Tesis sobre el tema "Sensors selection"

The past decade has seen great progress in research on large scale mapping and perception in static environments. Real world perception requires handling uncertain situations with multiple possible interpretations: e.g. changing appearances, dynamic objects, and varying motion models. These aspects of perception have been largely avoided through the use of heuristics and preprocessing. This thesis is motivated by the challenge of including discrete reasoning directly into the estimation process. We approach the problem by using Conditional Linear Gaussian Networks (CLGNs) as a generalization of least-squares estimation which allows the inclusion of discrete model selection variables. CLGNs are a powerful framework for modeling sparse multi-modal inference problems, but are difficult to solve efficiently. We propose the Iterative Local Model Selection (ILMS) algorithm as a general approximation strategy specifically geared towards the large scale problems encountered in tracking and mapping. Chapter 4 introduces the ILMS algorithm and compares its performance to traditional approximate inference techniques for Switching Linear Dynamical Systems (SLDSs). These evaluations validate the characteristics of the algorithm which make it particularly attractive for applications in robot perception. Chief among these is reliability of convergence, consistent performance, and a reasonable trade off between accuracy and efficiency. In Chapter 5, we show how the data association problem in multi-target tracking can be formulated as an SLDS and effectively solved using ILMS. The SLDS formulation allows the addition of additional discrete variables which model outliers and clutter in the scene. Evaluations on standard pedestrian tracking sequences demonstrates performance competitive with the state of the art. Chapter 6 applies the ILMS algorithm to robust pose graph estimation. A non-linear CLGN is constructed by introducing outlier indicator variables for all loop closures. The standard Gauss-Newton optimization algorithm is modified to use ILMS as an inference algorithm in between linearizations. Experiments demonstrate a large improvement over state-of-the-art robust techniques. The ILMS strategy presented in this thesis is simple and general, but still works surprisingly well. We argue that these properties are encouraging for wider applicability to problems in robot perception.

Uma das alternativas bastante abordada na literatura para a melhoria do gerenciamento da adubação nitrogenada nas culturas é o sensoriamento remoto, tendo destaque a utilização de sensores espectrais na região do visível e infravermelho. Neste trabalho, buscou-se estabelecer as relações existentes entre variações no teor foliar de nitrogênio (TFN) e a resposta espectral da folha de cana-de-açúcar, utilizando um sensor hiperespectral, com avaliações em três áreas experimentais do estado de São Paulo, com diferentes solos e variedades. Cada experimento foi alocado em blocos ao acaso, com parcelas subdividas e quatro repetições. Foram aplicadas doses de 0, 50, 100 e 150 kg de nitrogênio por hectare. A análise espectral foi realizada na folha \"+1\" em laboratório, sendo coletadas 10 folhas por subparcela, estas foram posteriormente submetidas a análise química para o TFN. Observou-se que existe correlação significativa entre o TFN e as variações na resposta espectral da cana-de-açúcar, sendo que a região do verde e de transição entre o vermelho e o infravermelho próximo (\"red-edge\") foram as mais consistentes e estáveis entre as áreas em estudo e safras avaliadas. A análise de componentes principais permitiu reforçar estes resultados, uma vez que as pontuações (\"scores\") dos componentes que apresentaram correlações significativas com o TFN, tiveram maiores pesos (\"loadings\") nas regiões espectrais citadas anteriormente. A partir das curvas espectrais foram também realizados os cálculos dos índices de vegetação já descritos em literatura, e estes submetidos a análise de regressão simples para predição do TFN, sendo os modelos calibrados com dados da safra 2012/13 e validados com os dados da safra 2013/14. Índices espectrais calculados com a combinação dos comprimentos de onda do verde e/ou \"red-edge\" com comprimentos de onda do infravermelho próximo tiveram bom desempenho na fase de validação, sendo que os cinco mais estáveis foram os índices BNi (500, 705 e 750 nm), GNDVI (550 e 780 nm), NDRE (790 e 720 nm), RI-1db (735 e 720 nm) e VOGa (740 e 720 nm). A variedade SP 81 3250 foi cultivada nas três áreas experimentais, o que permitiu a comparação do potencial de modelos calibrados por área, com um modelo generalista para uma mesma variedade cultivada em diferentes condições edáficas. Observou-se que embora o modelo generalista apresente parâmetros estatísticos significativos, existe redução expressiva da sensibilidade de predição quando comparado aos modelos calibrados por área experimental. Empregou-se também nesta pesquisa a análise de regressão linear múltipla por \"stepwise\" (RLMS) que gerou modelos com boa precisão na estimativa do TFN, mesmo quando calibrados por área experimental, independentes da variedade, utilizando de 5 a 6 comprimentos de onda. Concluímos com a presente pesquisa que comprimentos de onda específicos estão associados a variação do TFN em cana-de-açúcar, e estes são reportados na região do verde (próximos a 550 nm) e na região de transição entre os comprimentos de onda do vermelho e infravermelho próximo (680 a 720 nm). Apesar da baixa correlação entre a região do infravermelho próximo com o TFN, índices de vegetação calculados a partir destes comprimentos de onda ou a inserção destes na geração de modelos lineares foram importantes para melhorar a precisão da predição.<br>An alternative method, quite cited in literature to improve nitrogen fertilization management on crops is the remote sensing, highlighted with the use of spectral sensors in the visible and infrared region. In this work, we sought to establish the relationship between variations in leaf nitrogen content and the spectral response of sugarcane leaf using a hyperspectral sensor, with assessments in three experimental areas of São Paulo state, Brazil, with evaluations in different soils and varieties. Each experimental area was allocated in randomized block, with splitted plots and four repetition, hence, receiving doses of 0, 50, 100 and 150 kg of nitrogen per hectare. Spectral analysis was performed on the \"+1\" leaf in laboratory; we collected 10 leaves per subplots; which were subsequently subjected to chemical analysis to leaf nitrogen content determination. We observed a significant correlation between leaf nitrogen content and variations in sugarcane spectral response, we noticed that the region of the green light and red-edge were the most consistent and stable among the studied area and the crop seasons evaluated. The principal component analysis allowed to reinforce these results, since that the scores for principal components showed significant correlations with the leaf nitrogen content, had higher loadings values for the previous spectral regions mentioned. From the spectral curves were also performed calculations of spectral indices previously described in literature, being these submitted to simple regression analysis to direct prediction of leaf nitrogen content. The models were calibrated with 2012/13 and validated with 2013/14 crop season data. Spectral indices that were calculated with green and/or red-edge, combined with near-infrared wavelengths performed well in the validation phase, and the five most stable were the BNi (500, 705 and 750 nm), GNDVI (550 and 780 nm), NDRE (790 and 720 nm), IR-1dB (735 and 720 nm) and VOGa (740 and 720 nm). The variety SP 81 3250 was cultured in the three experimental areas, allowing to compare the performance of a specific site model with a general model for the same variety growing on different soil conditions. Although the general model presents meaningful statistical parameters, there is a significant reduction in sensitivity to predict leaf nitrogen content of sugarcane when compared with specific site calibrated models. We also used on this research the stepwise multiple linear regression (SMLR) that generated models with good precision to estimate the leaf nitrogen content, even when models are calibrated for an experimental area, regardless of spectral differences between varieties, using 5 to 6 wavelengths. This study shows that specific wavelengths are associated with variation in leaf nitrogen content of sugarcane, and these are reported in the region of green (near to 550 nm) and red-edge (680 to 720nm). Despite the low correlation observed between the infrared wavelengths to the leaf nitrogen content of sugarcane, vegetation indices calculated from these wavelengths, or its insertion on linear models generation were important to improve prediction accuracy.

The purpose of this research work is to propose an improved method for real-time sensitivity analysis (SA) applicable to large-scale complex systems. Borrowed from the EventTracker principle of the interrelation of causal events, it deploys the Rank Order Clustering (ROC) method to automatically group every relevant system input to parameters that represent the system state (i.e. output). The fundamental principle of event modelling is that the state of a given system is a function of every acquirable piece of knowledge or data (input) of events that occur within the system and its wider operational environment unless proven otherwise. It therefore strives to build the theoretical and practical foundation for the engineering of input data. The event modelling platform proposed attempts to filter unwanted data, and more importantly, include information that was thought to be irrelevant at the outset of the design process. The underpinning logic of the proposed Event Clustering technique (EventiC) is to build causal relationship between the events that trigger the inputs and outputs of the system. EventiC groups inputs with relevant corresponding outputs and measures the impact of each input variable on the output variables in short spans of time (relative real-time). It is believed that this grouping of relevant input-output event data by order of its importance in real-time is the key contribution to knowledge in this subject area. Our motivation is that components of current complex and organised systems are capable of generating and sharing information within their network of interrelated devices and systems. In addition to being an intelligent recorder of events, EventiC could also be a platform for preliminary data and knowledge construction. This improvement in the quality, and at times the quantity of input data, may lead to improved higher level mathematical formalism. It is hoped that better models will translate into superior controls and decision making. It is therefore believed that the projected outcome of this research work can be used to predict, stabilize (control), and optimize (operational research) the work of complex systems in the shortest possible time. For proof of concept, EventiC was designed using the MATLAB package and implemented using real-time data from the monitoring and control system of a typical cement manufacturing plant. The purpose for this deployment was to test and validate the concept, and to demonstrate whether the clusters of input data and their levels of importance against system performance indicators could be approved by industry experts. EventiC was used as an input variable selection tool for improving the existing fuzzy controller of the plant. Finally, EventiC was compared with its predecessor EventTracker using the same case study. The results revealed improvements in both computational efficiency and the quality of input variable selection.

