Rozprawy doktorskie na temat „SENSITIVE QAC”

This thesis explores two novel ways of reducing the data complexity of tactile sensing. The thesis begins by examining the state-of-the art in tactile sensing, not only examining the sensor construction and interpretation of data but also the motivation for these designs. The thesis then proposes two methods for reducing the complexity of data in tactile sensing. The first is a low-power tactile sensing array exploiting a novel application of a pressure-sensitive material called quantum tunnelling composite. The properties of this material in this array form are shown to be beneficial in robotics. The electrical characteristics of the material are also explored. A bit-based structure for representing tactile data called Bitworld is then defined and its computational performance is characterised. It is shown that this bit-based structure outperforms floating-point arrays by orders of magnitude. This structure is then shown to allow high-resolution images to be produced by combining low resolution sensor arrays with equivalent functional performance to a floating-point array, but with the advantages of computational efficiency. Finally, an investigation into making Bitworld robust in the presence of positional noise is described with simulations to verify that such robustness can be achieved. Overall, the sensor and data structure described in this thesis allow simple, but effective tactile systems to be deployed in robotics without requiring a significant commitment of computational or power resources on the part of a robot designer.

Presented in this thesis is a summary of the development of gallium arsenide position sensitive detectors. These are aimed at various applications with the original motivation being experimental particle physics. An account is given of basic semiconductor physics relevant to particle detectors. A review of electronics applicable to reading out charged signals from semiconductor detectors is included. Applications of gallium arsenide X-ray detectors are described together with results from a Monte-Carlo simulation of the spectrum obtained from an X-ray source. The design, fabrication and laboratory testing of detectors is presented for pixel and microstrip detectors and other test structures. Test beam results are also presneted for pixel and microstrip detectors. An introduction to ATLAS detector simulation is also given, with examples of detector descriptions for the GaAs Forward Semiconductor Tracker. Results from a generator level study of b-jets from the process Hbb and predictions of influences in the Forward Tracker are also given.

This thesis describes an experimental study of the heat transport between the tip of a Scanning Thermal Microscope –Atomic Force Microscope (SThM-AFM) and nanofabricated sample. SThM is one of the most flexible tools available for measuring thermal transport at the nanoscale. However, heat transfer between the probe and sample has a complicated nature as it depends on different parameters such as the nature of the surrounding gas (pressure, temperature, degree of humidity), and that of the mechanical contact between probe tip and sample (surface roughness and topography, mechanical properties of the tip surface and sample surface and the force applied). So understanding the heat flow between tip and sample requires careful study of all of these parameters. Thus quantification of tip and sample thermal contact is a key problem in the interpretation of SThM measurements. In this work the fabrication of devices to quantify SThM tip and sample surface thermal resistance in one single contact are presented. The fabrication was performed using lithographic nanofabrication and MEMS technology. Thermally grown silicon dioxide was chosen as a membrane material for all devices due to its low thermal and electrical conductivity and ease of fabrication.

This thesis consists of three topics whose over-arching theme is based on risk sensitive stochastic control. In the �first topic (chapter 2), we study a problem on benchmark out-performance. We model this as a zero-sum risk-sensitive stochastic game between an investor who as a player wants to maximize the risk-sensitive criterion while the other player ( a stochastic benchmark) tries to minimize this maximum risk-sensitive criterion. We obtain an explicit expression for the strategies for both these two players. In the second topic (chapter 3), we consider a finite horizon risk-sensitive asset management problem. We study it in the context of a zero-sum stochastic game between an investor and the second player called the "market world" which provides a probability measure. Via this game, we connect two (somewhat) disparate areas in stochastics; namely, stochastic stability and risk-sensitive stochastic control in mathematical finance. The connection is through the Follmer-Schweizer minimal martingale measure. We discuss the impact of this measure on the investor's optimal strategy. In the third topic (chapter 4), we study the sufficient stochastic maximum principle of semi-Markov modulated jump diffusion. We study its application in the context of a quadratic loss minimization problem. We also study the finite-horizon risk-sensitive optimization in relation to the underlying sufficient stochastic maximum principle of a semi-markov modulated diffusion.

