (FEE) durchgeführt. Als Modellsystem diente Triglyzinsulfat (TGS). Als spektromikroskopische Methode kam die Emissions-Elektronenmikroskopie zum Einsatz. Typische Schaltfelder betrugen 2 kV/mm, angelegt wurde eine sinusoïdale Wechselspannung mit 300 Hz. Die Temperatur, bei der die FEE verschwindet (32°C), liegt unterhalb der Curie-Temperatur des TGS (TC=49°C). Dieser Unterschied kann auf den Einfluss des Extraktionsfeldes des Emissions-Elektronenmikroskops (1 kV/mm) zurückgeführt werden. Oberhalb der Curie-Temperatur konnte keine Emission beobachtet werden. Die Elektroden vor und nach der Messung waren identisch, d.h. nicht zerstört, wie man es erwarten würde, wenn ein Oberflächenplasma gezündet wurde. Bei ca. 150 V/mm beginnt die Intensität der beobachteten Emission Schwankungen aufzuweisen. Dies könnte die Ursache in dem Einsatz von ersten Zündungen eines Mikroplasmas mit destruktiver Wirkung haben. Die ortsintegrierte Energieverteilung weist bei Spannungsamplituden bis 300 V zwei Maxima auf. Dies deutet auf zwei Emissionsmechanismen hin, einen sekundären (ca. 10 eV) und einen primären (ca. 13 bis 45 eV) Effekt. Die Hochenergie-Abschneidekanten korrelieren im Bereich bis 200 V bis auf wenige eV mit der angelegten Spannungsamplitude. Die Messung der ortsaufgelösten Energieverteilung zeigt, dass die primäre Emission aus den Bereichen ohne Elektrode stammt. Sie wird der FEE zugeschrieben. Diese Elektronen können– auf Grund der lokalen Felder – auf die Elektroden beschleunigt werden und hier sekundäre Prozesse auslösen (niederenergetischer Bereich des Spektrums). Dies wird durch die lokalen Spektren dieser Bereiche bestätigt.

The 1452 or 1453 A.D. Kuwae Eruption Signal Derived from Multiple Ice Core Records: Greatest Volcanic Sulfate Event of the Past 700 Years

We combined 33 ice core records, 13 from the Northern Hemisphere and 20 from the Southern Hemisphere, to determine the timing and magnitude of the great Kuwae eruption in the mid-15th century. We extracted volcanic deposition signals by applying a high-pass loess filter to the time series and examining peaks that exceed twice the 31 year running median absolute deviation. By accounting for the dating uncertainties associated with each record, these ice core records together reveal a large volcanogenic acid deposition event during 1453 - 1457 A. D. The results suggest only one major stratospheric injection from the Kuwae eruption and confirm previous findings that the Kuwae eruption took place in late 1452 or early 1453, which may serve as a reference to evaluate and improve the dating of ice core records. The average total sulfate deposition from the Kuwae eruption was 93 kg SO4/km(2) in Antarctica and 25 kg SO4/km(2) in Greenland. The deposition in Greenland was probably underestimated since it was the average value of only two northern Greenland sites with very low accumulation rates. After taking the spatial variation into consideration, the average Kuwae deposition in Greenland was estimated to be 45 kg SO4/km(2). By applying the same technique to the other major eruptions of the past 700 years our result suggests that the Kuwae eruption was the largest stratospheric sulfate event of that period, probably surpassing the total sulfate deposition of the Tambora eruption of 1815, which produced 59 kg SO4/km(2) in Antarctica and 50 kg SO4/km(2) in Greenland.

An automated method for mapping physical soil and water conservation structures on cultivated land using GIS and remote sensing techniques

An efficient and reliable automated model that can map physical Soil and Water Conservation (SWC) structures on cultivated land was developed using very high spatial resolution imagery obtained from Google Earth and ArcGIS, ERDAS IMAGINE, and SDC Morphology Toolbox for MATLAB and statistical techniques. The model was developed using the following procedures: (1) a high-pass spatial filter algorithm was applied to detect linear features, (2) morphological processing was used to remove unwanted linear features, (3) the raster format was vectorized, (4) the vectorized linear features were split per hectare (ha) and each line was then classified according to its compass direction, and (5) the sum of all vector lengths per class of direction per ha was calculated. Finally, the direction class with the greatest length was selected from each ha to predict the physical SWC structures. The model was calibrated and validated on the Ethiopian Highlands. The model correctly mapped 80% of the existing structures. The developed model was then tested at different sites with different topography. The results show that the developed model is feasible for automated mapping of physical SWC structures. Therefore, the model is useful for predicting and mapping physical SWC structures areas across diverse areas.

