Thesis (M.S.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Electrical Engineering, 2008. / Includes bibliographical references.

Thesis (Ph. D.)--Auburn University, 2008. / Abstract. Includes bibliographical references (p. 119-122).

Thesis (Ph. D.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Electrical and Computer Engineering, 2009. / Includes bibliographical references.

In many machine learning problems and application domains, the data are naturally organized by groups. For example, a video sequence is a group of images, an image is a group of patches, a document is a group of paragraphs/words, and a community is a group of people. We call them the collective data. In this thesis, we study how and what we can learn from collective data. Usually, machine learning focuses on individual objects, each of which is described by a feature vector and studied as a point in some metric space. When approaching collective data, researchers often reduce the groups into vectors to which traditional methods can be applied. We, on the other hand, will try to develop machine learning methods that respect the collective nature of data and learn from them directly. Several different approaches were taken to address this learning problem. When the groups consist of unordered discrete data points, it can naturally be characterized by its sufficient statistics – the histogram. For this case we develop efficient methods to address the outliers and temporal effects in the data based on matrix and tensor factorization methods. To learn from groups that contain multi-dimensional real-valued vectors, we develop both generative methods based on hierarchical probabilistic models and discriminative methods using group kernels based on new divergence estimators. With these tools, we can accomplish various tasks such as classification, regression, clustering, anomaly detection, and dimensionality reduction on collective data. We further consider the practical side of the divergence based algorithms. To reduce their time and space requirements, we evaluate and find methods that can effectively reduce the size of the groups with little impact on the accuracy. We also proposed the conditional divergence along with an efficient estimator in order to correct the sampling biases that might be present in the data. Finally, we develop methods to learn in cases where some divergences are missing, caused by either insufficient computational resources or extreme sampling biases. In addition to designing new learning methods, we will use them to help the scientific discovery process. In our collaboration with astronomers and physicists, we see that the new techniques can indeed help scientists make the best of data.

L'utilisation des réseaux de capteurs sans fil (WSN) ne cesse d'augmenter au point de couvrir divers domaines et applications. Cette tendance est supportée par les avancements techniques achevés dans la conception des capteurs, qui ont permis de réduire le coût ainsi que la taille de ces composants. Toutefois, il reste plusieurs défis qui font face au déploiement et au bon fonctionnement de ce type de réseaux et qui parviennent principalement de la limitation des ressources de capteurs ainsi de l'imperfection des données collectées. Dans cette thèse, on adresse le problème de collecte de données et de détection d'anomalies dans les réseaux de capteurs. Nous visons à assurer ces deux fonctionnalités tout en économisant l'utilisation des ressources de capteurs et en prolongeant la durée de vie de réseaux. Tout au long de ce travail, nous présentons plusieurs solutions qui permettent une collecte efficace de données de capteurs ainsi que une bonne détection des éventuelles anomalies. Dans notre première contribution, nous décrivons une solution basée sur la technique Compressive Sensing (CS) qui permet d'équilibrer le trafic transmis par les nœuds dans le réseau. Notre approche diffère des solutions existantes par la prise en compte de la corrélation temporelle ainsi que spatiale dans le processus de décompression des données. De plus, nous proposons une nouvelle formulation pour détecter les anomalies. Les simulations réalisées sur des données réelles prouvent l'efficacité de notre approche en termes de reconstruction de données et de détection d'anomalies par rapport aux approches existantes. Pour mieux optimiser l'utilisation des ressources de WSNs, nous proposons dans une deuxième contribution une solution de collecte de données et de détection d'anomalies basée sur la technique Matrix Completion (MC) qui consiste à transmettre un sous ensemble aléatoire de données de capteurs. Nous développons un algorithme qui estime les mesures manquantes en se basant sur plusieurs propriétés des données. L'algorithme développé permet également de dissimuler les anomalies de la structure normale des données. Cette solution est améliorée davantage dans notre troisième contribution, où nous proposons une formulation différente du problème de collecte de données et de détection d'anomalies. Nous reformulons les connaissances a priori sur les données cibles par des contraintes convexes. Ainsi, les paramètres impliqués dans l'algorithme développé sont liés a certaines propriétés physiques du phénomène observé et sont faciles à ajuster. Nos deux approches montrent de bonnes performances en les simulant sur des données réelles. Enfin, nous proposons dans la dernière contribution une nouvelle technique de collecte de données qui consiste à envoyer que les positions les plus importantes dans la représentation parcimonieuse des données uniquement. Nous considérons dans cette approche le bruit qui peut s'additionner aux données reçues par le nœud collecteur. Cette solution permet aussi de détecter les pics dans les mesures prélevées. En outre, nous validons l'efficacité de notre solution par une analyse théorique corroborée par des simulations sur des données réelles / The use of Wireless Sensor Networks (WSN)s is steadily increasing to cover various applications and domains. This trend is supported by the technical advancements in sensor manufacturing process which allow a considerable reduction in the cost and size of these components. However, there are several challenges facing the deployment and the good functioning of this type of networks. Indeed, WSN's applications have to deal with the limited energy, memory and processing capacities of sensor nodes as well as the imperfection of the probed data. This dissertation addresses the problem of collecting data and detecting anomalies in WSNs. The aforementioned functionality needs to be achieved while ensuring a reliable data quality at the collector node, a good anomaly detection accuracy, a low false alarm rate as well as an efficient energy consumption solution. Throughout this work, we provide different solutions that allow to meet these requirements. Foremost, we propose a Compressive Sensing (CS) based solution that allows to equilibrate the traffic carried by nodes regardless their distance from the sink. This solution promotes a larger lifespan of the WSN since it balances the energy consumption between sensor nodes. Our approach differs from existing CS-based solutions by taking into account the sparsity of sensory representation in the temporal domain in addition to the spatial dimension. Moreover, we propose a new formulation to detect aberrant readings. The simulations carried on real datasets prove the efficiency of our approach in terms of data recovering and anomaly detection compared to existing solutions. Aiming to further optimize the use of WSN resources, we propose in our second contribution a Matrix Completion (MC) based data gathering and anomaly detection solution where an arbitrary subset of nodes contributes at the data gathering process at each operating period. To fill the missing values, we mainly relay on the low rank structure of sensory data as well as the sparsity of readings in some transform domain. The developed algorithm also allows to dissemble anomalies from the normal data structure. This solution is enhanced in our third contribution where we propose a constrained formulation of the data gathering and anomalies detection problem. We reformulate the textit{a prior} knowledge about the target data as hard convex constraints. Thus, the involved parameters into the developed algorithm become easy to adjust since they are related to some physical properties of the treated data. Both MC based approaches are tested on real datasets and demonstrate good capabilities in terms of data reconstruction quality and anomaly detection performance. Finally, we propose in the last contribution a position based compressive data gathering scheme where nodes cooperate to compute and transmit only the relevant positions of their sensory sparse representation. This technique provide an efficient tool to deal with the noisy nature of WSN environment as well as detecting spikes in the sensory data. Furthermore, we validate the efficiency of our solution by a theoretical analysis and corroborate it by a simulation evaluation