Phase sensitive amplification (PSA) is a remarkably powerful tool if implemented correctly - it allows amplification with a noise figure lower than quantum mechanics normally dictates, and also allows the development of systems that coherently process the phase of their optical inputs. PSA at infrared wavelengths can be achieved using second or third order optical nonlinearities - indeed the first demonstrations of PSA utilised degenerate three photon mixing in bulk crystals, allowing the observation of effects such as quantum quadrature squeezing. This PhD research project was aimed at translating the fascinating science of PSAs into applications, based on Kerr nonlinearity in optical fibres, capable of being deployed in modern core optical networks runnning at 40 Gbit/s and higher. This objective was an integral part of the European Union FP7 project PHASORS. Studies, both theoretical and experimental, were carried out on wideband non-degenerate PSAs. The inline cascaded fiber optic parametric amplifier (FOPA), in which a first phase insensitive FOPA is used to generate phase locked signal-idler pairs, followed by a second FOPA in which PSA takes place, was used to experimentally demonstrate PSA gain characteristics in linear and saturated modes of operation, with PSA obtained over 20 nm in the telecom C-band. Focus was then re-directed towards applying a dual pump PSA to DPSK regeneration. The amplitude and phase limiting characteristics of these devices were experimentally studied, in particular revealing that amplitude saturation arises due to an interplay between input phase and nonlinear phase matching along the nonlinear fibre. This feature was used to identify a regime of operation for DPSK regeneration combining simultaneous phase and amplitude regeneration. A practical DPSK regenerator based on a degenerate dual pump phase sensitive FOPA was built. The device advanced the state of the art by incorporating a pump synthesis stage that allowed black-box operation. Detailed measurements using noise sources with varying frequency distributions in both amplitude and in phase, are presented, and the ability of the system to improve the phase and amplitude characteristics of signals at its input was verified. Also presented are results showing the regenerator installed in the middle of an 800 km dark fiber link. Finally, a novel scheme was proposed and demonstrated that utilised parametric mixing to perform arbitrary phase quantization. This relied on the coherent addition of an M-level PSK signal with a conjugated (M-1)th phase harmonic, scaled by a coefficient m, to achieve M-level phase quantization. The concept was successfully demonstrated with QPSK data signals at 56 Gbaud, as well as used to quantize lower bandwidth test signals from 2 to 6 phase levels.

The upgrade of the Continuous Electron Beam Accelerator Facility (CEBAF) Large Acceptance Spectrometer (CLAS) to CLAS12 will offer unique possibilities to study the strong interaction and the internal nucleon dynamics. For this, excellent hadron identification over its full kinematic range is essential and a Ring Imaging CHerenkov detector (RICH) has been proposed for installation into CLAS12 to achieve this. This thesis describes studies performed towards the selection of a photon detector for use in the CLAS12 RICH, which will be a crucial component for the success of the RICH performance. MultiAnode PhotoMultiplier Tubes (MAPMTs) have been selected as the most promising photon detectors for the CLAS12 RICH. A high-precision laser test-stand was developed to characterise candidate MAPMTs and the results were used to determine the Hamamatsu H8500 MAPMT as the optimal device. Throughout the laser tests the H8500 MAPMT proved to be single photon sensitive as demanded, the spatial resolution was found to satisfy the required value of less than 1cm by 1cm and the device exhibited sufficiently low crosstalk levels of less than 3%. The response and crosstalk of the device showed a dependency upon the MAPMT construction, the magnitude of which was shown to be negligible and overall the H8500 provides a plane of suitable uniformity to satisfy the imaging functionality of the CLAS12 RICH. To further confirm the H8500 MAPMT choice, its performance in response to Cherenkov light within two prototype setups was evaluated. First a small-scale prototype was designed and constructed, incorporating one H8500 MAPMT to image Cherenkov rings created by cosmic muons traversing aerogel radiators. Extraction of the results required full understanding of the device based upon calibrations made with the laser tests. The prototype was also simulated in detail, allowing for a model description of the MAPMT to be validated. Secondly, 28 H8500 MAPMTs were used in a large-scale prototype to image Cherenkov rings produced by mixed hadrons traversing aerogel radiators, to evaluate their performance in a geometry and an environment similar to that expected in the CLAS12 RICH. Both prototypes revealed that the H8500 MAPMT can be used to successfully detect the required 7 photons per Cherenkov event to achieve pion/kaon separation in the CLAS12 RICH. Furthermore, a pion/kaon separation of more than 3 sigma at 6GeV/c was observed through a preliminary analysis of data extracted with the large-scale prototype. The prototype studies also confirmed the low-noise behaviour of the H8500 MAPMT. As a result of the laser test-stand and prototype tests, the choice of the H8500 MAPMT as the photon detector for the CLAS12 RICH was validated.