L'eccezionale sviluppo della microelettronica e dei sistemi informatici verificatosi nell'ultimo decennio ha consentito la nascita di sensori e dispositivi mobili con caratteristiche senza precedenti; l'elevata potenza di calcolo, il basso costo e le ridotte dimensioni hanno fatto diventare questi strumenti parte della nostra vita quotidiana. Gli smartphone dei nostri giorni integrano funzionalità di comunicazione ed elevate capacità computazionali e di sensing, attraverso un largo numero di sensori embedded. I dati provenienti dai sensori presenti sullo smartphone possono fornire informazioni utili circa l'ambiente nel quale si trova l'utente che sta utilizzando il dispositivo; come raccogliere e analizzare queste grandi quantità di dati rispettando la privacy dell'utente costituisce un'importantissima sfida per il futuro, nonché un'area di ricerca molto attiva. Sofisticate tecniche di riconoscimento si occupano di identificare automaticamente l'attività che l'utente sta svolgendo o il veicolo con il quale egli si sta spostando utilizzando i dati provenienti dai sensori dello smartphone; questo processo di inferenza statistica è possibile grazie ai modelli prodotti dagli algoritmi di classificazione. Partendo dallo stato dell'arte nel campo della cosiddetta Human Activity Recognition, si evidenziano alcune importanti domande ancora senza risposta: è possibile costruire un sistema di classificazione affidabile tracciando il maggior numero possibile di sensori? Il processo di classificazione e di creazione del modello può essere spostato su un server remoto in grado di effettuare velocemente operazioni costose? In questa tesi di laurea si cerca di rispondere alle domande appena poste, implementando un sistema di identificazione automatica dell'attività svolta dall'utente costituito da un'applicazione Android e da un server remoto dedicato al processo di classificazione.

Un dels majors problemes experimentats pels sistemes de detecció de gasos basats en sensors d'òxids metàl·lics és la seva manca de reproduibilitat, estabilitat i selectivitat. A fi i a efecte d'intentar resoldre aquest problemes, diferents estratègies han estat desenvolupades en paral·lel. Algunes es relacionen a la millora dels materials i d'altres impliquen el condicionament o el pre-tractament de les mostres. Les més emprades han consistit en aprofitar que els sensors presenten sensibilitats solapades per construir matrius de sensors i emprar tècniques de processament del senyal o bé utilitzar característiques de la resposta dinàmica dels sensors. En els darrers anys, modular la temperatura de treball del sensors d'òxids metàl·lics s'ha convertit en un dels mètodes més utilitzats per incrementar-ne la selectivitat. Això s'esdevé així donat que la resposta del sensor varia amb la seva temperatura de treball. Per això, en determinats casos, mesurant la resposta d'un sensor a n temperatures de treball diferents pot ser equivalent a tenir una matriu de n sensors diferents. Això permet obtenir informació multivariant de cada sensor individualment i ajuda a mantenir baixa la dimensionalitat del sistema de mesura per resoldre una determinada aplicació. Malgrat que molts i bons resultats han estat publicats dins aquest àmbit, la tria de les freqüències emprades en la modulació de la temperatura de treball dels sensor ha consistit fins ara en un procés empíric que no garanteix la obtenció dels millors resultats per una determinada aplicació. En aquest context, el principal objectiu d'aquesta tesi doctoral ha consistit en desenvolupar un mètode sistemàtic que permeti determinar quines són les freqüències de modulació òptimes que caldria emprar per resoldre un determinat problema d'anàlisi de gasos. Aquest mètode, extret del camp d'identificació de sistemes, ha esta desenvolupat i implementat per primer cop dins l'àmbit dels sensors de gasos. Aquest consisteix en estudiar la resposta dels sensors en presència de gasos mentre la temperatura de treball dels sensors és modulada per un senyal pseudo-aleatori de longitud màxima. Aquest senyals comparteixen algunes propietats amb el soroll blanc, i per tant poden ajudar a estimar la resposta lineal d'un sistema amb no-linealitats (per exemple, la resposta impulsional d'un sistema sensor-gas). El procés d'optimització es duu a terme mitjançant la selecció entre els components espectrals de les estimacions de la resposta impulsional, d'aquells que millor ajuden bé a discriminar o a quantificar els gasos objectiu dins una aplicació d'anàlisi de gasos donada. Tenint en compte que els components espectrals estan directament relacionats amb les xvii Improving the performance of micro-machined metal oxide gas sensors: Optimization of the temperature modulation mode via pseudo-random sequences. freqüències de modulació, la tria d'uns pocs components espectrals resulta en la determinació de les freqüències òptimes de modulació. En els primer experiments, senyals binaris pseudo-aleatoris van ser emprats per modular la temperatura de treball de sensors de gasos basats en òxids metàl·lics micro-mecanitzats dins d'un rang comprès entre 0 i 112,5 Hz. La freqüència superior és lleugerament superior a la frequència de tall de les membranes dels sensors. El resultat principal derivat d'aques estudi va ser que les freqüències de modulació interessants es trobaven en un rang comprès entre 0 i 1 Hz. Això és comprensible donat que la cinètica de les reaccions i dels processos d'adsorció que es produeixen en la superfície dels sensors són lentes i si aquestes s'han de veure modificades per la modulació térmica, llavors caldran senyals de modulació de baixa freqüència. Això explica perquè s'han vingut emprant senyals moduladores de temperatura en el rang dels mHz, malgrat que les membranes d'un dispositiu micromecanitzat presenten respostes tèrmiques molts més ràpides (típicament de l'ordre de 100 Hz). En els experiments que continuaren els primers, un mètode evolucionat per determinar les freqüències de modulació tèrmica òptimes va ser implementat. Aquest es basa en l'ús de seqüències pseudo-aleatòries multi-nivell de longitud màxima. Els senyals de tipus multi-nivell van ser considerats en substitució dels senyals binaris ja que els primers permeten obtenir una millor estimació que els segons de la dinàmica lineal d'un sistema amb no linealitats. I és ben conegut que els sensors de gasos basats en òxids metàl·lics presenten no linealitat en la seva resposta. Aquests estudis sistemàtics van ser completament validats mitjançant la síntesi de senyals multi-sinusoïdals amb les freqüències prèviament identificades emprant sequències pseudo-aleatòries. Quan la temperatura de treball dels sensors va ser modulada amb un senyal, el contingut freqüencial del qual era l'òptim, els gasos i les mescles de gasos considerades van poder ser discriminades perfectament i es va mostrar la possibilitat d'obtenir models de calibració acurats per predir la concentració dels gasos. En alguns casos, aquest procés de validació es va portar a terme emprant sensors que no havien estat utilitzats durant el procés d'optimització (per exemple, una agrupació de sensors diferent però del mateix lot de fabricació). En resum, el nou mètode desenvolupat en aquesta tesi per seleccionar les freqüències de modulació òptimes s'ha mostrat consistent i efectiu. El mètode és d'aplicació general i podria ser emprat en qualsevol problema d'anàlisi de gasos o bé estès a altres tipus de sensors (per exemple sensors polimèrics). Les contribucions científiques d'aquesta tesi s'han recollit en quatre articles en revistes internacionals i 13 llibres d'actes de conferències.<br>Uno de los mayores problemas experimentados en los sistemas de detección de gases basados en dispositivos de óxidos metálicos es su falta de reproducibilidad, estabilidad y selectividad. Con el fin de intentar resolver estos problemas, diferentes estrategias han sido desarrolladas en paralelo. Algunas de ellas se relacionan con la mejora de los materiales y otras implican acondicionamiento o pre-tratamiento de las muestras. Otras estrategias ampliamente empleadas consisten en aprovechar que los sensores presentan sensibilidades solapadas para construir matrices de sensores y emplear técnicas de procesamiento de señal o bien utilizar características de la respuesta dinámica de los sensores.En los últimos años, modular la temperatura de trabajo de los sensores de óxidos metálicos se ha convertido en uno de los métodos más utilizados para incrementar su selectividad. Esto se debe a, dado que la respuesta del sensor varía con su propia temperatura de trabajo, entonces, en determinados casos, midiendo la respuesta de un sensor a n temperaturas de trabajo diferentes, es equivalente a tener una matriz de n sensores diferentes. Esto permite obtener información multivariante de cada sensor individualmente y ayuda a mantener baja la dimensionalidad del sistema de medida para resolver una determinada aplicación. A pesar de los buenos resultados que han sido publicados dentro de este ámbito, la selección de las frecuencias empleadas en la modulación de la temperatura de trabajo de los sensores ha consistido, hasta el momento, en un proceso empírico lo que no garantiza la obtención de los mejores resultados para una determinada aplicación.En este contexto, el principal objetivo de esta tesis doctoral ha consistido en desarrollar un método sistemático que permita determinar cuales son las frecuencias de modulación óptimas que podrían emplearse para resolver un determinado problema de análisis de gases. Este método, extraído del campo de identificación de sistemas, ha sido desarrollado e implementado por primera vez dentro del ámbito de los sensores de gases. Éste consiste en estudiar la respuesta de los sensores en presencia de gases mientras la temperatura de trabajo de los sensores es modulada mediante una señal pseudo-aleatoria de longitud máxima. Estas señales comparten algunas propiedades con el ruido blanco, y por tanto pueden ayudar a estimar la respuesta lineal de un sistema con no-linealidades (por ejemplo, la respuesta impulsional de un sistema sensor-gas).El proceso de optimización es llevado a cabo mediante la selección entre las componentes espectrales de las estimaciones de la respuesta impulsional, de aquellas que más ayudan ya sea a discriminar o a cuantificar los gases objetivo dentro de una aplicación de análisis de gases dada. Teniendo en cuenta que las componentes espectrales están directamente relacionadas con las frecuencias de modulación, la selección de unas pocas componentes espectrales resulta en la determinación de las frecuencias optimas de modulación.En los primeres experimentos, señales binarias pseudo-aleatorias fueron utilizadas para modular la temperatura de trabajo de los sensores de gases basados en óxidos metálicos micro-mecanizados en un rango comprendido entre 0 a 112.5 Hz. La frecuencia superior es ligeramente mayor a la frecuencia de corte de las membranas de los sensores. El resultado principal derivado de estos estudios fue que las frecuencias de modulación interesantes se encuentran en un rango comprendido entre 0 y 1 Hz. Esto es comprensible dado que la cinética de las reacciones y de los procesos de adsorción que se producen en la superficie del sensor son lentos y si estos se han de alterar mediante la modulación térmica, se habrá de elaborar señales de modulación a bajas frecuencias. Esto explica por que se han venido empleado señales moduladoras de temperatura en el rango de los mHz, a pesar que las membranas de un dispositivo micro-mecanizado presentan respuestas mucho más rápidas (típicamente en el orden de los 100 Hz).En los experimentos posteriores a los primeros, un método evolucionado para determinar las frecuencias de modulación óptimas de los sensores micro-mecanizados fue implementado, el cual se basa en el uso de secuencias pseudo-aleatorias multi-nivel de longitud máxima (MLPRS). Las señales de tipo multi-nivel fueron consideradas en lugar de las binarias ya que las primeras permiten obtener una mejor estimación que las segundas de la dinámica lineal de un sistema con no linealidades. Y es bien conocido que los sensores de gases basados en óxidos metálicos presentan no-linealidades en su respuesta.Estos estudios sistemáticos fueron completamente validados mediante la síntesis de señales multi-senoidales con las frecuencias previamente identificadas utilizando secuencias pseudo-aleatorias. Cuando la temperatura de trabajo de los sensores fue modulada por una señal, el contenido frecuencial de la cual es el óptimo, los gases y mezclas de gases considerados pudieron ser discriminados perfectamente y se verificó la posibilidad de obtener modelos de calibración precisos para predecir la concentración de los gases. En algunos casos, estos procesos de validación se llevaron a cabo con sensores que no habían sido utilizados durante el proceso de optimización (por ejemplo, una agrupación de sensores diferentes pero del mismo lote de fabricación).En resumen, El nuevo método desarrollado in esta tesis para seleccionar las frecuencias de modulación optimas se a mostrado consistente y efectivo. El método es de aplicación general y podría ser utilizado en cualquier problema de análisis de gases o bien extendido a otro tipo de sensores (por ejemplo sensores poliméricos).Las contribuciones científicas de esta tesis se han recogido en 4 artículos en revistas internacionales y trece actas de conferencias.<br>One of the major problems in gas sensing systems that use metal oxide devices is the lack of reproducibility, stability and selectivity. In order to tackle these troubles experienced with metal oxide gas sensors, different strategies have been developed in parallel. Some of these are related to the improvement of materials, or the use of sample conditioning and pre-treating methods. Other widely used techniques include taking benefit of the unavoidable partially overlapping sensitivities by using sensor arrays and pattern recognition techniques or the use of dynamic features from the gas sensor response.In the last years, modulating the working temperature of metal oxide gas sensors has been one of the most used methods to enhance sensor selectivity. This occurs because, since, the sensor response is different at different working temperatures, and therefore, measuring the sensor response at n different temperatures is, in some cases, similar to the use of an array comprising n different sensors. This allows for measuring multivariate information from every single sensor and helps in keeping low the dimensionality of the measurement system needed to solve a specific application. Although the good results reported, until now, the selection of the frequencies used to modulate the working temperature remained an empirical process and that is not an accurate method to ensure that the best results are reached for a given application.In view of this context, the principal objective of this doctoral thesis was to develop a systematic method to determine which are the optimal temperature modulation frequencies to solve a given gas analysis problem. This method, which is borrowed from the field of system identification, has been developed and introduced for the first time in the area of gas sensors. It consists of studying the sensor response to gases when the operating temperature is modulated via maximum-length pseudo-random sequences. Such signals share some properties with white noise and, therefore, can be of help to estimate the linear response of a system with non-linearity (e.g., the impulse response of a sensor-gas system).The optimization process is conducted by selecting among the spectral components of the impulse response estimates, the few that better help either discriminating or quantifying the target gases of a given gas analysis application. Since spectral components are directly related to modulating frequencies, the selection of spectral components results in the determination of the optimal temperature modulating frequencies.In the first experiments, pseudo-random binary signals (PRBS) were employed to modulate the working temperature of micro-machined metal oxide gas sensors in a frequency range from 0 up to 112.5 Hz. The upper frequency is slightly higher than the cutoff frequency of the sensor membranes. The outcome of this initial study was that the important modulating frequencies were in the range between 0 and 1 Hz. This is understandable, since the kinetics of reaction and adsorption processes taking place at the sensor surface (i.e., physisorption/chemisorption/ionosorption) are slow and if these are to be altered by the thermal modulation, low frequency modulating signals need to be devised. This explains why low-frequency temperature-modulating signals (i.e. in the mHz range) have been used with micro-hotplate gas sensors, even though the thermal response of their membranes is much faster (typically, near 100 Hz).In the experiments that followed the first ones, an evolved method to determine the optimal temperature modulating frequencies for micro-hotplate gas sensors was introduced, which was based on the use of maximum length multilevel pseudo-random sequences (MLPRS). Multilevel signals were considered instead of the binary ones because the former can provide a better estimate than the latter of the linear dynamics of a process with non-linearity. And it is well known that temperature-modulated metal oxide gas sensors present non-linearity in their response.These systematic studies were fully validated by synthesizing multi-sinusoidal signals at the optimal frequencies previously identified using pseudo-random sequences. When the sensors had their operating temperatures modulated by a signal with a frequency content that corresponded to the optimal, the gases and gas mixtures considered could be perfectly discriminated and the building of accurate calibration models to predict gas concentration was found to be possible. In some cases, the validation process was conducted on sensors that had not been used for optimization purposes (e.g. a different sensor array from the same fabrication batch).Summarizing, the new method developed in this thesis for selecting the optimal modulating frequencies is shown to be consistent and effective. The method applies generally and could be used in any gas analysis problem or extended to other type of sensors (e.g. conducting polymer sensors).The scientific contributions of this thesis are collected in four journal papers and thirteen conference proceedings.