In this thesis we consider concentration inequalities and the concentration of measure phenomenon from a variety of angles. Sharp tail bounds on the deviation of Lipschitz functions of independent random variables about their mean are well known. We consider variations on this theme for dependent variables on the Boolean cube. In recent years negatively associated probability distributions have been studied as potential generalizations of independent random variables. Results on this class of distributions have been sparse at best, even when restricting to the Boolean cube. We consider the class of negatively associated distributions topologically, as a subset of the general class of probability measures. Both the weak (distributional) topology and the total variation topology are considered, and the simpler notion of negative correlation is investigated. The concentration of measure phenomenon began with Milman's proof of Dvoretzky's theorem, and is therefore intimately connected to the field of high-dimensional convex geometry. Recently this field has found application in the area of compressed sensing. We consider these applications and in particular analyze the use of Gordon's min-max inequality in various compressed sensing frameworks, including the Dantzig selector and the matrix uncertainty selector. Finally we consider the use of concentration inequalities in developing a theoretically sound anomaly detection algorithm. Our method uses a ranking procedure based on KNN graphs of given data. We develop a max-margin learning-to-rank framework to train limited complexity models to imitate these KNN scores. The resulting anomaly detector is shown to be asymptotically optimal in that for any false alarm rate α, its decision region converges to the α-percentile minimum volume level set of the unknown underlying density.

abstract: Photovoltaics (PV) is an important and rapidly growing area of research. With the advent of power system monitoring and communication technology collectively known as the "smart grid," an opportunity exists to apply signal processing techniques to monitoring and control of PV arrays. In this paper a monitoring system which provides real-time measurements of each PV module's voltage and current is considered. A fault detection algorithm formulated as a clustering problem and addressed using the robust minimum covariance determinant (MCD) estimator is described; its performance on simulated instances of arc and ground faults is evaluated. The algorithm is found to perform well on many types of faults commonly occurring in PV arrays. Among several types of detection algorithms considered, only the MCD shows high performance on both types of faults. / Dissertation/Thesis / M.S. Electrical Engineering 2012