In this theses I will concentrate on femtosecond laser induced modification in silica glass. One type of modification in fused silica is subwavelength nanogratings. This modification exhibits form birefringence and therefore optical elements can be fabricated based on nanogratings. The main goal of my work was development and fabrication of practical optical elements based on femtosecond laser induced nanogratings. In order to be able to fabricate optical elements, laser induced modification had to be uniform and exhibit strong birefringence. Also fabrication speed has to be high in order to write reasonable size optical element in short time. Optimal fabrication parameters were found by studying physical and optical properties of nanogratings. The first polarization sensitive optical element to be demonstrated was polarization converter, which transforms Gaussian beam with uniform polarization distribution into a cylindrical vector beam. Later on, polarization converters for different types of laser beams and spectral ranges were demonstrated. Properties of generated beams were measured and compared with theoretical modelling. Generated beams were also successfully exploited for material modification and results are compared with theoretical simulations. Same converters can be also used for optical vortices generation. Unique property of such converters is that they allow tuning orbital angular momentum of photons in the light beam and therefore adjusting the torque of the vortex beam. The important feature of this vortex generation technique is that optical angular momentum is changed without affecting intensity distribution of the laser beam. The benefits of tunable optical angular momentum were demonstrated in optical trapping experiment where silica micro spheres were trapped into a rotating ring is performed. The ability to tune the torque of the beam is proved by changing rotation speed change of the beads’ ring. Birefringence can be exploited not only for polarization control but also for phase. Half-wave retarder can induce phase delay to circularly polarized light depending on the azimuth of slow axis of the retarder. Based on that, converter for “diffraction-free” Airy and Bessel beams are fabricated. Properties of generated beams are measured and compared with theoretical simulations. Airy beam converter based on laser induced nanogratings has unique ability to generate dual Airy beam which is superposition two Airy beams accelerating in opposite direction. Generated Airy beam is also used in material modification. Self-detachment of fibre-like structure is observed from laser modified substrate which is speculated to be caused by properties of Airy beam. Finally, an array of micro waveplates is fabricated which together with a linear polarizer and a CCD camera allows real time polarization imaging. Full state of polarization can be measured. For the demonstration, polarization distribution of cylindrical vector beams was measured.

This thesis studies the risk-sensitive control problem for a class of non-linear stochastic systems and its financial applications. The nonlinearity is of the square- root type, and is inspired by applications. The problems of optimal investment and consumption are also considered under several different assumptions on the stochastic interest rate and stochastic volatility. At the beginning, we systematically investigate the nonlinearity of risk-sensitive control problem. It consists of quadratic and square-root terms in the state. Such an optimal control problem can be solved in an explicit closed form by the completion of squares method. As an application of the risk-sensitive control in financial mathematics, the optimal investment problem will be described in the Chapter 4. A new interest rate, which follows the stochastic process with mixed Cox- Ingersoll-Ross (CIR) model and quadratic affine term structure model (QATSM) is introduced. Such an interest rate model admits an explicit price for the zero- coupon bond. In Chapter 5, we consider a portfolio optimization problem on an infinite time horizon. The stochastic interest rate consists not only of the quadratic terms, but also of the square-root terms. On the other hand, the double square root process is also introduced to establish the interest rate model. Under some sufficient conditions, the unique solution of the optimal investment problem is found in an explicit closed form. Furthermore, the optimal consumption problem is considered in Chapter 6 and 7. It can be solved in an explicit closed form via the methods of completion of squares and the change of measure. We provide a detailed discussion on the existence of the optimal trading strategies. Such trading strategies can be deduced for both finite and infinite time horizon cases.