We develop a family of Bayesian algorithms built around Gaussian processes for various problems posed by sensor networks. We firstly introduce an iterative Gaussian process for multi-sensor inference problems, and show how our algorithm is able to cope with data that may be noisy, missing, delayed and/or correlated. Our algorithm can also effectively manage data that features changepoints, such as sensor faults. Extensions to our algorithm allow us to tackle some of the decision problems faced in sensor networks, including observation scheduling. Along these lines, we also propose a general method of global optimisation, Gaussian process global optimisation (GPGO), and demonstrate how it may be used for sensor placement. Our algorithms operate within a complete Bayesian probabilistic framework. As such, we show how the hyperparameters of our system can be marginalised by use of Bayesian quadrature, a principled method of approximate integration. Similar techniques also allow us to produce full posterior distributions for any hyperparameters of interest, such as the location of changepoints. We frame the selection of the positions of the hyperparameter samples required by Bayesian quadrature as a decision problem, with the aim of minimising the uncertainty we possess about the values of the integrals we are approximating. Taking this approach, we have developed sampling for Bayesian quadrature (SBQ), a principled competitor to Monte Carlo methods. We conclude by testing our proposals on real weather sensor networks. We further benchmark GPGO on a wide range of canonical test problems, over which it achieves a significant improvement on its competitors. Finally, the efficacy of SBQ is demonstrated in the context of both prediction and optimisation.