A ionosfera é uma das principais fontes de erro sistemático das observáveis GPS (Global Positioning System - Sistema de Posicionamento Global), pois, por ser um meio dispersivo, ela afeta a propagação de ondas eletromagnéticas, fazendo com que a modulação e a fase das ondas portadoras transmitidas pelos satélites GPS sofram, respectivamente, um retardo e um avanço, o que, por sua vez, provoca um erro na distância medida entre o satélite e o receptor. Esse erro é inversamente proporcional ao quadrado da freqüência do sinal e diretamente proporcional ao TEC (Total Electron Content - Conteúdo Total de Elétrons), ou seja, à densidade de elétrons presentes na ionosfera ao longo do caminho entre o satélite e a antena receptora. O TEC sofre variações regulares, cujo comportamento pode ser verificado ao longo do dia, ao longo das estações do ano e também ao longo de ciclos de aproximadamente onze anos (associados à ocorrência de manchas solares). Além dessas variações, eventos solares extremos (explosões solares, ejeções coronais de massa, entre outros) podem causar abruptas e significativas mudanças no comportamento do TEC, exercendo grande influência no posicionamento com GPS, principalmente com receptores de uma freqüência. No Brasil, o fator ionosfera é ainda mais relevante, pois essa região é afetada por fenômenos como a Anomalia Equatorial (AE), a Anomalia Magnética do Atlântico Sul (AMAS) e até mesmo pela ocorrência de irregularidades ionosféricas. Pretendendo aprofundar o entendimento da relação entre a ionosfera e o posicionamento com GPS na região brasileira, essa pesquisa analisou dados de TEC e dados de GPS em períodos de alta e baixa atividade solar, bem como em um período geomagneticamente perturbado. Os resultados demonstraram uma relação direta entre a redução do TEC, no período de baixa atividade solar, e a melhora no posicionamento com GPS. Essa melhora se traduziu, no posicionamento por ponto, por uma redução de 59% no erro planimétrico e 64% no erro altimétrico e, no posicionamento relativo, por uma redução de 65% no erro planimétrico e 63% no erro altimétrico. Já durante o período afetado por uma severa tempestade geomagnética verificou-se um comportamento completamente atípico da ionosfera, piorando muitos os resultados do posicionamento relativo, em horários e locais inesperados. / The ionosphere is one of the main sources of systemathic error of the observable GPS (Global Positioning System) because as it is a dispersive environment it affects the propagation of electromagnetics waves making the modulation and the phase of signals transmitted by GPS sattelites go through, respectivelly, delay and advance which will cause an error in the measure of the distance between the sattelite and the receptor. This error is inversely proportional to the square of the frequency of the signal and directly proportional to the TEC (Total Electron Content), what means the density of electrons on the ionosphere between the sattelite and the reception antenna. The TEC goes through regular variances, which behaviour can be verified during the day, throughout seasons and also throughout cycles of approximately eleven years (related to the ocorrence of sunspot). Besides these variances, extreme solar events such as solar flares and coronal mass ejection may cause abrupt and significant changes to TEC behavior, exerting big influence in GPS positioning, mainly to monofrequency receptors. In Brazil, the ionosphere factor is even more relevant because this region is affected by phenomena such as the Equatorial Anomaly (EA), the South Atlantic Magnetic Anomaly (SAMA) and even by the ocorrence of ionospheric irregularities. In order to develop knowledge about the relation between ionosphere and GPS positioning in Brazil, on this research TEC and GPS data were analised in periods of high and low solar activity, as well as in a geomagnetic perturbed period. The results showed direct relation between the decreasing of TEC, in the low solar activity period, and the improving of GPS positioning. This improving has resulted in a reduction of 59% in the planimetric error and 64% in the altimetric error in the point positioning and a reduction of 65% in the planimetric error and 63% in the altimetric error in the relative positioning. During the period affected by a severe geomagnetic storm, a completely atypical behavior was identified in the ionosphere, making the results of the relative positioning much worse in unexpected times and locations.

