Medições em túnel de vento e em escala real têm mostrado que as pressões do vento natural não estão totalmente correlacionadas espacialmente. O uso da Decomposição Ortogonal Própria (DOP) permite a obtenção dos desvios padrões das respostas da estrutura a partir da matriz de coeficientes de correlação espacial das pressões do vento nos diferentes pontos de tomadas em túnel. O método indicado pela NBR6123 - Forças devido ao vento em edificações supõe um vento de rajada totalmente correlacionado espacialmente dentro das dimensões da edificação. O objetivo deste trabalho é comparar a resposta estrutural obtida pelos métodos da Decomposição Ortogonal Própria (MDOP) e da NBR6123. Para tal foi escolhida uma estrutura tipo treliça espacial em arco semicircular com 53,00 m de largura, 26,50m de flecha e 92,80m de comprimento. A escolha da estrutura tipo treliça espacial tem o motivo de este tipo de estrutura possuir uma continuidade estrutural capaz de difundir o desvio padrão devido às pressões em uma tomada em uma extremidade da construção para uma resposta em um elemento estrutural na outra extremidade. Foi construído um modelo em escala reduzida para realização de ensaios no túnel de vento do Laboratório de Aerodinâmica das Construções (LAC) da Universidade Federal do Rio Grande do Sul - UFRGS. A partir daí empregou-se o MDOP para a obtenção da resposta estrutural e comparou-se com a resposta obtida pelo método da NBR6123. Os resultados demonstram que os distintos padrões de escoamento em torno do modelo resultam em um carregamento com níveis também distintos de correlação e que o MDOP é uma poderosa ferramenta de análise, permitindo um projeto mais seguro e otimizado da estrutura. / Both wind tunnel and real scale measurements have been indicating that the natural wind pressures are not entirely correlated spatially. The use of the Proper Orthogonal Decomposition (POD) allows the acquisition of the structural responses standard deviations from the wind pressures spatial correlation coefficients matrix on the various measurement points in a tunnel. The method shown by the "NBR6123 - Forças devido ao vento em edificações", on the other hand, assumes a gust of wind entirely correlated spatially. The objective of this study was to compare the structural response obtained by the proper orthogonal decomposition methods (POD) with that obtained by the NBR6123 methods. In order to do that, a spatial truss-type structure in semicircular arch was chosen, with a 53,00 m width, a 26,50 m arrow and a 92,80 m length. The choice for a spatial truss structure was made because it has a structural continuity capable of spreading the pressure's standard deviation from a given measurement point on one extremity of the construction, to a response in a structural element on the other extremity. A reduced scale model was built in the Laboratório de Aerodinâmica das Construções (LAC) from the Universidade Federal do Rio Grande do Sul - UFGRS -, in order to perform the tests. Then, the POD\ was employed to obtain the structural responses, which were compared to those obtained employing the NBR6123. The results show that the distinct flow patterns around the model result in a load with also distinct levels of correlation. In conclusion, the POD is a powerful analysis tool that allows a safer and optimized structural design.