In this thesis, pH and potassium all-solid-state ISE based on potentiometry and bioimpedance sensors were designed, fabricated and integrated in a miniaturized array for its application in endoscopic surgery for in vivo ischemia detection inside the stomach. To achieve this goal, the developed array withstood the low pH and corrosive condition in the gastric juice of the stomach, by modifying the surface with a conductive Ag/AgCl ink containing hydrophilic and hydrophobic groups. That creates a stable and robust candidate for low pH applications. However, these sensors have to demonstrate besides stability, high sensitivity, and selectivity. For this purpose, different ionophores specific to a single ion were tested. Octadecyl isonicotinate was the one that shown better results as pH ionophore and valinomycin, bis [(benzo-15-crown-4)-4-ylmethyl] pimelate for potassium detection. All these ionophores were embedded in PVC polymer membrane containing also plasticizers such as 2-nitrophenyl octyl ether, bis (1-butylpentyl) adipate (BBPA) and liphophilic anionic additives such as potassium tetrakis (4-chlorophenyl) borate (KTpClPB). The specific compositions of membranes to detect potassium or pH were optimized for the better performance of the sensors. pH ISE sensor shows a nernstian behavior (-54,38 mV/pH) at low pH and a nearly nernstian behavior at physiological pH (-34,899 mV/pH). Bioimpedance sensor was tested and optimized in vitro with different solutions of ions concentration to mimic ischemia detection and with different kinds of tissues from different nature. For this purpose, chicken fat and breast tissues were taken as a model for mimicking non-ischemic and ischemic states respectively. The effect of electrodes insulation as well as the pressure applied on the tissue was studied. The dependence of the impedance response with different pressure applied to the sensor was overcome by applying magnetic field attachment. The sensor array was modified with ring magnets which were attracted by an external magnet, giving stable and reliable signal discarding mechanical motion. The shape and size of the sensor array were designed for being adapted to the commercially available gastroendoscopes. Round shaped cylinder of 7 mm diameter was fabricated with 12 electrodes pin of 600 µm diameter, containing 3 RE, 3 pH and 2 potassium all-solid-state sensors and 4 electrodes in a row for impedance measurements. The sensor array was successfully integrated in commercial endoscope and inserted inside the pig stomach. The blood flow of certain area of the stomach was interrupted by ligating or crossclamping vessels and organ wall. Ischemia and reperfusion steps were sensed successfully with potassium and pH sensors. These results also indicate that information about hypoxic tissue damage can be collected with this array. Ischemia was also sensed on small intestine tissue by opening the abdominal part of the body and getting the sensor array in contact with the intestine. By crossclamping of mesenteric artery by tourniquets and scissors, ischemic and reperfusion states were controlled. Results proved that ischemia and reperfusion can be monitored by our integrated sensor array. As a conclusion, a novel all-solid-state potentiometric, miniaturized, low cost and mass producible pH, potassium all-solid-state ISE and impedance sensors integrated in an array was successfully fabricated for detecting ischemia inside the stomach by means of endoscopic techniques and also on small intestine. This array was tested in vitro and vivo giving reproducible and reliable results. The developed all-solid-state pH sensors permit low pH sensing from 0.7-2.5, which is the only example in the literature that allows so low pH detection, and so make this sensor a unique device for stomach detection.<br>El diagnóstico médico es uno de los campos que han obtenido más ventajas de la capacidad de los electrodos selectivos de iones (ESI) para la detección de iones, ya que los cambios en la concentración de estos elementos están directamente relacionados con diferentes enfermedades. La detección de isquemia es una de las favorecidas por estos sensores. La isquemia es una disminución del suministro de sangre a un órgano y se requiere una detección rápida y precisa. Los métodos de detección in situ en el tejido de los órganos conllevan una detección temprana de la isquemia y el estómago es uno de los órganos más importantes en la detección de Ischemia. Sin embargo, el bajo pH del jugo gástrico del estómago hace difícil la fabricación de sensores de estado sólido con ESI estables y funcionales, principalmente debido a la interferencia de aniones y al problema de la adhesión entre la membrana ESI y la superficie del electrodo. En esta tesis, se han diseñado y fabricado electrodos selección de iones de pH y potasio ESI de estado sólido basados en la potenciometría y sensores de bioimpedancia y se han integrado en una matriz en miniatura para su aplicación en la cirugía endoscópica para la detección de isquemia in vivo en el interior del estómago. El conjunto de sensores se integró con éxito en endoscopio comercial y se inserto en el interior del estómago de un cerdo. El flujo de sangre de cierta área del estómago se interrumpió mediante la ligación o pinzamiento de los vasos sanguineos y la pared del órgano. Los pasos de isquemia y reperfusión fueron detectados con éxito con los sensores de potasio y de pH. Estos resultados también indican que se puede obtener información sobre el daño en el tejido hipóxico recogido con esta matriz. Los sensores de pH de sólido desarrollados permiten la detección pH bajos de 0,7 a 2,5, que es el único ejemplo en la literatura de detección de pH tan bajos con este tipo de sensores y por lo tanto hacen que sea este sensor de un dispositivo único para la detección de isquemia en el estómago.

We consider the problem of selecting k sensors out of m available (linear) sensors, so that the error in estimating some parameters is minimized. When the sensor noises are uncorrelated, the sensor selection problem can be (approximately) solved by a method recently suggested by Joshi and Boyd, which relies on a convex relaxation of the underlying combinatorial optimization problem. This thesis describes a non-trivial extension of the relaxation method to the case when the measurement noises are correlated, as occurs, for example, in a sensor scheduling problem in a dynamic system. We develop several new semidenite programming (SDP) relaxations for the problem, which give provable bounds on the attainable performance, as well as suboptimal sensor selections. Numerical experiments for sensor scheduling suggest that the methods work well.

In the modern world, systems such as aircraft systems are becoming increasingly complex, often consisting of a large number of components. As no component is perfectly reliable, they can fail, some in many different ways, leading to a large number of potential component failures on complex systems. Component failures can have detrimental effects on the performance of the system, with some component failures even causing system failure, potentially damaging the system, or more importantly, potentially endangering human life. In order to be able to detect component failures on complex systems, the inclusion of sensors is becoming increasingly common. In addition to being able to detect component failures, the sensors can be used to diagnose component failures, with certain symptoms and the resultant sensor readings being produced by certain component failures. Another benefit is that it may also be possible to prevent system failure by detecting component failures early, activating redundant components and enabling the mission to be completed. However, including sensors in the system increases the cost of the system, can add weight to the system and require space for installation, a factor of particular importance for weight critical systems, such as aircraft systems. Therefore, a balance between being able to detect and diagnose failures in systems and the cost, weight and space requirements of the sensors needs to be achieved. In this thesis, a novel sensor selection methodology is proposed, which is based on a performance metric. Individual sensors, and combinations of sensors are ranked based on their performance of detecting faults and diagnosing failures in the system. In addition to the sensors’ detection and diagnostic performance, the metric also considers the effect that the component failures have on the functionality of the system, where sensors that detect critical failures are favoured over sensors that do not detect such failures. The performance metric is then extended to consider the time taken to detect and diagnose component failures, as the sooner component failures are detected, the more likely system failure can be prevented. This is important in a system that operates in a phased mission. In addition, a proposed two-level Genetic Algorithm is used in order to efficiently determine a suitable combination of sensors for larger systems, where an exhaustive calculation of the performance metric for all combinations of sensors is not feasible. For a simple flow system, a Bayesian Belief Network (BBN) is used to model the effects of component failures, and sensor readings. During the fault diagnostic process, observed sensor readings can be introduced in the BBN, which then can be used to identify the failed components. However, an alternative system modelling and fault diagnostic technique is proposed as a part of this thesis which can be used on larger systems, and can determine sensor readings and component failures more quickly than the BBN method. This method is based on a series of if-then-else statements in order to determine the effect that the component states have on the performance of the system. The work proposed in this thesis is applied to three example systems: a simple flow system, an example aircraft fuel system and the fuel system for an Airbus A380-800 aircraft.

This research contributed to the development of advanced feature selection model, hyperparameter optimization and temporal multimodal deep learning model to improve the performance of dimensional emotion recognition. This study adopts different approaches based on portable wearable physiological sensors. It identified best models for feature selection and best hyperparameter values for Long Short-Term Memory network and how to fuse multi-modal sensors efficiently for assessing emotion recognition. All methods of this thesis collectively deliver better algorithms and maximize the use of miniaturized sensors to provide an accurate measurement of emotion recognition.