This thesis describes the development, characterisation and application of a portable spectroscopy system for ultra-sensitive, real-time detection of breath ethane. In healthcare, breath ethane is a widely accepted marker of free radical-induced cell damage and may be used to indicate changes in oxidative stress. The aim was to deliver a compact instrument capable of long-term, on-site use in a clinical environment, while also retaining the high performance previously achieved by lab-based systems at the University of Glasgow. The newly developed instrument has a sensitivity of 70 parts per trillion with a 1 Hz sampling rate. The system incorporates a cryogenicallycooled lead-salt laser and uses a second derivative wavelength modulation detection scheme. A thermally-managed closed-loop refrigeration system has eliminated the need for liquid coolants. The instrument has been field-tested to ensure target performance is sustained in a range of environments, both indoor and outdoor. It has since been used in a number of pilot clinical studies, both off-site and on-site, in which breath ethane was monitored as a marker of oxidative stress. The three main clinical areas investigated were dialysis, radiotherapy and intensive care. In the intensive care study, the instrument was modified to enable automatic breath sampling of inspired and expired gases of ventilated patients. This technique proved highly successful and the instrument then remained at the Southern General hospital, where it continued to be used as part of a wider study into breath ethane in intensive care patients. The use of the new spectroscopy system has enabled ultra-sensitive, rapid analysis of a large number of breath samples. The use of the new instrument, in particular for continual breath monitoring, has enabled the detection of short-lived fluctuations in breath ethane, yielding some interesting findings in a number of pilot clinical studies. Our results suggest that breath ethane may be used as an indicator of dynamic changes in oxidative stress. Further studies will be required to determine if such monitoring is of clinical benefit. Chapter 1 gives a general introduction to spectroscopy and some background to our project. A number of spectroscopic techniques and laser sources are discussed, along with a review of previous work in ethane detection. In chapter 2 some background theory of molecular spectroscopy is given, with a more detailed discussion of the wavelength modulation technique. Chapter 3 describes in detail the development of the portable spectroscopy system. The achieved performance and factors contributing to this performance are discussed in chapter 4. The field test of the instrument is reported on in chapter 5. In chapter 6 the application of the technology to breath analysis and the current challenges in this field are discussed. Example breath ethane measurements for healthy controls are provided. The clinical pilot studies conducted using the new system in areas of dialysis, intensive care and radiotherapy are discussed in chapters 7, 8, and 9 respectively. Chapter 10 contains the thesis summary and conclusions, with suggestions for future work.

A Flight Management System (FMS) performs 4 Dimensional flight planning
Lateral Planning (Calculation of the latitude and longitudes of waypoints), Vertical Planning (Calculation of the altitudes of waypoints) and Temporal Planning(Calculation of Estimated Time of Arrival). Correct and accurate calculation of4D flight path and then guiding the pilot/airplane to track the route in specified accuracy limits in terms of lateral (i.e Required Navigational Performance RNP), vertical (Reduced Vertical Seperation Minima RVSM), and time (Required Time of Arrival RTA) is what FMS performs in brief. Any deviation of planned input values versus actual input values, especially during the emergency cases (i.e burning outoneof engines etc.), causes the aircraft to deviate the plan and requires replanning now taking into consideration the currentsituation. In emergency situations especially in Oceaning Flights (flights whose cruise phase lasts more than 5 hour is called as &ldquo
Oceaning Flights&rdquo
) Optimum Cruise Phase Flight Route Planning plays a vital role. In avionics domain &ldquo
Optimum&rdquo
does not mean &ldquo
shortest path&rdquo
mainly due to the effect of weather data as wind speed and direction directly affects the groundspeed. In the scope of the current thesis, an algorithm employing dynamic programming paradigms will be designed and implemented to find the optimum flight route planning. A top down approach by making use of aircraft route planning ontology will be implemented to fill the gap between the flight plan specific domain knowledge and optimization techniques employed. Where as the algorithm will be generic by encapsulating the aircraft&rsquo
s performance characteristics
it will be evaluated on C-130 aircraft.

