In this thesis, a system to locate an electrostatic discharge (ESD) event within an electronic device has been developed. ESD can cause a device to fail legally required radiated emissions limits as well as disrupt intended operation. The system used a fast oscilloscope with four channels, each channel attached to a high frequency near-field antenna. These antennas were placed at known locations in three dimensional space to measure the fields radiated from the ESD event. A Time-Difference-of-Arrival technique was used to calculate the location of the ESD event. Quick determination of the ESD event location provides developers with a tool that saves them time and money by eliminating the time-consuming and tedious method of general ESD mitigation within a product.

Power transformers are one of the most important components in the electrical energy network. Extending transformer life is very economically valuable due to power outage. Therefore the development of instruments to monitor the transformer condition is of great interest. Detection of partial discharges (PDs) in power transformers is an effective diagnostic because it may reveal and quantify an important aging factor and provide information on the condition of the transformer. However, partial discharge diagnostics are still not effectively used for online monitoring of transformers because of the complexity of PD measurements and difficulties of discriminating of PDs and other noise sources. This thesis presents a further study of detection and location of partial discharges in power transformers based on previous work conducted at the Center for Photonics Technology (CPT) at Virginia Tech. The detection and positioning system consists of multiple extrinsic Fabry-Parot interferometric (EFPI) fiber acoustic sensors which can survive the harsh environment of oil-filled transformers. This thesis work is focused on optimal arrangement of multiple sensors to monitor and locate PD activities in a power transformer. This includes the following aspects. First, the sensor design requirements are discussed in order to successfully detect and accurately position the PD sources. In the following sections, Finite Element Method (FEM) is used to model the EFPI sensor fabricated at CPT. Experiments were conducted to measure the angular dependence of the frequency response of the sensor. It is shown that within the range of ±45º incident angles, the sensitivity varies by 3-5dB. Finally, the thesis demonstrates a PD positioning experiment in a 500 gallon water tank (R à H = 74" à 30" cylinder) using a hyperbolic positioning algorithm and time difference of arrival (TDOA). Finally we demonstrated that 100% of the positioning data is bounded by a 22.7à 4.1à 5.3 mm₃ cube, with a sensing range of 810 mm using the leading edge method with FIR filtering. / Master of Science

A partial discharge (PD) is the dissipation of energy caused by the buildup of localized electric field intensity. In high voltage devices such as transformers, this buildup of charge and its release can be symptomatic of problems associated with aging, such as floating components and insulation breakdown. This is why PD detection is used in power systems to monitor the state of health of high voltage transformers. If such problems are not detected and repaired, the strength and frequency of PDs increases and eventually leads to the catastrophic failure of the transformer, which can cause external equipment damage, fires and loss of revenue due to an unscheduled outage. Reliable online PD detection is a critical need for power companies to improve personnel safety and decrease the potential for loss of service. The PD phenomenon is manifested in a variety of physically observable signals including electric and acoustic pulses and is currently detected using a host of exterior measurement techniques. These techniques include electrical lead tapping and piezoelectric transducer (PZT) based acoustic detection. Many modern systems use a combination of these techniques because electrical detection is an older and proven technology and acoustic detection allows for the source to be located when several sensors are mounted to the exterior of the tank. However, if an acoustic sensor could be placed inside the tank, not only would acoustic detection be easier due to the increased signal amplitude and elimination of multipath interference, but positioning could also be performed with more accuracy in a shorter time. This thesis presents a fiber optic acoustic sensing system design that can be used to detect and locate PD sources within a high voltage transformer. The system is based on an optical acoustic (OA) sensor that is capable of surviving the harsh environment of the transformer interior while not compromising the transformer's functionality, which allows for online detection and positioning. This thesis presents the theoretical functionality and experimental validation of a band-limited OA sensor with a usable range of 100-300 kHz, which is consistent with the frequency content of an acoustic pulse caused by a PD event. It also presents a positioning system using the time difference of arrival (TDOA) of the acoustic pulse with respect to four sensors that is capable of reporting the three-dimensional position of a PD to within ±5cm on any axis. / Master of Science