Han estat preparats diversos sensors potenciomètrics basats en parells iònics insolubles (1:1) provinents de l'associació d'ions tensioactius, en particular el catió tetradodecilamoni i l'anió dodecilbenzesulfonat (DBS-). Els materials sensors, incorporats a una matriu polimèrica de clorur de polivinil (PVC) que contenia com a agent plastificant l'o-nitrofeniloctil èter, van ser aplicats sobre una resina conductora per tal de fornir elèctrodes sense solució interna de referència. La resposta d'aquests elèctrodes a dodecilsulfat (DS-) i DBS- així com la interferència de diversos anions inorgànics habituals i cations tensioactius va ser igualment examinada. Es presenten els principals paràmetres corresponents a l'avaluació d'aquests elèctrodes i els seus temps de resposta. Les membranes estudiades mostren un bon comportament en ser utilitzades com a sensors potenciomètrics per a tensioactius aniònics.<br/><br/>Aquests elèctrodes van ser emprats en la determinació de mostres comercials d'anions tensioactius per mitjà de valoracions potenciomètriques amb el reactiu valorant Hyamine 1622 (Hy+), sense aparèixer discrepàncies en els resultats quan aquests es comparaven amb el mètode de la valoració amb indicador mixt de les dues fases. Les característiques globals de funcionament milloren les corresponents a l'elèctrode comercial de tensioactius Orion, permetent la valoració de tensioactius a concentracions inferiors a 10mM i la seva aplicabilitat en anàlisis de rutina.<br/><br/>Ha estat desenvolupat un sistema d'injecció en flux amb detecció potenciomètrica per a la monitorització del contingut de tensioactiu aniònic en solucions de rentat de planxes metàl·liques, emprant elèctrodes tubulars all-solid-state revestits de membrana de PVC. La mostra, amb una força iònica elevada i un pH extremadament bàsic, es determinada després de ser condicionada a partir d'un sistema de flux bicanal. Amb aquest procediment s'obtenen uns errors mitjans al voltant de l'1 % i una velocitat d'anàlisi d'unes 30 mostres/hora.<br/><br/>Fent ús de la mateixa membrana polimèrica, es van desenvolupar transistors d'efecte de camp selectius a ions (ISFETs) per a tensioactius aniònics. En aquest cas, els dispositius van mostrar uns temps de vida superiors a 4 mesos, millorant els resultats habituals per a ISFETs basats en membranes de PVC. Altres característiques a ressaltar són els pendents nernstians aconseguits (59 a 62 mV/dec), límits de detecció propers a 10-6 M, una bona linealitat i una adequada resposta tant a diversos anions tensioactius com al reactiu emprat en la determinació potenciomètrica, el que va permetre l'avaluació del contingut global dels tensioactius aniònics presents en diferents mostres. No van aparèixer diferències significatives durant la comparació dels resultats obtinguts emprant el mètode proposat que utilitza els ISFETs com a indicadors de punt final i el mètode de valoració en dues fases.<br/><br/>Van ser estudiades noves formulacions fotocurables en un intent de superar les limitacions inherents a les membranes de PVC i compatibilitzar la preparació de les membranes amb les tècniques fotolitogràfiques que poden produir sensors miniaturitzats. Aquestes membranes estan basades en un polímer d'uretà-acrilat que empra 2-cianofeniloctil èter com a plastificant compatible amb el procés de fotopolimerització. Aquestes membranes han resultat altament selectives als anions tensioactius assajats al mateix temps que els anions inorgànics més habituals no interfereixen de forma important. Es proposa una metodologia per tal d'optimitzar la composició de la membrana que condueix a una formulació apta per a un ús general. Aquesta membrana és completament caracteritzada, incloent els calibratges a anions tensioactius com el DBS-, el tetrapropilenbenzesulfonat i el DS-, així com als cations Hy+ i cetiltrimetilamoni. Les dades corresponents a l'ió principal mostren una sensibilitat de 58.1 mV/dec, una resposta lineal situada entre 1 x 10-3 M i 3 x 10-6 M, un límit de detecció corresponent a 0.26 ppm DBS- i una reproductibilitat del pendent entre dies, expressada com a desviació estàndard relativa, del 2.8 %.<br/><br/>Finalment, es descriu la preparació de nous ISFETs basats en la membrana fotocurable prèviament optimitzada. Aquests dispositius no presenten diferències significatives (P=0.05) en termes de sensibilitat i reproductibilitat en ser comparats amb els corresponents elèctrodes selectius i presenten, addicionalment, una millora en els temps de resposta. Es mostra l'aplicació d'aquests ISFETs en la monitorització dels processos de fotodegradació, emprant dispersions de diòxid de titani, per a dos anions tensioactius: el DBS-, que conté una part aromàtica en la seva estructura i el DS-, de cadena alquílica. La determinació del contingut de tensioactiu va ser portada a terme emprant una metodologia d'addició estàndard fent ús dels ISFETs com a sensors potenciomètrics, sense la necessitat d'etapes de separació prèvies. La cinètica de degradació, en ambdós casos, va resultar de primer ordre, amb uns temps de vida mitjana de 31.5 min per a DBS- i 52.0 min per a DS-.<br>Potentiometric sensors, based on ion-pair 1:1 water-insoluble salts, were prepared using tetradodecylammonim as cationic and dodecylbenzenesulfonate (DBS-) as anionic surfactant. Sensing materials were incorporated in poly(vinyl chloride) (PVC) matrix containing o-nitrophenyl octyl ether as a plasticizer and applied on a support of conductive resin without inner reference solution. The response of these electrodes to dodecylsulfate (DS-) and DBS- as well as the interferences of several common inorganic anions and cationic surfactants were examined. The main parameters of evaluation of these electrodes and response times are presented and its behaviour is discussed. The membranes show good performance for use as a general potentiometric sensor responsive to anionic surfactants.<br/><br/>These electrodes were applied to the determination of anionic surfactants by potentiometric titration using Hyamine 1622 (Hy+) as the titrant. The results from this method compare favourably with those of the two-phase mixed indicator titrationt method for several commercial anionic surfactants. The performance characteristics improve the commercially available Orion surfactant electrode allowing the titration of surfactants at concentrations down to 10mM and its applicability in routine analysis.<br/><br/>A flow-injection system based on potentiometric detection and designed for the monitoring of anionic surfactant content of printing plates washing solutions was developed. PVC membrane ion-selective electrodes, constructed with all-solid-state tubular flow-through design were used. The sample, with high ionic strength and extremely alkaline pH, was conditioned in a two-channel flow injection system, allowing for surfactant determination. Errors of approximately 1 % may be obtained and the sample throughput achieved is around 30 samples per hour.<br/><br/>Using the same PVC membrane, ion-selective field-effect transistors (ISFETs) to anionic surfactants were developed. These devices showed a lifetime longer than four months, improving values of PVC membrane-ISFETs. Othes characteristics are Nernstian slopes from 59 to 62 mV/dec, detection limits of about 10-6 M and good linearity. The also showed response to several anionic surfactant species and to the reagent used for the potentiometric titration. This allowed the measurement of the overall anonic surfactant content in different samples. In a comparative study, there were no significant diferences between the results produced with the standard, two-phase titration method and the proposed potentiometric titration method using surfactant ISFETs as end-point indicators.<br/><br/>Overcoming the PVC membrane limitations, new photocurable formulations were studied, were the membrane manufacture could be compatible with massive photolithographic techniques to produce miniaturized sensors. These membranes are based on a urethane-acrylate polymer, and use 2-cyanophenyl octyl ether (CPOE), a plasticizer compatible with the photocuring process. These membranes are highly selective to the anionic surfactants assayed while common inorganic anions did not interfere. An optimization methodology is proposed for their formulation, suited to the final application. A membrane with a general-purpose formulation is fully characterized, including calibration results for anionic surfactants such as DBS-, tetrapropylenebenzenesulfonate and DS-, or cationic surfactants such as Hy+ and cetyltrimethylammonium ion. With the primary ion DBS- we verified a 58.1 mV/dec sensitivity, a linear response between 1 x 10-3 M and 3 x 10-6 M, a detection limit corresponding to 0.26 ppm DBS- and a slope precision of 2.8 % RSD betweeen days.<br/><br/>Finally, the preparation of a new ISFET based on the optimized previous membrane is described. When compared to convencional ion-selective electrodes, the prepared ISFETs do not show significant diferences in sensitivity and reproducibility (P=0.05), improving the response time. As the application, photodegradation process using titanium dioxide dispersions, were monitorized for two anionic surfactants: DBS-, being aromatic, and the more alkylic DS-. The determination of surfactant concentration was performed following a standard addition methodology, using ISFETs as the sensors, and without any previous separation stages. The degradation kinetics in both cases are first-order processes, with half-life times of 31.5 min for DBS- and 52.0 min for DS-.

Nitrate determination and ion-selective electrodes, with particular emphasis on the nitrate selective electrode, are reviewed. A new electrode for nitrate has been developed, by covalent attachment of appropriate sensor groups to a robust polymer matrix, with a view to improving the lifetime and the applications of the electrode to include hostile environments. The QAS (I) were prepared as sensor materials.(1) R' = R" = R''' = ethyl(2) R' = allylR" = R''' = methyl, ethyl, propyl(3) R' = R" = allylR''' = methyl, ethyl, propyl, butyl(4) R' = R" = R''' = allylX- = Cl, Br. (I) [diagram]. The production of membranes by cross-linking the sensors (I) with styrene-b-butadiene-b-styrene triblock elastomer (SBS) using a free radical initiated cyclopolymerisation is described. Various methods of membrane preparation have been investigated, and a convenient one-step solvent casting procedure identified and optimised. A range of membranes has been prepared and their physical and electroanalytical properties evaluated. The cross-linked density of the polymers was calculated from swelling measurements and the Flory-Rehner equation, the extent of covalent attachment has been derived from Kjeldahl analyses. The performance of the membranes as ion-selective electrodes is presented and discussed. Several electrodes showed fast response, long lifetime and Nernstian behaviour in the range 10-1 - 10-4 mol dm-3, but their selectivity was inferior to present commercial electrodes. A number of materials were introduced as mediators in order to improve the selectivity of the membranes. A possible mediator (O-nitrophenyloctyl ether) was identified and an analogue (O-nitrophenyl-w-undecylenyl ether) prepared which had suitable functionality for covalent attachment to the polymer. Unfortunately the inclusion of this material did not improvethe selectivity to chloride and the reasons for this are discussed.

Thesis (M.S.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Systems Science and Industrial Engineering, 2007.<br>Includes bibliographical references.

Boronic acids in orgamc chemistry are extensively used as building blocks in coupling reactions. Moreover, boronic acids also act as Lewis acids which enable them to easily form reversible covalent complexes with sugars, amino acids and other Lewis base donors. Taking this unique characteristic that two phenylboronic acids separated by a six-carbon amino linker are especially selective for D-glucose. The major part of this research is to modify boronic acid sensors with various fluorophores and material supports, not just for saccharide detection, but also for other applications such as protein / peptide separation in electrophoresis or surface plasmon resonance analysis. Saccharides also form glycoconjugates which are involved in cell-cell recognition / interactions in biological systems; therefore, the development of new boronic acid fluorescent sensors for bioimaging is also included within the research.