Magnetic susceptibility and electric conductivity of marine surficial sediments by benthic electromagnetic profiling

Distribution, accumulation and diagenesis of surficial sediments in coastal and continental shelf systems follow complex chains of localized processes and form deposits of great spatial variability. Given the environmental and economic relevance of ocean margins, there is growing need for innovative geophysical exploration methods to characterize seafloor sediments by more than acoustic properties. A newly conceptualized benthic profiling and data processing approach based on controlled source electromagnetic (CSEM) imaging permits to coevally quantify the magnetic susceptibility and the electric conductivity of shallow marine deposits. The two physical properties differ fundamentally insofar as magnetic susceptibility mostly assesses solid particle characteristics such as terrigenous or iron mineral content, redox state and contamination level, while electric conductivity primarily relates to the fluid-filled pore space and detects salinity, porosity and grain-size variations. We develop and validate a layered half-space inversion algorithm for submarine multifrequency CSEM with concentric sensor configuration. Guided by results of modeling, we modified a commercial land CSEM sensor for submarine application, which was mounted into a nonconductive and nonmagnetic bottom-towed sled. This benthic EM profiler Neridis II achieves 25 soundings/second at 3-4 knots over continuous profiles of up to hundred kilometers. Magnetic susceptibility is determined from the 75 Hz in-phase response (90% signal originates from the top 50 cm), while electric conductivity is derived from the 5 kHz out-of-phase (quadrature) component (90% signal from the top 92 cm). Exemplary survey data from the north-west Iberian margin underline the excellent sensitivity, functionality and robustness of the system in littoral (~0-50 m) and neritic (~50-300 m) environments. Susceptibility vs. porosity cross-plots successfully identify known lithofacies units and their transitions. All presently available data indicate an eminent potential of CSEM profiling for assessing the complex distribution of shallow marine surficial sediments and for revealing climatic, hydrodynamic, diagenetic and anthropogenic factors governing their formation.

First results from stable-istope measurements on ice cores from Dronning Maud Land

The European Project for Ice Coring in Antarctica (EPICA) focuses on the drilling of two deep ice cores, the first at Dome C and the second at Kohnen station (75°00' S, 0°04' E) in Dronning Maud Land (DML). This paper deals with stable-isotope records from ice cores drilled in DML. In the first season, the deep EPICA DML core reached a depth of 450 m, recovering ice approximately 7000 years old. Generally, the d18O record indicates a stable Holocene climate and shows low variability. However, during the last 4000 years (based on a preliminary time-scale) the d18O values decrease continuously by about 0.6%, and the deuterium excess values increase by about 0.5%. The correlation between d18O and the deuterium excess d is investigated for a 50m long core section and the near-surface snow. High-pass filtered profiles are positively correlated, whereas the correlation between low-pass filtered profiles is negative. A post-depositional effect due to diffusion processes can be seen in a sub-annually resolved profile from snow-pit samples. Changes in the seasonality of the evolution of the snow cover and the consequences for stable-isotope content are demonstrated with data from ice core B31.