Dans le monde d'aujourd'hui où la plupart des aspects de la vie moderne sont traités par des systèmes informatiques, la vie privée est de plus en plus une grande préoccupation. En outre, les données ont été générées massivement et traitées en particulier dans les deux dernières années, ce qui motive les personnes et les organisations à externaliser leurs données massives à des environnements infonuagiques offerts par des fournisseurs de services. Ces environnements peuvent accomplir les tâches pour le stockage et l'analyse de données massives, car ils reposent principalement sur Hadoop MapReduce qui est conçu pour traiter efficacement des données massives en parallèle. Bien que l'externalisation de données massives dans le nuage facilite le traitement de données et réduit le coût de la maintenance et du stockage de données locales, elle soulève de nouveaux problèmes concernant la protection de la vie privée. Donc, comment on peut effectuer des calculs sur de données massives et sensibles tout en préservant la vie privée. Par conséquent, la construction de systèmes sécurisés pour la manipulation et le traitement de telles données privées et massives est cruciale. Nous avons besoin de mécanismes pour protéger les données privées, même lorsque le calcul en cours d'exécution est non sécurisé. Il y a eu plusieurs recherches ont porté sur la recherche de solutions aux problèmes de confidentialité et de sécurité lors de l'analyse de données dans les environnements infonuagique. Dans cette thèse, nous étudions quelques travaux existants pour protéger la vie privée de tout individu dans un ensemble de données, en particulier la notion de vie privée connue comme confidentialité différentielle. Confidentialité différentielle a été proposée afin de mieux protéger la vie privée du forage des données sensibles, assurant que le résultat global publié ne révèle rien sur la présence ou l'absence d'un individu donné. Enfin, nous proposons une idée de combiner confidentialité différentielle avec une autre méthode de préservation de la vie privée disponible.
In nowadays world where most aspects of modern life are handled and managed by computer systems, privacy has increasingly become a big concern. In addition, data has been massively generated and processed especially over the last two years. The rate at which data is generated on one hand, and the need to efficiently store and analyze it on the other hand, lead people and organizations to outsource their massive amounts of data (namely Big Data) to cloud environments supported by cloud service providers (CSPs). Such environments can perfectly undertake the tasks for storing and analyzing big data since they mainly rely on Hadoop MapReduce framework, which is designed to efficiently handle big data in parallel. Although outsourcing big data into the cloud facilitates data processing and reduces the maintenance cost of local data storage, it raises new problem concerning privacy protection. The question is how one can perform computations on sensitive and big data while still preserving privacy. Therefore, building secure systems for handling and processing such private massive data is crucial. We need mechanisms to protect private data even when the running computation is untrusted. There have been several researches and work focused on finding solutions to the privacy and security issues for data analytics on cloud environments. In this dissertation, we study some existing work to protect the privacy of any individual in a data set, specifically a notion of privacy known as differential privacy. Differential privacy has been proposed to better protect the privacy of data mining over sensitive data, ensuring that the released aggregate result gives almost nothing about whether or not any given individual has been contributed to the data set. Finally, we propose an idea of combining differential privacy with another available privacy preserving method.

In this thesis we present a theoretical approach to the study of the charge recombination reaction in dye sensitised solar cells. An expression for the charge transfer rate at the semiconductor-molecule interface has been derived in chapter 2, within the theoretical framework of the non-adiabatic electron transfer. Computational procedures are presented in chapters 3 and 4 for realistic systems and the comparison with experimental data is reported. In chapter 5 we explore some strategies to impede the charge recombination reaction and we assess the impact of lattice defects on these strategies. In chapters 6 and 7 we present two innovative approaches to minimise the charge recombination rate, based on the partitioning scheme applied in chapter 2. We then suggest alternative designs for the molecular sensitiser and test the rationale of the strategies proposed with calculations of realistic systems at the Density Functional level of theory. A phenomenological model of the electron at the semiconductor-solvent interface is presented in chapter 8 to test one of the assumptions underpinning the theoretical scheme.