Optimum Sensor Localization/Selection in A Diagnostic/Prognostic Architecture Guangfan Zhang 107 Pages Directed by Dr. George J. Vachtsevanos This research addresses the problem of sensor localization/selection for fault diagnostic purposes in Prognostics and Health Management (PHM)/Condition-Based Maintenance (CBM) systems. The performance of PHM/CBM systems relies not only on the diagnostic/prognostic algorithms used, but also on the types, location, and number of sensors selected. Most of the research reported in the area of sensor localization/selection for fault diagnosis focuses on qualitative analysis and lacks a uniform figure of merit. Moreover, sensor localization/selection is mainly studied as an open-loop problem without considering the performance feedback from the on-line diagnostic/prognostic system. In this research, a novel approach for sensor localization/selection is proposed in an integrated diagnostic/prognostic architecture to achieve maximum diagnostic performance. First, a fault detectability metric is defined quantitatively. A novel graph-based approach, the Quantified-Directed Model, is called upon to model fault propagation in complex systems and an appropriate figure-of-merit is defined to maximize fault detectability and minimize the required number of sensors while achieving optimum performance. Secondly, the proposed sensor localization/selection strategy is integrated into a diagnostic/prognostic system architecture while exhibiting attributes of flexibility and scalability. Moreover, the performance is validated and verified in the integrated diagnostic/prognostic architecture, and the performance of the integrated diagnostic/prognostic architecture acts as useful feedback for further optimizing the sensors considered. The approach is tested and validated through a five-tank simulation system. This research has led to the following major contributions: ??generalized methodology for sensor localization/selection for fault diagnostic purposes. ??quantitative definition of fault detection ability of a sensor, a novel Quantified-Directed Model (QDG) method for fault propagation modeling purposes, and a generalized figure of merit to maximize fault detectability and minimize the required number of sensors while achieving optimum diagnostic performance at the system level. ??novel, integrated architecture for a diagnostic/prognostic system. ??lidation of the proposed sensor localization/selection approach in the integrated diagnostic/prognostic architecture.

This thesis presents the design, fabrication, characterisation and testing of a chemically modified electrothermally actuated microgripper. The chemical modification involves the integration of a potentiometric ion selective electrode (ISE) onto a bare electrode fabricated within the tip of the microgripper. This microgripper sensor device is intended for use in the application of detecting, in real time, the movement of key ions that are involved in intercellular communication from a mechanically stressed cell. An optimised fabrication route for the specifically designed microgrippers, which have tip dimensions of 10 – 60 µm, is described in detail. The fabrication route delivers a high yield (95%) of operational unmodified devices. An 1800 ± 20 µm2 bare gold electrode that is fabricated at the tip of the microgripper is modified into an all solid state ISE that uses PEDOT as the ion-to-electron solid contact. Suitable ionophores that selectively detect K+, Na+ and Ca2+ are used to fabricate potassium, sodium and calcium ion selective microgripper sensor devices. The quality control and testing characteristics that follow the guidelines defined by IUPAC are performed to ascertain the sensitivity, selectivity and stability of the microgripper sensor devices. Good selectivity is achieved, with limits of detection of 2.4 x 10-4 M, 1.8 x 10-4 M and 2.0 x 10-5 M for the K+, Na+ and Ca2+ devices respectively. Proof of concept experiments of the real life testing of the K+ ISE device used to mechanically stress mouse oocytes gave preliminary measurements that indicate that stress signalling occurs via a switch on mechanism, and that there is a small increase in K+ concentration as applied stress increases. Due to the high systematic error within the calibration process the magnitude of this concentration increase is unknown. The Na+ and Ca2+ ISE devices suffer from interference and sensitivity restrictions respectively so a signal response vs. applied cell stress relationship of these ions is currently unobtainable.

Les câbles électriques sont utilisés dans tous les secteurs pour transférer de l'énergie ou de l'information. Pendant le fonctionnement, les câbles peuvent être sujets à des défauts francs (circuit ouvert ou court-circuit) ou des défauts non-francs (endommagement de l'isolant, pincement, etc.) dus à une mauvaise utilisation, aux conditions environnementales ou au vieillissement. Ces défauts doivent être détectés à leur stade le plus précoce pour éviter une interruption de la fonction ou des conséquences plus graves. Parmi les méthodes de diagnostic des réseaux filaires qui ont été étudiées dans la littérature, la réflectométrie électrique a été considérée la plus efficace surtout dans le cas d'un défaut franc. Cependant, cette méthode s'avère moins fiable en présence d'un défaut non-franc caractérisé, généralement, par une signature de faible amplitude sur le réflectogramme qui dépend non seulement de la variation de l'impédance caractéristique du câble au niveau du défaut mais également de la configuration du signal de test telle que sa bande passante. En effet, l'augmentation de la fréquence maximale du signal de test améliore la résolution ''spatiale'' de l'information des défauts non-francs. Cependant, elle accentue, en même temps, les phénomènes d'atténuation et de dispersion du signal de test rendant ainsi la détection de ces défauts moins fiable, et surtout dans le cas des réseaux filaires complexes où la réflectométrie pourrait souffrir de problèmes d'ambiguïté liée à la localisation des défauts. Dans ce cadre, la réflectométrie distribuée où plusieurs capteurs sont installés aux extrémités du réseau sous test est appliquée entrainant l'apparition d'autres problématiques telles que le partage des ressources, la fusion de capteurs pour la prise de décision, la consommation d'énergie, etc.Dans ce contexte, cette thèse propose de développer deux approches : la première permet de choisie la meilleure fréquence maximale à appliquer au signal de test pour la détection des défauts non-francs. La seconde approche a pour objectif de choisir les capteurs les plus pertinents pour leur diagnostic dans les réseaux filaires complexes. Pour cela, une combinaison entre les données basées sur la réflectométrie et l'algorithme d'analyse en composantes principales (PCA) est utilisée. Le modèle de la PCA est développé pour détecter les défauts non francs existants. Associé à une analyse statistique basée sur Hotelling’s T² et Squared Prediction Error (SPE), les paramètres requis sont identifiés. Une étude expérimentale est réalisée, et une analyse de leurs performances en environnement bruité est effectuée<br>Electrical cables are used in all sectors to transfer energy or information. During operation, the cables may be subject to hard faults (open circuit, short circuit) or soft faults (isolation damage, pinching, etc.) due to misuse, environmental conditions, or aging. These faults must be detected at their earliest stage to avoid interruption of the function or more serious consequences. Even though several electric and non-electric wire diagnosis methods have been studied and developed throughout the last few decades, reflectometry-based techniques have provided effective results with hard faults. However, they have been shown to be less reliable whenever soft faults are addressed.Indeed, soft faults are characterized by a small impedance variation, resulting in a low amplitude signature on the corresponding reflectograms. Accordingly, the detection of these faults depends strongly on the test signal configuration, such as its bandwidth. Although the increase of the maximal frequency of the test signal enhances the soft fault's ''spatial'' resolution, its performance is limited by signal attenuation and dispersion. Moreover, although reflectometry offers good results in point-to-point topology networks, it suffers from ambiguity related to fault location in more complex wired networks (Multi-branched). As a solution, distributed reflectometry method, where sensors are implemented in the extremities of the network under test, is used. However, several issues are enforced, from the computing complexities and sensors fusion problems to the energy consumption.In this context, this Ph.D. dissertation proposes to develop two approaches: the first selects the best maximal frequency for soft fault detection, and the second selects the most relevant sensors to monitor and diagnose those faults in multi-branched wired networks. The proposed solution is based on a combination between reflectometry and Principal Component Analysis (PCA). The PCA model coupled with statistical analysis based on Hotelling’s T² and Squared Prediction Error (SPE) is used to detect the soft faults and select the required parameters. Experimental validation is carried out, and performance analysis in the presence of noise is investigated

The visual system receives a dynamic stream of information, but it has a limited capacity and must deploy its resources to behaviourally relevant stimuli - a process referred to as “attention”. Rapid serial visual presentation (RSVP) is an experimental method for investigating attention’s time course by presenting a rapid sequence of stimuli at a single location. Attentional selection in both naturalistic viewing and RSVP is limited by masking, and many models of selection in RSVP assume that masking terminates sensory memory for stimuli that are no longer present. However, there is indirect evidence that information about unselected RSVP stimuli may persist in a buffer despite masking. In this thesis we directly investigate buffering and selection of a cued item from one of multiple simultaneous RSVP streams. We use mixture modelling to analyse reports from only those trials in which participants identified a letter in response to the cue, and outline a novel quantitative test for buffering (Chapter 2). This provides new insights into the temporal variability of selection with exogenous and endogenous cues (Chapter 3). A series of experiments show that participants can select buffered representations, despite masking, and this appears to be related to the number of simultaneous RSVP streams (Chapter 4). We also investigate possible contributions of crowding and eccentricity to selection (Chapter 5). RSVP provides a measure of attention’s timing that replicates classic attentional effects. However, participants appear to dedicate attention to the streams prior to the cue’s appearance. When there are few streams, this leads to attentional speeds fast enough to select a stimulus representation that persists briefly, despite the masking inherent in RSVP. This falsifies theoretical claims about masking in RSVP, and demonstrates that the dynamic nature of naturalistic viewing does not prevent selection from sensory memory.