Submitted by (ana.araujo@ufrpe.br) on 2016-08-08T13:12:45Z No. of bitstreams: 1 Marcos Flaviano Matos Soares.pdf: 1499686 bytes, checksum: 363f7807680627fa59738a37fa059d53 (MD5) / Made available in DSpace on 2016-08-08T13:12:45Z (GMT). No. of bitstreams: 1 Marcos Flaviano Matos Soares.pdf: 1499686 bytes, checksum: 363f7807680627fa59738a37fa059d53 (MD5) Previous issue date: 2010-08-27 / The use of modern computational resources together with the development of mathematical, statistical and computational techniques in the treatment of geographical information, has been important a better understanding of patterns of geographically distributed data. An application of these resources is the detection and interpretation of the spatial distribution of hotspots, a term used for elements registered by satellite sensors of surface regions with high temperature. The number of hot pixels should not be identified with the number of fires, because there are factors that can influence real fire detection, where both omission and false identification are possible. The present work aims at identification of space correlations of hotspots in Brazil detected by the satellite NOAA 12, during the period 1998- 2006, and provide aid in the choice of theoretical models inferential express themselves through spatial distribution across the spatial correlation property of stochastic processes generating this phenomenon, using the method to calculate the fractal dimension developed by Grassberger and Proccacia. It is found that spatial distribution of hot pixels for individual years under study demonstrates fractal behavior with the correlation dimension approximately DCORR 1.6. The value of the fractal dimension for data grouped by month along the whole period, is also close to 1.6, except for the months of january and april which display two regions of fractal behavior, with DCORR 1.0 for distances below 10km, and DCORR 1.6 for larger distance. This behavior suggests possible multifractality, and requires further phenomenological studies to be well understood. The results of the current work should be taken into account in the development and validation of theoretical and computational models of the stochastic processes of this phenomenon, as well as related phenomena, such as e.g. carbon emission. / Uso de recursos computacionais através do desenvolvimento crescente de técnicas matemáticas, estatísticas e computacionais no tratamento de informação geográfica tem sido importante no desenvolvimento de estudos de padrões de distribuição geográfica. Uma aplicação destes recursos é a detecção de focos de calor, que é uma expressão utilizada para interpretar o registro de calor captado na superfície do solo pelos sensores dos satélites. Para tanto o número de focos de calor registrados não se deve afirmar como sendo igual ao número de queimadas, pois existem fatores que podem influenciar na identificação desses focos de calor como os erros de omissão e inclusão. O presente trabalho tem como objetivo identificar correlações espaciais de focos de calor no Brasil detectados pelo satélite NOAA 12, durante o período 1998-2006, e fornecer auxílio na escolha de modelos teóricos inferenciais que se expressem por meio de distribuição espacial através da correlação espacial, propriedade dos processos estocásticos gerados deste fenômeno, usando o método para calcular a dimensão fractal desenvolvido por Grassberger e Proccacia. A distribuição espacial dos focos de calor para os anos individuais observados demonstra fractalidade, com dimensão de correlação aproximadamente DCORR 1.6. Valor da dimensão fractal para os dados agrupados por mês em todo período observado também tem valor próximo a 1.6, exceto meses de janeiro e abril que demonstram duas regiões de fractalidade, com DCORR 1.0 para distâncias abaixo de 10 Km, e DCORR 1.6 para distâncias maiores. Este comportamento exige estudos fenomenológicos para seu melhor entendimento, e sugere possível multifractalidade. Os resultados alcançados neste trabalho devem ser levados em consideração no desenvolvimento e validação dos modelos teóricos e computacionais dos processos estocásticos geradores deste fenômeno, bem como fenômenos relacionados, como, por exemplo, emissão de carbono.

This study analyzes data acquired in French Polynesia in the Pacific and The Bahamas (Atlantic), both oceans affected by recent, well documented and sequential disturbances. For the purposes of this study, a disturbance is defined as a perturbation of environmental, physical or biological conditions that causes a distinct change in the ecosystem. After several decades of coral bleaching events, biological change, and anthropogenic impacts, rapid assessments of the coral community were accomplished by collecting photo-transects across the reefs to extract size structure of the corals, percent live tissue cover and perform a faunal evaluation. Cluster analyses and spatial autocorrelation tests were done to examine the community structure and dynamics at both locations. All multivariate analyses pointed to a disturbed ecosystem and the lack of spatial correlation indicated the impact of a local disturbance over that of a regional event. In assessing the spatial coral community structure, different responses to large versus small scales of disturbance were found. This emphasizes the importance of tailoring management of coral reefs to specific impacts. These two distinct regions were shown to have correlated spatial response patterns to sequential disturbances, supporting the idea of community pattern signatures for different scales of disturbance and the need for an adjustment in management protocols.

Full dimension (FD) multiple-input multiple-output (MIMO) technology has recently attracted substantial research attention in the 3rd Generation Partnership Project (3GPP) as a promising technique for the next-generation of wireless communication networks. FD-MIMO scenarios utilize a planar two-dimensional (2D) active antenna system (AAS) that not only allows a large number of antenna elements to be placed within feasible base station (BS) form factors, but also provides the ability of elevation beamforming. This dissertation presents the elevation beamforming analysis for cellular networks utilizing FD massive MIMO antenna arrays. In particular, two architectures are proposed for the AAS - the uniform linear array (ULA) and the uniform circular array (UCA) of antenna ports, where each port is mapped to a group of vertically arranged antenna elements with a corresponding downtilt weight vector. To support FD-MIMO techniques, this dissertation presents two different 3D ray-tracing channel modeling approaches, the ITU based ‘antenna port approach’ and the 3GPP technical report (TR) 36.873 based ‘antenna element approach’. The spatial correlation functions (SCF)s for both FD-MIMO arrays are characterized based on the antenna port approach. The resulting expressions depend on the underlying angular distributions and antenna patterns through the Fourier series coefficients of the power spectra and are therefore valid for any 3D propagation environment. Simulation results investigate the performance patterns of the two arrays as a function of several channel and array parameters. The SCF for the ULA of antenna ports is then characterized in terms of the downtilt weight vectors, based on the more recent antenna element approach. The derived SCFs are used to form the Rayleigh correlated 3D channel model. All these aspects are put together to provide a mathematical framework for the design of elevation beamforming schemes in single-cell and multi-cell scenarios. Finally, this dissertation proposes to use the double scattering channel to model limited scattering in realistic propagation environments and derives deterministic equivalents of the signal-to-interference-plus-noise ratio (SINR) and ergodic rate with regularized zeroforcing (RZF) precoding. The performance of a massive MIMO system is shown to be limited by the number of scatterers. To this end, this dissertation points out future research directions