Utilizando simulação de dinâmica molecular estudamos as propriedades termodin âmicas da água em um sistema com macromoléculas hidrofóbicas e hidrofílicas. Em um primeiro momento utilizamos um sistema com uma proteína imersa em água. Os dois modelos atomísticos utilizados foram o modelo SPC/E e o modelo TIP4P-2005 e a proteína utilizada foi a toxina oriunda do escorpião brasileiro, TS-Kappa. Observamos para o sistema bulk e para o sistema com a proteína hidratada um máximo valor de densidade. Porém, analisando a densidade próximo à superfície da proteína, o comportamento anômalo não está mais presente. Para baixas temperaturas densidade da água próximo à proteína é maior do que no bulk. Nossos resultados mostram que o número de ligações de hidrogênio entre as moléculas de água na camada de hidratação da proteína, é menor que o número de ligações de hidrogênio para o sistema puro. A água na primeira camada de hidratação conecta-se com a superfície da proteína, além de ligar-se com outras moléculas de água vizinhas. medida que a temperatura diminui as moléculas de água tornam-se mais estruturadas próximo à região hidrofílica, e a densidade da água nessa região é maior do que a densidade da água na região hidrofóbica. Os resultados para a difusão nos mostram que a água diminui sua mobilidade na superfície da proteína, indicando uma maior conexão com sua superfície. Com o objetivo de estudar o comportamento da água em um sistema com interações hidrofílicas e hidrofóbicas, usamos um modelo mínimo com duas placas paralelas imersas em um uido do tipo-água. Nossos resultados indicam que a densidade da água aumenta com a diminuição da temperatura e a água torna-se mais estruturada a medida que a temperatura diminui. / Using a molecular dynamics simulation we studied the thermodynamic properties of water in a system with a hydrophilic and hydrophobic macromolecule. At rst, we used a system with a protein immersed in water. The models used were the SPC/E and the TIP4P-2005 water model and the chosen protein was the Brazilian scorpion toxin TS-Kappa. For bulk system and the protein hydrated system, the maximum value of density is present. However, the density of water near to the protein surface, the anomalous behavior is no longer veri ed. At lower temperatures the density of water near to protein surface is higher than the bulk. Our results show that the number of hydrogen bonds between water molecules in the hydration shell is lower than the number of hydrogen bonds in bulk water. The water in the rst hydration shell connect with the hydrophilic protein surface besides the hydrogen bonds with other water molecules. As the temperature is decreased the water is more structured near the hydrophilic amino acid and the density in this region is higher than the density of water in the hydrophobic region. The di usion values of water shows that hydration water is more connected with the protein surface, so the di usion coe cient decreases in this region. In order to investigate the behavior of water in a hydrophobic and a hydrophilic system, we used a simple model of parallel plates immersed in a water-like uid. Our results suggests the density of water increases with the decrease of temperature and the water is more structured as the temperatures is decreased.

A origem e evolução da Bacia de Pelotas está diretamente relacionada com os processos tectônicos, que por sua vez, condicionam vários processos sedimentares. Assim, o estudo da compartimentação morfoestrutural é extremamente importante no seu entendimento evolutivo. O objetivo principal foi o levantamento gravimétrico terrestre da área de estudo, seu processamento e interpretação para posterior identificação de suas principais estruturas, assim como investigar a configuração morfoestrutural da porção central da Planície Costeira do Rio Grande do Sul. O método do potencial gravimétrico (gravimetria) é amplamente usado na prospecção mineral, análise de bacias, e mapeamentos geológicos por causa do seu baixo custo e rapidez nos resultados e fornecem informação quanto a geologia e a delimitação de estruturas e descontinuidades geológicas, além de fornecer importantes informações a respeito do embasamento subjacente no caso de bacias sedimentares. A área de estudo escolhida para a aplicação da metodologia proposta compreende a parte emersa da Bacia de Pelotas, na região central da Planície Costeira do Rio Grande do Sul, aproximadamente entre as latitudes 30° e 32°10’S e as longitudes 50°40’ e 52°40’W; situa-se em uma área que abrange os municípios costeiros do litoral médio do Rio Grande do Sul entre os municípios de São José do Norte e Palmares do Sul. O principal resultado gerado foi mapas de anomalia Bouguer e seus derivados identificando quatro setores de anomalias distintas, evidenciados por altos e baixos gravimétricos e sua possível correlação com as feições estruturais do embasamento geradas ou reativadas por rifteamento proveniente da origem da Bacia de Pelotas. Futuramente, outros métodos geofísicos podem ser utilizados e integrados para contribuir com o modelo proposto neste estudo. / The origin and evolution of the Pelotas Basin is directly related to tectonic processes, which in turn affect various sedimentary processes. Thus, the study of morphostructural compartments is extremely important in its evolution understanding. The main objective was the terrestrial gravimetric survey of the study area, its processing and interpretation for later identification of its main structures, as well as to investigate the morphostructural configuration of the middle portion of coastal plain Rio Grande do Sul. The method of gravity potential (gravity), is widely used in mineral prospecting, basin analysis, and geological mapping because of its low cost and speed of results and provide information about geology and delineation of structures and geological discontinuities, and provide essential information about the underlying basis in the case of sedimentary basins. The study area chosen for the implementation of the proposed methodology comprises the emerged part of the Pelotas Basin, in central Rio Grande do Sul coastal province, between latitudes 30° and 32°10'S and longitudes 50°4' and 52°40'W, in an area covering the coastal municipalities of the middle coast of Rio Grande do Sul State between the cities of São José do Norte and Palmares do Sul. The main results generated were Bouguer anomaly maps and derivatives by identifying four sectors of distinct anomalies highlighted by high and low gravity and its possible correlation with the structural features of the basement generated or reactivated by rifting related to Pelotas Basin origin. In the future, other geophysical methods can be used and integrated to contribute to the model suggested in this study.