Vision systems have become ubiquitous. They are used for traffic monitoring, elderly care, video conferencing, virtual reality, surveillance, smart rooms, home automation, sport games analysis, industrial safety, medical care etc. In most vision systems, the data coming from the visual sensor(s) is processed before transmission in order to save communication bandwidth or achieve higher frame rates. The type of data processing needs to be chosen carefully depending on the targeted application, and taking into account the available memory, computational power, energy resources and bandwidth constraints. In this dissertation, we investigate how a vision system should be built under practical constraints. First, this system should be intelligent, such that the right data is extracted from the video source. Second, when processing video data this intelligent vision system should know its own practical limitations, and should try to achieve the best possible output result that lies within its capabilities. We study and improve a wide range of vision systems for a variety of applications, which go together with different types of constraints. First, we present a modulo-PCM-based coding algorithm for applications that demand very low complexity coding and need to preserve some of the advantageous properties of PCM coding (direct processing, random access, rate scalability). Our modulo-PCM coding scheme combines three well-known, simple, source coding strategies: PCM, binning, and interpolative coding. The encoder first analyzes the signal statistics in a very simple way. Then, based on these signal statistics, the encoder simply discards a number of bits of each image sample. The modulo-PCM decoder recovers the removed bits of each sample by using its received bits and side information which is generated by interpolating previous decoded signals. Our algorithm is especially appropriate for image coding.<br>Morbee, M. (2011). Optimized information processing in resource-constrained vision systems. From low-complexity coding to smart sensor networks [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/12126<br>Palancia

Web services and sensor networks have been undergoing tremendous growth during the past few years. Integrating web services technology to sensor networks domain offers many advantages. For example, the existing Internet protocols can be used instead of designing proprietary ones, 3rd party can deliver value-added web services to end users easily, etc. However, wireless sensor networks are constrained by limited system resources, such as memory, power, and computation capacity. In this thesis, we propose a web service selection algorithm by introducing QoS-based cost function concept. The mechanism we used is choosing only a few web services from each service community according to their cost function. Then we compose them together to find the optimal composite service. The proposed selection algorithm consumes only limited system resources and has the characteristics of efficiency, dynamics, and fault tolerance. It guarantees the composed web service has the optimal performance in terms of cost function.

Thesis (S.M.)--Massachusetts Institute of Technology, System Design and Management Program, February 2007.<br>Includes bibliographical references (leaves 97-100).<br>The role of data fusion in sensor platforms is becoming increasingly important in various domains of science, technology and business. Fusion pertains to the merging or integration of information towards an enhanced level of awareness. This thesis provides a canonical overview of several major fusion architectures developed from the remote sensing and defense community. Additionally, it provides an assessment of current sensors and their platforms, the influence of reliability measures, and the connection to fusion applications. We present several types of architecture for managing multi-sensor data fusion, specifically as they relate to the tracking-correlation function and blackboard processing representations in knowledge engineering. Object-Process Methods are used to model the information fusion process and supporting systems. Several mathematical techniques are shown to be useful in the fusion of numerical properties, sensor data updating and the implementation of unique detection probabilities. Finally, we discuss the importance of fusion to the concept and operation of the Semantic Web, which promises new ways to exploit the synergy of multi-sensor data platforms. This requires the synthesis of fusion with ontology models for knowledge representation. We discuss the importance of fusion as a reuse process in ontological engineering, and review key lifecycle models in ontology development. The evolutionary approach to ontology development is considered the most useful and adaptable to the complexities of semantic networks. Several potential applications for data fusion are screened and ranked according to the Joint Directors of Laboratories (JDL) process model for information fusion. Based on these predetermined criteria, the case of medical diagnostic imaging was found to offer the most promising applications for fusion, on which future product platforms can be built.<br>by Atif R. Mirza.<br>S.M.

Energy is the main bottleneck for wireless sensor networks and it has dominating effects on network lifetime. Sensors have finite energy and during the process of sensing and transmitting data to the cluster head they lose energy. Sensors that are furthest away from the cluster head require more energy than the closer ones. These losses of energy cause sensors to die faster, lower coverage area and hence network death. In this paper we investigate techniques to maximize the network lifetime by selecting the most optimal cluster head. The proposed technique is in context of cluster-based wireless sensor networks, the others being un-clustered sensor networks. The cluster head selection technique is based upon distance and node degree. The key idea of the scheme is to choose new cluster heads according to their distance and node degree toward the original cluster head location. The benefit of such a technique is to reduce the overall consumption of energy, thus increasing the network lifetime.

Kyoto University (京都大学)<br>0048<br>新制・課程博士<br>博士(農学)<br>甲第22508号<br>農博第2412号<br>新制||農||1078(附属図書館)<br>学位論文||R2||N5288(農学部図書室)<br>京都大学大学院農学研究科応用生命科学専攻<br>(主査)教授 加納 健司, 教授 三芳 秀人, 教授 宮川 恒<br>学位規則第4条第1項該当

Previous work has produced a tin dioxide based semiconducting gas sensor which is selective to CO in the presence of the lower hydrocarbons. The sensing material comprises a mixture of bismuth oxide and tin dioxide. When this mixture is heated at 800<SUP>o</SUP>C all of the bismuth oxide is converted to bismuth stannate (Bi<SUB>2</SUB>Sn<SUB>2</SUB>O<SUB>7</SUB>). This reaction commences at ≃ 650°C. This is one of a group of mixed oxides having the formula M_2Sn_2O_7 which possess a pyrochlore structure. The thesis describes the characterisation of novel sensors. These were produced with the aim of reproducing the behaviour of the original device. Sensor materials comprised mixtures of metal oxides (M_2O_3 where M = lanthanum, neodymium, samarium, gadolinium, ytterbium, dysprosium, thulium, holmium and lutetium, and also yttrium) and tin dioxide which formed pyrochlore stannates at 1500^oC. Furthermore, scandium oxide (Sc_2O_3) and antimony oxide (Sb_2O_3) mixed with tin dioxide and heated at appropriate temperatures produced stannate materials which do not possess the pyrochlore structure. Other sensor materials comprised titanium dioxide and lanthanum oxide or neodymium oxide which formed pyrochlore titanates at suitable temperatures. Tin dioxide sintered at 1500°C exhibits resistance increases or decreases upon exposure to identical gases at different operating temperatures. In a further attempt to understand the gas-sensing behaviour of these materials their response upon exposure to reducing gases, for a range of sintering temperatures ≤1500°C, was compared to the behaviour of a tin dioxide sensor identically treated. The physical properties of unheated and heated sensor materials were investigated by x-ray diffraction (room temperature and temperature-programmed), fourier transform infra-red (FTIR) spectroscopy, thermogravimetry and differential scanning calorimetry (TG/DSC) and B.E.T. surface area equipment. Hence the temperature of production of stannate and titanate compounds and the chemical and physical processes which occurred during heating were determined.

Direct potentiometric techniques are becoming an invaluable analytical resource in the clinical laboratory. One important aspect of their use is determining or monitoring the activities of ions present in physiological media such as blood, urine, cerebral fluid, serum etc. This work is based, mainly, on the measurement of the blood electrolytes sodium, potassium, calcium and pH. Various automatic, multi-electrolyte, direct potentiometric, ion-selective analysers are available. The problem, to date, has been twofold - firstly different instruments give different results with the same sample and secondly, direct potentiometry senses the activity of the free (hydrated) ions - an entity unfamiliar to clinicians who have been used to flame photometric concentrations. To resolve these problems it is necessary to establish an operational pION scale by international consensus. This study was aimed at contributing towards achieving this end. The first requirement for the establishment of a pION scale is a set of universally accepted standards. Multi-electrolyte calibration standards using "Good" buffers have been proposed. Analyte binding to these buffers have been evaluated using two non-linear least squares programs SCOGS2 and SUPERQUAD . Commercial and home made ion-selective electrodes and ion-selective field effect transistors were calibrated and tested in the calibration solutions by measuring transfer potentials. A flow through rig was set up using several different ion-selective electrodes from manufacturers' instruments but with a common reference electrode and this system was tested with the aqueous calibration solutions, plasma and serum. The second requirement for the establishment of a pION scale is, resolution of the problem of residual liquid junction potentials. There are two aspects of this problem. The first aspect is the variations arising from the difference between the 'junction' potential with the calibration solution and the junction potential with plasma. The second aspect is the 'inter' instrument differences arising because of choice of ions in the salt bridge solutions and different liquid junction geometries. To minimise the first aspect, the ionic composition of the calibration solutions were formulated to reflect the ionic composition of plasma. Residual liquid junction potentials of the calibration solutions against the NBS blood phosphate buffer were determined. The second aspect has been discussed. Other aspects of standardisation such as sample pre-treatment and measurement protocol have been studied briefly. These include heparin binding and carbon-dioxide contamination. Reporting results in activities based on activity co-efficients calculated by Covington and Ferris (Pitzer and Hydration models) rather than the conventional flame photometric concentrations, has also been discussed.

It is often the case that small faults in a structure lead to irreparable damages that deliver a huge financial loss or even pose safety risks. Thus, an early fault detection is necessary, such that these unfortunate events can be avoided. In this thesis, the problem of structural damage detection is considered. In particular there are 3 main contributions: First, a novel sensors selection algorithm based on the concepts of entropy and information gain from information theory is developed, to reduce the number of sensors without affecting, or even improving, model accuracy; Second, a novel technique based on Kalman filtering and on a combination of Regression Trees theory from Machine Learning and Auto Regressive (AR) system identification from control theory is derived, to build models that can be used to detect structural damages. Finally, a new fault detection algorithm based on Poly-Exponential (PE) models and nonlinear Kalman filtering on the residual is introduced, which is able to enhance the sensitivity of the proposed fault detection algorithm and improve the data prediction quality for some accelerometers in a notably margin. The presented techniques are validated on three different experimental datasets, providing evidence that the proposed algorithms outperform some previous approaches, improving the prediction accuracy and the damage detection sensitivity while reducing the number of sensors.