The angular location of an acoustic source can be estimated by measuring an acoustic direction of incidence based solely on the noise produced by the source. Methods for determining the direction of incidence based on sound intensity, the phase of cross-spectral functions, and cross-correlation functions are available. In this current work, we implement Dominant Frequency SElection (DFSE) algorithm. Direction of arrival (DOA) estimation usingmicrophone arrays is to use the phase information present in signals from microphones that are spatially separated. DFSE uses the phase difference between the Fourier transformedsignals to estimate the direction ofarrival (DOA)and is implemented using a three-element ’L’ shaped microphone array, linear microphone array, and planar 16-microphone array. This method is based on simply locating the maximum amplitude from each of the Fourier transformed signals and thereby deriving the source location by solving the set of non-linear least squares equations. For any pair of microphones, the surface on whichthe time difference ofarrival (TDOA) is constant is a hyperboloidoftwo sheets. Acoustic source localization algorithms typically exploit this fact by grouping all microphones into pairs, estimating the TDOA of each pair, then finding the point where all associated hyperboloids most nearly intersect. We make use of both closed-form solutions and iterative techniques to solve for the source location.Acoustic source positioned in 2-dimensional plane and 3-dimensional space have been successfully located.

Autant le GPS tend à s'imposer pour la localisation à l'extérieur des bâtiments, autant la situation est beaucoup plus ouverte pour la localisation à l'intérieur des bâtiments. De nombreux réseaux WiFi sont déployés dans les bâtiments. Ils diffusent des informations de puissance du signal permettant de remonter à la position d'un mobile. La technique du fingerprinting par puissance WiFi permet de localiser le mobile. Cependant, l'utilisation de cette technique de localisation requière une base de données correspondant à la couverture radio WiFi dans l'environnement.<br />L'utilisation d'une technique de localisation basée sur des mesures temporelles est moins contraignante que le fingerprinting. L'émission d'impulsions radio très brèves confère à la technologie 802.15.4a un fort pouvoir séparateur des multi-trajets. Le phénomène de multi-trajets est la principale contrainte au déploiement d'une technologie de localisation par mesures temporelles. La détection du premier trajet est très importante.<br />Des estimateurs comme le filtre de Kalman ou le filtre particulaire sont nécessaires pour limiter les effets des multi-trajets, des bruits de mesure, etc. Ces filtres peuvent aussi intégrer des informations de cartographie. Bien souvent, l'exploitation d'une seule technologie est insuffisante. La fusion d'informations de localisation est une étape supplémentaire pour améliorer la localisation. Des architectures de fusion robustes permettent de corriger les défauts de chacune des technologies pour conduire à un système plus robuste et plus précis en toutes circonstances.<br />Ce travail présente une approche innovante pour la localisation WiFi avec l'exploitation de cartographie dans l'estimateur tout en gardant une faible complexité suivant la plate-forme de déploiement visée. L'exploration des capacités de la localisation par ULB est proposée dans un second temps, avant d'aborder une réflexion sur les méthodes de fusion multi-capteurs.

[pt] O problema de localização de estações móveis pessoais em sistemas celulares de comunicações vem recebendo grande atenção nos últimos anos, tanto por questões ligadas à segurança como por suas amplas aplicações comerciais no desenvolvimento de novos serviços e aplicações. Nesta dissertação foi desenvolvido um ambiente de simulação de localização de estações móveis em ambiente micro celulares empregando um programa de traçado de raios pelo método da força bruta (lançamento de raios), já disponível, para estimar os comprimentos de percursos e tempos de chegada entre diversas estações rádio base e a estação móvel em cenários urbanos modelados por sólidos multifacetados. Os perfis de retardo gerados por este programa são usados como dados de entrada para um programa desenvolvido nesta dissertação que estima a localização dos móveis utilizando os métodos de Taylor e de Chan. O processo desenvolvido foi testado em ambientes de geometria simples fornecendo resultados bastante consistentes e mostrando que a técnica de traçado de raios é uma ferramenta útil para a simulação e desenvolvimento de algoritmos de localização, cujo teste em situações reais exige grande volume de medidas de alta complexidade cujos exemplos na literatura técnica são escassos. Com base nas simulações foi investigadas a influência do número de estações rádio base na precisão das estimativas de localização e realizada uma comparação do desempenho dos métodos em situações com visibilidade (LOS) e sem visibilidade (NLOS). Foi analisado ainda o efeito da altura das estações na precisão dos resultados de localização. / [en] The location of mobile terminals in mobile cellular systems has been receiving increasing attention in the last few years. This interest in focused not only in security aspects but also in the development of new services for commercial application. In this Dissertation a simulation environment for mobile stations location in microcellular systems was developed. The simulation tools include a ray tracing software, previously implemented using the ray launching technique, to estimate the path lengths and time of arrival of signals from the mobile station to several radio base stations, and new software implementing terminal location methods using the Taylor linearization and the Chan methods. The simulation tools were tested in scenarios of simple geometry producing consistent results and showing that ray tracing can be a useful tool for simulation and development of location algorithms. The simulations allowed the investigation of location precision dependence on the number of radio bases employed and the evaluation of the estimation methods in visibility (LOS) and non- visibility (NLOS) conditions. The influence of base station antennas heights was also investigated.