Cognitive radio technology provides a solution for the spectrum scarcity issue by allowing the unlicensed users which are the cognitive radio devices to share the licensed band with the licensed (primary) users. The abilities of cognitive radio device help the secondary (unlicensed) nodes to observe the licensed band and to adjust their transmission parameters for maintaining the primary communication since in cognitive radio networks, it is essential that the existence of the unlicensed users must not harm the licensed network. Under these circumstances, we consider vertical cooperative transmission where primary transmission is not severely damaged by the secondary interference since each primary pair (transmitter and receiver) selects a neighbor secondary user as a cooperative relay to assist their transmission. Cooperation provides an increase in the signal-tointerference-ratio (SIR) of the primary network which can be harnessed by the secondary network as an additional bandwidth for their communication. We propose three relay selection rules so the influence of the relay’s position over the temporal and spatial correlations can be evaluated for different network conditions. Additionally, we implement primary exclusive region (PER) for each primary pair in the network which covers primary nodes, and all secondary users locate inside the zone become inactive. According to the outage model proposed in paper [2], temporal and spatial correlations are assumed to be 1 and 0 respectively; however it is estimated that regarding the environmental factors and the relay’s location, these correlation values may vary. This thesis work is based on the validation of the assumptions provided in paper [2] and our results demonstrate that the temporal and spatial correlation values changes under different circumstances and with different relay selection rules. The simulation results also show that PER significantly stimulates the cooperation performance thereby increases the transmission quality of the primary network.

While multivariate data analysis is concerned with data in the form of random vectors, functional data analysis goes one big step farther, focusing on data that are infinite-dimensional, such as curves, shapes and images. We focus on functional data that are measured over time across multiple subjects. The first part of the thesis focuses on spatially correlated functional data. This correlation is modeled by correlating functional principal component scores. We propose a Spatial Principal Analysis by Conditional Expectation framework to explicitly estimate spatial correlations and reconstruct individual curves. This approach works even when the observed data per curve are extremely sparse. Assuming spatial stationarity, empirical between-curve correlations are calculated as the ratio of eigenvalues of the smoothed covariance surface Cov(Xi(s),Xi(t)) and cross-covariance surface Cov(Xi(s),Xj(t)). Then a parametric spatial correlation model is employed to fit empirical correlations. Finally, principal component scores are estimated to reconstruct the sparsely observed curves. This framework could naturally accommodate arbitrary covariance structures, but there is an enormous reduction in computation if one can assume the separability of temporal and spatial components. We propose hypothesis tests to examine the separability and isotropy effect of spatial correlation. Simulation studies and applications of empirical data show improvements in the curve reconstruction using our framework over the method where curves are assumed to be independent. In addition, asymptotic properties of estimates are discussed in details. In the second part of this work, we present a new approach to factor rotation for functional data. This is achieved by rotating the functional principal components toward a predefined space of periodic functions designed to decompose the total variation into components that are nearly-periodic and nearly-aperiodic with a predefined period. We show that the factor rotation can be obtained by the calculation of canonical correlations between appropriate spaces. Moreover, we demonstrate that our proposed rotations provide stable and interpretable results in the presence of highly complex covariance. This work is motivated by the goal of finding interpretable sources of variability in a gridded time series of vegetation index measurements obtained from remote sensing, and we demonstrate our methodology through the application of factor rotation of this data.