Anions are ubiquitous in a myriad of chemical, biological and environmental processes. The ability to selectively recognise target anions in competitive solvents has potential uses in biomedical and environmental applications. Macrocyclic structures such as cyclic peptides in particular, provide excellent scaffolds for anionic recognition as they allow for a preorganised binding site; however, their synthesis is laborious and typically involves numerous purification steps. Therefore, the work described in this thesis focuses on the development of small molecular scaffolds for the selective recognition of anions. Chapter 2 describes the design of a family of novel linear peptide-based bis[zinc(II)– dipicolylamine] receptors and an investigation of their anion binding abilities. In order to maximise the effectiveness of the synthetic strategy, the solid phase peptide synthesis approach was employed. An investigation of the binding abilities of these receptors under mimicked physiological conditions was carried out using an indicator displacement assay approach with pyrocatechol violet as the indicator. Based on the results of the UV-Vis spectroscopic binding studies performed with these receptors, it was found that all receptors exhibit excellent affinity for the target anion, pyrophosphate, with observed selectivity over adenosine triphosphate and adenosine diphosphate. Chapter 3 describes the design of two new covalently linked receptors [a dipicolylamine–naphthalimide derivative and a bis(isoquinoline) derivative] and an investigation of their metal and anion sensing properties. Preliminary metal sensing studies under mimicked physiological conditions revealed that for both derivatives, coordination to different metal ions afforded different fluorescence outputs. Furthermore, anion sensing studies using the respective dipicolylamine–naphthalimide and bis(isoquinoline) metal complexes highlighted the potential of these two derivatives operating as multi anion probes upon complexation to different metal ions. Chapter 4 describes the design of a series of novel linear peptide-based receptors (decorated with either a thiourea group or a squaramide moiety) and an investigation of their anion binding abilities. In a similar manner to the synthetic strategy employed in Chapter 2, we have also utilised the solid phase peptide synthesis approach for the preparation of these linear peptide-based receptors. 1H NMR spectroscopic titrations performed in 0.5% v/v H2O/DMSO-d6 and in a more competitive media of 20% v/v H2O/DMSO-d6 indicated that all receptors demonstrate strong binding affinity and selectivity for the target anion, sulfate. Such remarkable binding to sulfate is proposed to arise from a synergistic interaction between the amide backbone and the thiourea/squaramide NH protons analogous to that observed in the sulfate binding protein. The final chapter, Chapter 5, describes the first study of a novel family of croconamides as anion receptors. Access to these compounds was achieved via a three-step route from croconic acid. 1H NMR titrations with these croconamides in both CD3CN-d3 and DMSO-d6 revealed a strong preference for the chloride anion. Interestingly, unique binding behaviours that are contrary to those reported for a structurally similar class of compounds (oxosquaramides) were observed.

ITC/USA 2013 Conference Proceedings / The Forty-Ninth Annual International Telemetering Conference and Technical Exhibition / October 21-24, 2013 / Bally's Hotel & Convention Center, Las Vegas, NV<br>Relay selection for optimal communication as well as multiple source localization is studied. We consider the use of dual-role nodes that can work both as relays and also as anchors. The dual-role nodes and multiple sources are placed at fixed locations in a two-dimensional space. Each dual-role node estimates its distance to all the sources within its radius of action. Dual-role selection is then obtained considering all the measured distances and the total SNR of all sources-to-destination channels for optimal communication and multiple source localization. Bit error rate performance as well as mean squared error of the proposed optimal dual-role node selection scheme are presented.

Zinc is an important micronutrient but the biological function of its labile form is poorly understood. Zinc selective fluorescence sensors, recognized as the major tool to gain information about the role of zinc in living systems, have been attracting more and more interest. The most promising solution currently being studied comes in the form of ratiometric sensors. Unlike sensors based on the switch-on mechanism, ratiometric sensors determine the free metal concentration directly from the ratio of the emission intensities at two wavelengths. The major restriction on the design of this type of sensor is from the necessity for a spectral-shift upon binding metal ions. To develop novel ratiometric sensors, we have developed designs based on excited-state intramolecular proton transfer (ESIPT). In the absence of ZnII at neutral pH, the 2-(2 -sulfonamidophenyl)benzimidazole family undergoes ESIPT to yield a highly Stokes-shifted emission from the proton-transfer tautomer. Coordination of ZnII inhibits the ESIPT process and yields a significant hypsochromic shift of the fluorescence emission maximum. By implementing structural modifications, we were able to gauge free ZnII concentrations in the millimolar to picomolar range. To tune the peak excitation towards lower energy, a property that is of particular importance in the light of biological applications, we modified the platform molecule with extended pi-conjugation and by substituent engineering. The position of the modification and the nature of the substituents strongly influenced the photophysical properties of the investigated derivatives. Several fluorophores revealed emission ratiometric properties with a large dynamic range combined with a peak absorption beyond 350 nm, rendering these probes promising candidates for applications. To further understand the origin of the substituent effect, we studied five derivatives for the solvatochromic shift analysis and quantum chemical studies. The results showed that the negative solvatochromic shift behavior was most pronounced in protic solvents presumably due to specific hydrogen-bonding interactions. The extrapolated gas-phase emission energies correlated qualitatively with the trends in Stokes shifts, suggesting that solute-solvent interactions do not play a significant role in explaining the divergent emission energy shifts. Detailed quantum chemical calculations not only confirmed the moderately polarized nature of the ESIPT tautomers but also provided a rationale for the observed emission shifts based on the differential change in the HOMO and LUMO energies. This study revealed the great potential of 2-(2 -arylsulfonamidophenyl)- benzimidazoles, such as tunable peak absorption and emission, a very wide dynamic range regarding to zinc binding, very little solvent polarity dependence, and especially, the emission ratiometric property. All these properties make this system a unique candidate to tackle the problems in the research of zinc biology.

In Underwater Acoustic Sensor Networks (UWASNs), cameras have recently been deployed for enhanced monitoring. However, their use has faced several obstacles. Since video capturing and processing consume significant amounts of camera battery power, they are kept in sleep mode and activated only when ultrasonic sensors detect a target. The present study proposes a camera relocation structure in UWASNs to maximize the coverage of detected targets with the least possible vertical camera movement. This approach determines the coverage of each acoustic sensor in advance by getting the most applicable cameras in terms of orientation and frustum of camera in 3-D that are covered by such sensors. Whenever a target is exposed, this information is then used and shared with other sensors that detected the same target. Compared to a flooding-based approach, experiment results indicate that this proposed solution can quickly capture the detected targets with the least camera movement.

Chemical sensors are generally used as one-dimensional devices, where one measures the sensor’s response at a fixed setting, e.g., infrared absorption at a specific wavelength, or conductivity of a solid-state sensor at a specific operating temperature. In many cases, additional information can be extracted by modulating some internal property (e.g., temperature, voltage) of the sensor. However, this additional information comes at a cost (e.g., sensing times, power consumption), so offline optimization techniques (such as feature-subset selection) are commonly used to identify a subset of the most informative sensor tunings. An alternative to offline techniques is active sensing, where the sensor tunings are adapted in real-time based on the information obtained from previous measurements. Prior work in domains such as vision, robotics, and target tracking has shown that active sensing can schedule agile sensors to manage their sensing resources more efficiently than passive sensing, and also balance between sensing costs and performance. Inspired from the history of active sensing, in this dissertation, we developed active sensing algorithms that address three different computational problems in chemical sensing. First, we consider the problem of classification with a single tunable chemical sensor. We formulate the classification problem as a partially observable Markov decision process, and solve it with a myopic algorithm. At each step, the algorithm estimates the utility of each sensing configuration as the difference between expected reduction in Bayesian risk and sensing cost, and selects the configuration with maximum utility. We evaluated this approach on simulated Fabry-Perot interferometers (FPI), and experimentally validated on metal-oxide (MOX) sensors. Our results show that the active sensing method obtains better classification performance than passive sensing methods, and also is more robust to additive Gaussian noise in sensor measurements. Second, we consider the problem of estimating concentrations of the constituents in a gas mixture using a tunable sensor. We formulate this multicomponent-analysis problem as that of probabilistic state estimation, where each state represents a different concentration profile. We maintain a belief distribution that assigns a probability to each profile, and update the distribution by incorporating the latest sensor measurements. To select the sensor’s next operating configuration, we use a myopic algorithm that chooses the operating configuration expected to best reduce the uncertainty in the future belief distribution. We validated this approach on both simulated and real MOX sensors. The results again demonstrate improved estimation performance and robustness to noise. Lastly, we present an algorithm that extends active sensing to sensor arrays. This algorithm borrows concepts from feature subset selection to enable an array of tunable sensors operate collaboratively for the classification of gas samples. The algorithm constructs an optimized action vector at each sensing step, which contains separate operating configurations for each sensor in the array. When dealing with sensor arrays, one needs to account for the correlation among sensors. To this end, we developed two objective functions: weighted Fisher scores, and dynamic mutual information, which can quantify the discriminatory information and redundancy of a given action vector with respect to the measurements already acquired. Once again, we validated the approach on simulated FPI arrays and experimentally tested it on an array of MOX sensors. The results show improved classification performance and robustness to additive noise.

碩士<br>國立臺灣科技大學<br>電子工程系<br>94<br>Generally, color matching function as sensor of color stimulus is the common way to do the spectral reconstruction. However, the result of reconstruction is not as good as using least square estimation (LSE ) method, and the choice of sensor is usually neglected in spectral reconstruction. This paper propose a method of choosing sensors which can pick out N sensors as the best set of M sensors .The simulation result shows that the method we proposed can improve the Spectral reconstruction obviously.