In speech communication systems such as voice-controlled systems, hands-free mobile telephones and hearing aids, the received signals are degraded by room reverberation and background noise. This degradation can reduce the perceived quality and intelligibility of the speech, and decrease the performance of speech enhancement and source localization. These problems are difficult to solve due to the colored and nonstationary nature of the speech signals, and features of the Room Impulse Response (RIR) such as its long duration and non-minimum phase. In this dissertation, we focus on two topics of speech enhancement and speaker localization in noisy reverberant environments. A two-stage speech enhancement method is presented to suppress both early and late reverberation in noisy speech using only one microphone. It is shown that this method works well even in highly reverberant rooms. Experiments under different acoustic conditions confirm that the proposed blind method is superior in terms of reducing early and late reverberation effects and noise compared to other well known single-microphone techniques in the literature. Time Difference Of Arrival (TDOA)-based methods usually provide the most accurate source localization in adverse conditions. The key issue for these methods is to accurately estimate the TDOA using the smallest number of microphones. Two robust Time Delay Estimation (TDE) methods are proposed which use the information from only two microphones. One method is based on adaptive inverse filtering which provides superior performance even in highly reverberant and moderately noisy conditions. It also has negligible failure estimation which makes it a reliable method in realistic environments. This method has high computational complexity due to the estimation in the first stage for the first microphone. As a result, it can not be applied in time-varying environments and real-time applications. Our second method improves this problem by introducing two effective preprocessing stages for the conventional Cross Correlation (CC)-based methods. The results obtained in different noisy reverberant conditions including a real and time-varying environment demonstrate that the proposed methods are superior compared to the conventional TDE methods. / Graduate / 0544 / 0984 / saeed.mosayyebpour@gmail.com

In speech communication systems such as voice-controlled systems, hands-free mobile telephones and hearing aids, the received signals are degraded by room reverberation and background noise. This degradation can reduce the perceived quality and intelligibility of the speech, and decrease the performance of speech enhancement and source localization. These problems are difficult to solve due to the colored and nonstationary nature of the speech signals, and features of the Room Impulse Response (RIR) such as its long duration and non-minimum phase. In this dissertation, we focus on two topics of speech enhancement and speaker localization in noisy reverberant environments. A two-stage speech enhancement method is presented to suppress both early and late reverberation in noisy speech using only one microphone. It is shown that this method works well even in highly reverberant rooms. Experiments under different acoustic conditions confirm that the proposed blind method is superior in terms of reducing early and late reverberation effects and noise compared to other well known single-microphone techniques in the literature. Time Difference Of Arrival (TDOA)-based methods usually provide the most accurate source localization in adverse conditions. The key issue for these methods is to accurately estimate the TDOA using the smallest number of microphones. Two robust Time Delay Estimation (TDE) methods are proposed which use the information from only two microphones. One method is based on adaptive inverse filtering which provides superior performance even in highly reverberant and moderately noisy conditions. It also has negligible failure estimation which makes it a reliable method in realistic environments. This method has high computational complexity due to the estimation in the first stage for the first microphone. As a result, it can not be applied in time-varying environments and real-time applications. Our second method improves this problem by introducing two effective preprocessing stages for the conventional Cross Correlation (CC)-based methods. The results obtained in different noisy reverberant conditions including a real and time-varying environment demonstrate that the proposed methods are superior compared to the conventional TDE methods. / Graduate / 2015-04-23 / 0544 / 0984 / saeed.mosayyebpour@gmail.com