Técnicas de procesado digital de señal aplicadas a acelerómetros

En este proyecto se estudian y analizan las diferentes técnicas de procesado digital de señal aplicadas a acelerómetros. Se hace uso de una tarjeta de prototipado, basada en DSP, para realizar las diferentes pruebas. El proyecto se basa, principalmente, en realizar filtrado digital en señales provenientes de un acelerómetro en concreto, el 1201F, cuyo campo de aplicación es básicamente la automoción. Una vez estudiadas la teoría de procesado y las características de los filtros, diseñamos una aplicación basándonos sobre todo en el entorno en el que se desarrollaría una aplicación de este tipo. A lo largo del diseño, se explican las diferentes fases: diseño por ordenador (Matlab), diseño de los filtros en el DSP (C), pruebas sobre el DSP sin el acelerómetro, calibración del acelerómetro, pruebas finales sobre el acelerómetro... Las herramientas utilizadas son: la plataforma Kit de evaluación 21-161N de Analog Devices (equipado con el entorno de desarrollo Visual DSP 4.5++), el acelerómetro 1201F, el sistema de calibración de acelerómetros CS-18-LF de Spektra y los programas software MATLAB 7.5 y CoolEditPRO 2.0. Se realizan únicamente filtros IIR de 2º orden, de todos los tipos (Butterworth, Chebyshev I y II y Elípticos). Realizamos filtros de banda estrecha, paso-banda y banda eliminada, de varios tipos, dentro del fondo de escala que permite el acelerómetro. Una vez realizadas todas las pruebas, tanto simulaciones como físicas, se seleccionan los filtros que presentan un mejor funcionamiento y se analizan para obtener conclusiones. Como se dispone de un entorno adecuado para ello, se combinan los filtros entre sí de varias maneras, para obtener filtros de mayor orden (estructura paralelo). De esta forma, a partir de filtros paso-banda, podemos obtener otras configuraciones que nos darán mayor flexibilidad. El objetivo de este proyecto no se basa sólo en obtener buenos resultados en el filtrado, sino también de aprovechar las facilidades del entorno y las herramientas de las que disponemos para realizar el diseño más eficiente posible. In this project, we study and analize digital signal processing in order to design an accelerometer-based application. We use a hardware card of evaluation, based on DSP, to make different tests. This project is based in design digital filters for an automotion application. The accelerometer type is 1201F. First, we study digital processing theory and main parameters of real filters, to make a design based on the application environment. Along the application, we comment all the different steps: computer design (Matlab), filter design on the DSP (C language), simulation test on the DSP without the accelerometer, accelerometer calibration, final tests on the accelerometer... Hardware and software tools used are: Kit of Evaluation 21-161-N, based on DSP, of Analog Devices (equiped with software development tool Visual DSP 4.5++), 1201-F accelerometer, CS-18-LF calibration system of SPEKTRA and software tools MATLAB 7.5 and CoolEditPRO 2.0. We only perform 2nd orden IIR filters, all-type : Butterworth, Chebyshev I and II and Ellyptics. We perform bandpass and stopband filters, with very narrow band, taking advantage of the accelerometer's full scale. Once all the evidence, both simulations and physical, are finished, filters having better performance and analyzed and selected to draw conclusions. As there is a suitable environment for it, the filters are combined together in different ways to obtain higher order filters (parallel structure). Thus, from band-pass filters, we can obtain many configurations that will give us greater flexibility. The purpose of this project is not only based on good results in filtering, but also to exploit the facilities of the environment and the available tools to make the most efficient design possible.

New Intermediate band sulphide nanoparticles acting in the full visible light range spectra as an active photocatalyst