This dissertation addresses specific issues on domestic and foreign trade in India. The three chapters of the dissertation are summarized as follows. In the first chapter, we analyze the existence of regional wage differences across Indian states, and how domestic trade affects those premiums. We follow a two-step estimation process used in the literature on Labor Economics. Our empirical results show that higher level of domestic imports tends to reduce the state premiums, and higher domestic exports increase those premiums, which is consistent with a specific factor Ricardo-Viner model. Thus, promoting domestic trade by with states specializing in certain industries may lead to higher welfare within the country. In addition, we find, state premiums depend negatively on state-level amenities measured by per-capita power availability, and does not depend on the richness of the State measured by per-capita Net State Domestic Product. In the second chapter, we look at the pattern and determinants of inter-state manufacturing trade in India. In the paper, we use information on 12 manufacturing industries categorized under 5 sectors from 2005 to 2013 with two-year intervals in between. We find that a 1\% decrease in income ratio between importing state net state domestic product and exporting state net domestic product has significantly varying effects on trade flows across the different sectors. For coal and minerals, the effect is 36.8%, for chemical it is 105%, for metals it is 31.5% and for cement, it is 36.8%. In all these case a decrease in income ratio increases exports. For machinery, a 1% decrease in income ratio lead to approximately 9.3% reduction in trade. This suggests that machineries which are capital goods are more imported by richer states, whereas the other goods which can be classified as intermediate inputs are more imported by poorer states. We also find that infrastructure promotes trade and on average infrastructure reduces the effect of contiguity by around 28.6% and promotes trade even between non-contiguous states. Therefore, infrastructure in the form of roadways, highways, and railways must be built and maintained to promote facilitate trade in India. In chapter three, we compare the effects of tariffs and non-tariff barriers on Indian exports. We use Indian HS-96 four-digit industry level export data from COMTRADE and tariff data from TRAINS database for the study. The overall result suggests that input tariffs have the largest effect on exports, followed by final tariff and foreign tariffs. A 1% reduction in input tariff leads to around 8.6% increase in exports. A similar reduction in final tariffs and foreign tariffs lead to 3.6% and 2.8% increase, respectively in exports. Thus, we conclude that the supply side effect of exports dominates the demand side effects. From a policy perspective, if countries try to improve trade balance by imposing high tariffs, it may lead to a negative effect on exports through the input tariff effects.

Made available in DSpace on 2017-07-10T19:25:11Z (GMT). No. of bitstreams: 1 Ricardo.pdf: 1705247 bytes, checksum: 435e22f6b6cf7bf51e60d35de482adba (MD5) Previous issue date: 2012-04-09 / The agricultural production area has important information to understand the response of a crop yield. Precision agriculture helps the farmers on acquiring and treating in correct way the variations found in the area. Management zones (MZ) provide the precision agriculture implementation in a viable and relatively economical way, since they represent homogeneous regions, which can be treated differently. Thus, this trial aimed at generating MZ with different numbers of variables using fuzzy C-means algorithm, based on attributes spatially correlated with crop yield. The objective was to evaluate the relative efficiency of each management zone and verify if the different ways of variable compositions differ in the generation of MZ. Fifteen designs of MZ were generated and divided from two to five classes, based on attributes as: copper, silt, clay and altitude, which were correlated with crop yield. The copper variable showed the highest spatial correlation with crop yield, which promoted the generation of MZ with good results of relative efficiency and good separation of data among the classes. The relative efficiency did not present in an orderly manner, increasing or decreasing according to the number of variables used in division of MZ. The delineated MZ was based on variables as copper and clay and divided into five classes, so, it showed the highest relative efficiency / A área de produção agrícola contém informações importantes para se entender a resposta quanto à produtividade de uma cultura. A agricultura de precisão proporciona aos agricultores adquirirem e tratarem as variações encontradas na área de forma adequada. Unidades de manejo (UM) proporcionam a implantação da agricultura de precisão de forma viável e relativamente mais econômica, representando regiões homogêneas, que podem ser tratadas de forma diferenciada. Este trabalho consistiu em gerar unidades de manejo com diferentes números de variáveis através do algoritmo fuzzy C-means, baseando-se em atributos correlacionados espacialmente com a produtividade. O objetivo é avaliar a eficiência relativa de cada unidade de manejo e verificar se os diferentes modos de composição de variáveis se diferenciam na geração de UM. Geraram-se 15 delineamentos de UM, divididos de 2 a 5 classes, baseando-se nos atributos cobre, silte, argila e altitude, que foram correlacionados com a produtividade. A variável cobre teve a maior correlação espacial com a produtividade, a qual proporcionou a geração de UM com bons resultados de eficiência relativa e boa separação dos dados entre as classes. A eficiência relativa não se apresentou de forma ordenada, crescente ou decrescente, quanto ao número de variáveis utilizadas no delineamento em UM. A UM delineada a partir das variáveis Cu e argila, dividida em cinco classes, foi a que apresentou maior eficiência relativa