Speech is by far the most natural form of communication between human beings. It is intuitive, expressive and contains information at several cognitive levels. We as humans, are perceptive to several of these cognitive levels of information, as we can gather the information pertaining to the identity of the speaker, the speaker's gender, emotion, location, the language, and so on, in addition to the content of what is being spoken. This makes speech based human machine interaction (HMI), both desirable and challenging for the same set of reasons. For HMI to be natural for humans, it is imperative that a machine understands information present in speech, at least at the level of speaker identity, language, location in space, and the summary of what is being spoken. Although one can draw parallels between the human-human interaction and HMI, the two differ in their purpose. We, as humans, interact with a machine, mostly in the context of getting a task done more efficiently, than is possible without the machine. Thus, typically in HMI, controlling the machine in a specific manner is the primary goal. In this context, it can be argued that, HMI, with a limited vocabulary containing specific commands, would suffice for a more efficient use of the machine. In this thesis, we address the problem of ``Who spoke what and where", in the context of a machine understanding the information pertaining to identities of the speakers, their locations in space and the keywords they spoke, thus considering three levels of information - speaker identity (who), location (where) and keywords (what). This can be addressed with the help of multiple sensors like microphones, video camera, proximity sensors, motion detectors, etc., and combining all these modalities. However, we explore the use of only microphones to address this issue. In practical scenarios, often there are times, wherein, multiple people are talking at the same time. Thus, the goal of this thesis is to detect all the speakers, their keywords, and their locations in mixture signals containing speech from simultaneous speakers. Addressing this problem of ``Who spoke what and where" using only microphone signals, forms a part of acoustic scene analysis (ASA) of speech based acoustic events. We divide the problem of ``who spoke what and where" into two sub-problems: ``Who spoke what?" and ``Who spoke where". Each of these problems is cast in a generic latent variable (LV) framework to capture information in speech at different levels. We associate a LV to represent each of these levels and model the relationship between the levels using conditional dependency. The sub-problem of ``who spoke what" is addressed using single channel microphone signal, by modeling the mixture signal in terms of LV mass functions of speaker identity, the conditional mass function of the keyword spoken given the speaker identity, and a speaker-specific-keyword model. The LV mass functions are estimated in a Maximum likelihood (ML) framework using the Expectation Maximization (EM) algorithm using Student's-t Mixture Model (tMM) as speaker-specific-keyword models. Motivated by HMI in a home environment, we have created our own database. In mixture signals, containing two speakers uttering the keywords simultaneously, the proposed framework achieves an accuracy of 82 % for detecting both the speakers and their respective keywords. The other sub-problem of ``who spoke where?" is addressed in two stages. In the first stage, the enclosure is discretized into sectors. The speakers and the sectors in which they are located are detected in an approach similar to the one employed for ``who spoke what" using signals collected from a Uniform Circular Array (UCA). However, in place of speaker-specific-keyword models, we use tMM based speaker models trained on clean speech, along with a simple Delay and Sum Beamformer (DSB). In the second stage, the speakers are localized within the active sectors using a novel region constrained localization technique based on time difference of arrival (TDOA). Since the problem being addressed is a multi-label classification task, we use the average Hamming score (accuracy) as the performance metric. Although the proposed approach yields an accuracy of 100 % in an anechoic setting for detecting both the speakers and their corresponding sectors in two-speaker mixture signals, the performance degrades to an accuracy of 67 % in a reverberant setting, with a $60$ dB reverberation time (RT60) of 300 ms. To improve the performance under reverberation, prior knowledge of the location of multiple sources is derived using a novel technique derived from geometrical insights into TDOA estimation. With this prior knowledge, the accuracy of the proposed approach improves to 91 %. It is worthwhile to note that, the accuracies are computed for mixture signals containing more than 90 % overlap of competing speakers. The proposed LV framework offers a convenient methodology to represent information at broad levels. In this thesis, we have shown its use with three different levels. This can be extended to several such levels to be applicable for a generic analysis of the acoustic scene consisting of broad levels of events. It will turn out that not all levels are dependent on each other and hence the LV dependencies can be minimized by independence assumption, which will lead to solving several smaller sub-problems, as we have shown above. The LV framework is also attractive to incorporate prior knowledge about the acoustic setting, which is combined with the evidence from the data to derive the information about the presence of an acoustic event. The performance of the framework, is dependent on the choice of stochastic models, which model the likelihood function of the data given the presence of acoustic events. However, it provides an access to compare and contrast the use of different stochastic models for representing the likelihood function.

Wide Area Measurements (WAMs) have been widely used in the energy management system (EMS) of power system for monitoring, operation and control. In recent years, the advent of synchronized Phasor Measurements Unit (PMU) has added another dimension to the field of wide-area measurement. However, the high cost of the PMU, which includes the manufacture and deployment fee, is a hurdle to the wide use of the PMU in power systems. Unlike traditional PMUs, the frequency monitoring network (FNET) developed by the Virginia Tech Power IT lab is an Internet—based, GPS—synchronized, wide-area frequency monitoring network deployed at the distribution level, providing a low-cost and easily deployable WAMs solution. In this dissertation, the research work can be categorized into two parts: FNET Design and Situation Awareness Algorithm Development. / Ph. D.