Nowadays one of the challenges of materials science is to find new technologies that will be able to make the most of renewable energies. An example of new proposals in this field are the intermediate-band (IB) materials, which promise higher efficiencies in photovoltaic applications (through the intermediate band solar cells), or in heterogeneous photocatalysis (using nanoparticles of them, for the light-induced degradation of pollutants or for the efficient photoevolution of hydrogen from water). An IB material consists in a semiconductor in which gap a new level is introduced [1], the intermediate band (IB), which should be partially filled by electrons and completely separated of the valence band (VB) and of the conduction band (CB). This scheme (figure 1) allows an electron from the VB to be promoted to the IB, and from the latter to the CB, upon absorption of photons with energy below the band gap Eg, so that energy can be absorbed in a wider range of the solar spectrum and a higher current can be obtained without sacrificing the photovoltage (or the chemical driving force) corresponding to the full bandgap Eg, thus increasing the overall efficiency. This concept, applied to photocatalysis, would allow using photons of a wider visible range while keeping the same redox capacity. It is important to note that this concept differs from the classic photocatalyst doping principle, which essentially tries just to decrease the bandgap. This new type of materials would keep the full bandgap potential but would use also lower energy photons. In our group several IB materials have been proposed, mainly for the photovoltaic application, based on extensively doping known semiconductors with transition metals [2], examining with DFT calculations their electronic structures. Here we refer to In2S3 and SnS2, which contain octahedral cations; when doped with Ti or V an IB is formed according to quantum calculations (see e.g. figure 2). We have used a solvotermal synthesis method to prepare in nanocrystalline form the In2S3 thiospinel and the layered compound SnS2 (which when undoped have bandgaps of 2.0 and 2.2 eV respectively) where the cation is substituted by vanadium at a ?10% level. This substitution has been studied, characterizing the materials by different physical and chemical techniques (TXRF, XRD, HR-TEM/EDS) (see e.g. figure 3) and verifying with UV spectrometry that this substitution introduces in the spectrum the sub-bandgap features predicted by the calculations (figure 4). For both sulphide type nanoparticles (doped and undoped) the photocatalytic activity was studied by following at room temperature the oxidation of formic acid in aqueous suspension, a simple reaction which is easily monitored by UV-Vis spectroscopy. The spectral response of the process is measured using a collection of band pass filters that allow only some wavelengths into the reaction system. Thanks to this method the spectral range in which the materials are active in the photodecomposition (which coincides with the band gap for the undoped samples) can be checked, proving that for the vanadium substituted samples this range is increased, making possible to cover all the visible light range. Furthermore it is checked that these new materials are more photocorrosion resistant than the toxic CdS witch is a well know compound frequently used in tests of visible light photocatalysis. These materials are thus promising not only for degradation of pollutants (or for photovoltaic cells) but also for efficient photoevolution of hydrogen from water; work in this direction is now being pursued.

Monte Carlo limit cycle characterization

The fixed point implementation of IIR digital filters usually leads to the appearance of zero-input limit cycles, which degrade the performance of the system. In this paper, we develop an efficient Monte Carlo algorithm to detect and characterize limit cycles in fixed-point IIR digital filters. The proposed approach considers filters formulated in the state space and is valid for any fixed point representation and quantization function. Numerical simulations on several high-order filters, where an exhaustive search is unfeasible, show the effectiveness of the proposed approach.

Estudo e aplicação de guias de ondas integrados ao substrato em frequências de micro-ondas.

Esse trabalho de pesquisa apresenta um estudo detalhado sobre guias de ondas integrados ao substrato (SIW) operando em frequências de micro-ondas com base na teoria de guias de ondas retangulares (RWG). O estudo sobre guias SIW associa equações apresentadas na literatura e utiliza simulações eletromagnéticas para desenvolver um procedimento de projeto bem definido. É considerada a integração entre guias SIW e linhas de transmissão de microfita, projetando-se transições entre essas duas estruturas com o propósito de prover casamento de impedância e de viabilizar a caracterização em frequências de micro-ondas. São apresentadas considerações sobre processos de fabricação de circuitos SIW em substratos constituídos por laminados de alta frequência. Uma vez estabelecidos os procedimentos de fabricação e os critérios de projeto, a tecnologia SIW é aplicada ao projeto de três guias de ondas SIW nas bandas S e X, que foram fabricados empregando laminados de alta-frequência. Foram projetados dois filtros SIW passa-faixa empregando cavidades ressonantes e postes metálicos indutivos. Os dois filtros operam na frequência central de 10,61 GHz, sendo que um deles tem banda de passagem de 7,5%e é de 3ª ordem e o outro filtro tem banda de passagem de 15%, sendo de 5ª ordem. Foram realizadas comparações entre o desempenho simulado e experimental das estruturas SIW projetadas. Os resultados de simulações eletromagnéticas e experimentais demonstraram boa concordância. Os projetos em tecnologia SIW apresentados neste trabalho de pesquisa possuem perdas de retorno melhores que 10 dB na banda de operação e perdas por inserção de 1,0 dB a 1,5 dB. É apresentada a análise da sensibilidade do desempenho dos guias de ondas e filtros SIW projetados a desvios dimensionais típicos do processo de fabricação por microfresagem mecânica. Com os resultados experimentais e de simulação foi possível validar os procedimentos de projeto e de fabricação de circuitos SIW operando em frequências de micro-ondas.