Made available in DSpace on 2017-05-12T14:48:35Z (GMT). No. of bitstreams: 1 Ricardo.pdf: 1705247 bytes, checksum: 435e22f6b6cf7bf51e60d35de482adba (MD5) Previous issue date: 2012-04-09 / The agricultural production area has important information to understand the response of a crop yield. Precision agriculture helps the farmers on acquiring and treating in correct way the variations found in the area. Management zones (MZ) provide the precision agriculture implementation in a viable and relatively economical way, since they represent homogeneous regions, which can be treated differently. Thus, this trial aimed at generating MZ with different numbers of variables using fuzzy C-means algorithm, based on attributes spatially correlated with crop yield. The objective was to evaluate the relative efficiency of each management zone and verify if the different ways of variable compositions differ in the generation of MZ. Fifteen designs of MZ were generated and divided from two to five classes, based on attributes as: copper, silt, clay and altitude, which were correlated with crop yield. The copper variable showed the highest spatial correlation with crop yield, which promoted the generation of MZ with good results of relative efficiency and good separation of data among the classes. The relative efficiency did not present in an orderly manner, increasing or decreasing according to the number of variables used in division of MZ. The delineated MZ was based on variables as copper and clay and divided into five classes, so, it showed the highest relative efficiency / A área de produção agrícola contém informações importantes para se entender a resposta quanto à produtividade de uma cultura. A agricultura de precisão proporciona aos agricultores adquirirem e tratarem as variações encontradas na área de forma adequada. Unidades de manejo (UM) proporcionam a implantação da agricultura de precisão de forma viável e relativamente mais econômica, representando regiões homogêneas, que podem ser tratadas de forma diferenciada. Este trabalho consistiu em gerar unidades de manejo com diferentes números de variáveis através do algoritmo fuzzy C-means, baseando-se em atributos correlacionados espacialmente com a produtividade. O objetivo é avaliar a eficiência relativa de cada unidade de manejo e verificar se os diferentes modos de composição de variáveis se diferenciam na geração de UM. Geraram-se 15 delineamentos de UM, divididos de 2 a 5 classes, baseando-se nos atributos cobre, silte, argila e altitude, que foram correlacionados com a produtividade. A variável cobre teve a maior correlação espacial com a produtividade, a qual proporcionou a geração de UM com bons resultados de eficiência relativa e boa separação dos dados entre as classes. A eficiência relativa não se apresentou de forma ordenada, crescente ou decrescente, quanto ao número de variáveis utilizadas no delineamento em UM. A UM delineada a partir das variáveis Cu e argila, dividida em cinco classes, foi a que apresentou maior eficiência relativa

This thesis considers resource management issues in wireless sensor networks (WSNs), wireless local area networks (WLANs), and cognitive radio (CR) networks. Since energy is a critical resource in WSNs, we consider energy minimization techniques based on explicit node cooperation and distributed source coding (DSC). The explicit node cooperation based on space time block codes (STBC) improves energy efficiency of WSNs, by reducing the energy consumption per bit of each sensor node. The DSC on the other hand exploits the spatial correlation in WSNs, and thus reduces the data generated in a WSN. For the purpose of our analysis, we model the spatial correlation according to a linear Gauss-Markov model. Through our numerical results, we observe that the node cooperation combined with DSC can improve energy efficiency for many cases of interest. A unique aspect of our work is we obtain important structural results using the concepts from monotone comparative statics. These structural results provide insights into the general design of WSNs. Through our numerical results, we also demonstrate that, the cooperation based transmission can achieve better mutual information (MI)-energy tradeoff than the non-cooperation based transmission scheme. From the perspective of WLANs, we propose a price based approach to regulate the channel occupancy of low rate users, which is known to be the primary cause for low overall throughput in WLANs. Owing to the decentralized nature of WLANs we use non-cooperative game theory as a tool for analysis. Specifically, we use supermodular game theory. Through our analysis, we show that an increase in price leads to an increase in rate of WLAN users. We also prove that the best response dynamics indeed converge to the Nash equilibrium of the underlying non-cooperative game. Through our numerical results, we demonstrate that by proper tuning of the price, the proposed price based approach can lead to an improvement in overall throughput of a WLAN. Finally from the perspective of CR networks, we consider the impact of number of channels captured by a secondary user on its transmission control protocol (TCP) throughput. From our simulation results it was found that, there exists a definite optimal number of channels a secondary user needs to capture, to maximize its TCP throughput.