This thesis presents a comprehensive example framework on how current multiple model state estimation algorithms for jump Markov linear systems can be improved. The possible improvements are categorized as: -Design of multiple model state estimation algorithms using new criteria. -Improvements obtained using existing multiple model state estimation algorithms. In the first category, risk-sensitive estimation is proposed for jump Markov linear systems. Two types of cost functions namely, the instantaneous and cumulative cost functions related with risk-sensitive estimation are examined and for each one, the corresponding multiple model estate estimation algorithm is derived. For the cumulative cost function, the derivation involves the reference probability method where one defines and uses a new probability measure under which the involved processes has independence properties. The performance of the proposed risk-sensitive filters are illustrated and compared with conventional algorithms using simulations. The thesis addresses the second category of improvements by proposing -Two new online transition probability estimation schemes for jump Markov linear systems. -A mixed multiple model state estimation scheme which combines desirable properties of two different multiple model state estimation methods. The two online transition probability estimators proposed use the recursive Kullback-Leibler (RKL) procedure and the maximum likelihood (ML) criteria to derive the corresponding identification schemes. When used in state estimation, these methods result in an average error decrease in the root mean square (RMS) state estimation errors, which is proved using simulation studies. The mixed multiple model estimation procedure which utilizes the analysis of the single Gaussian approximation of Gaussian mixtures in Bayesian filtering, combines IMM (Interacting Multiple Model) filter and GPB2 (2nd Order Generalized Pseudo Bayesian) filter efficiently. The resulting algorithm reaches the performance of GPB2 with less Kalman filters.

Wide band gap metal oxides have recently become one of the most investigated materials in surface science. Among these metal oxides especially TiO2 attracts great interest, because of its wide range applications, low cost, biocompatibility and ease of analysis by all experimental techniques. The usage of TiO2 as a component in solar cell technology is one of the most investigated applications of TiO2 . The wide band gap of TiO2 renders it inecient for isolated use in solar cells. TiO2 surface are therefore coated with a dye in order to increase eciency. This type of solar cells are called dye sensitized solar cells . The eciency of dye sensitized solar cells is directly related with the absorbed light portion of the entire solar spectrum by the dye molecule. Inspite of the early dyes, recent dye molcules, which are called wider wavelength response dye molecules, can absorb a larger portion of entire solar spectrum. Thus, the eciency of dye sensitized solar cells is increased by a considerably amount. In this thesis the electronic structure of organic rings, which are the fundamental components of the dye molecules, adsorbed on anatase (001) surface is analyzed using density functionaltheory. The main goal is to obtain a trend in the electronic structure of the system as a function of increasing ring number. Electronic structure analysis is conducted through band structure and density of states calculations. Results are presented and discussed in the framework of dye sensitized solar cells theory.

Query Autocompletion is a leading attribute of Search Engines which makes the user’s search experience better by predicting the query. QAC methods suggest query suggestions to users, after they enter some of the keystrokes in the search engine. This is done by predicting the query using past query logs and other trends. Current QAC methods use the Most Popular Completions as the suggestion results. Context and Personalized techniques are proposed already but they are used separately. The present methods being incorporated are the location and past searches sensitive QAC. In this proposed work of thesis, we will talk about a hybrid technique by combining both the context sensitive, trending and personalized suggestions. The improvements which are made in the base paper are that a new approach can be proposed by combining the three techniques to create a hybrid technique. It intends to incorporate three major research works: Time sensitive (based on time series and trends), Context Sensitive (based on recent searches done) and Personalized (based on gender, location and age-group) query auto completion. Thus an algorithm that considers all these parameters will be better at predicting the user query. The results predicted are better in reducing the user keystrokes during the search and also reduces the searching time, and also enhances the reliability of the search engine. Further improvements can be done by extracting the user’s browsing history to determine keywords, interests and other user-specific data for enhancing the result predictions.