The research investigates several critical design issues of continuous-time (CT) [Sigma Delta] modulators. The first is to investigate the sensitivity of CT [Sigma Delta] modulators to high-frequency clock spurs. These spurs down-convert the high-frequency quantization noise, degrading the dynamic range of the modulator. The second is to study the robustness of continuous-time loop filters under large RC product variations. Large RC variations in the CMOS process strongly degrade the performance of continuous-time [Sigma Delta] modulators, and reduce the production yield. The third is to model the harmonic distortion of one-bit continuous-time [Sigma Delta] modulators due to the interaction between the first integrator and the feedback digital-to-analog converter (DAC). A closed-form expression of the 3'rd-order harmonic distortion is derived and verified. Conventional CT [Sigma Delta] modulators employ all active integrators: each integrator needs an active amplifier. The research proposes a 5th-order continuous-time [Sigma Delta] modulator with a hybrid active-passive loop filter consisting of only three amplifiers. The passive integrators save power, and introduce no distortion. The active integrators provide gain and minimize internal noise contributions. A single-bit switched-capacitor DAC is employed as the main feedback DAC for high clock jitter immunity. An additional current steering DAC stabilizes the loop with the advantage of simplicity. To verify the proposed techniques, a prototype continuous-time [Sigma Delta] modulator with 2-MHz signal bandwidth is designed in a 0.25-¹m CMOS technology targeting for GPS or WCDMA applications. The experimental results show that the prototype modulator achieves 68-dB dynamic range over 2-MHz bandwidth with a 150-MHz clock, consuming 1.8 mA from a 1.5-V supply.

Wireless sensor networks (WSN) are an emerging technology with a wide range of applications including environment monitoring, security and surveillance, health care, smart homes, etc. Subject to severe resource constraints in wireless sensor networks, in this research, we address the distributed estimation of unknown parameters by studying the correlation among resource, distortion, and lifetime, which are three major concerns for WSN applications. The objective of the proposed research is to design efficient distributed estimation algorithms for resource-constrained wireless sensor networks, where the major challenge is the integrated design of local signal processing operations and strategies for inter-sensor communication and networking so as to achieve a desirable tradeoff among resource efficiency (bandwidth and energy), system performance (estimation distortion and network lifetime), and implementation simplicity. More specifically, we address the efficient distributed estimation from the following perspectives: (i) rate-distortion perspective, where the objective is to study the rate-distortion bound for the distributed estimation and to design practical and distributed algorithms suitable for wireless sensor networks to approach the performance bound by optimally allocating the bit rate for each sensor, (ii) energy-distortion perspective, where the objective is to study the energy-distortion bound for the distributed estimation and to design practical and distributed algorithms suitable for wireless sensor networks to approach the performance bound by optimally allocating the bit rate and transmission energy for each sensor, and (iii) lifetime-distortion perspective, where the objective is to maximize the network lifetime while meeting estimation distortion requirements by jointly optimizing the source coding, source throughput and multi-hop routing. Also, energy-efficient cluster-based distributed estimation is studied, where the objective is to minimize the overall energy cost by appropriately dividing the sensor field into multiple clusters with data aggregation at cluster heads.

Στην παρούσα διπλωματική εργασία πραγματοποιείται μελέτη της επίδρασης ενός συστήματος διασύνδεσης ανεμογεννήτριας με το δίκτυο μέσης τάσης, όσον αφορά την έγχυση αρμονικών από την ανεμογεννήτρια προς το δίκτυο. Η μελέτη περιλαμβάνει τη μοντελοποίηση κάθε τμήματος ενός δικτύου μέσης τάσης με πραγματικά στοιχεία που πάρθηκαν από τη ΔΕΗ. Επιπλέον, μοντελοποιείται μια ανεμογεννήτρια τύπου σύγχρονης μόνιμων μαγνητών, καθώς και όλοι οι ηλεκτρονικοί μετατροπείς ισχύος που απαιτούνται για τη διασύνδεση αυτής με το δίκτυο μέσης τάσης. Όλα τα επιμέρους μοντέλα διασυνδέονται μεταξύ τους ώστε να αποτελέσουν το συνολικό μοντέλο που περιλαμβάνει το δίκτυο μέσης τάσης με τις γραμμές μεταφοράς, τους μετασχηματιστές, τα φορτία και τους πυκνωτές αντιστάθμισης που διαθέτει και την ανεμογεννήτρια μαζί με τους μετατροπείς συνδεδεμένη σε ένα ζυγό του δικτύου. Από το μοντέλο αυτό εξάγεται το αρμονικό περιεχόμενο τόσο στο ζυγό της ανεμογεννήτριας όσο και σε διπλανούς ζυγούς και συγκρίνεται κάθε φορά με το πρότυπο IEC 61000-3-6 που θέτει τα όρια για την έγχυση αρμονικών στο δίκτυο μέσης τάσης. Αρχικά, γίνεται μια αναφορά στα πλεονεκτήματα της αιολικής ενέργειας έναντι των άλλων συμβατικών μορφών ενέργειας. Επιπλέον, αναφέρονται τα προβλήματα που δημιουργούνται από την παρουσία ανώτερων αρμονικών στο δίκτυο. Έπειτα, πραγματοποιείται θεωρητική ανάλυση κάθε τμήματος του δικτύου μέσης τάσης. Για κάθε τμήμα δημιουργείται, επίσης, ένα μοντέλο στο Matlab/Simulink που βασίζεται σε στοιχεία από τη ΔΕΗ και εξισώσεις που αφορούν το εκάστοτε τμήμα. Στη συνέχεια, αναλύεται και μοντελοποιείται στο Matlab/Simulink και κάθε τμήμα του συστήματος διασύνδεσης της ανεμογεννήτριας με το δίκτυο μέσης τάσης που περιλαμβάνει την ανεμογεννήτρια τύπου σύγχρονης μόνιμων μαγνητών, ανορθωτική διάταξη, μετατροπέα συνεχούς τάσης σε συνεχή, τριφασικό αντιστροφέα, φίλτρο και μετασχηματιστή. Για κάθε μετατροπέα αναλύεται και μοντελοποιείται και έλεγχος κλειστού βρόχου. Τέλος, τα επιμέρους μοντέλα συγκροτούνται σε ένα συνολικό μοντέλο. Μεταβάλλοντας το συντελεστή φόρτισης των μετασχηματιστών του δικτύου, το συντελεστή ισχύος και το πλήθος των ανεμογεννητριών που συνδέονται στον ίδιο ζυγό, εξάγεται το αρμονικό περιεχόμενο για διάφορους ζυγούς και ελέγχεται αν αυτό συμφωνεί με το πρότυπο IEC 61000-3-6. Επιπλέον, εξετάζεται το αρμονικό περιεχόμενο της τάσης, όταν ο ζυγός διασύνδεσης αλλάζει θέση μέσα στο υπό μελέτη δίκτυο, καθώς και όταν αλλάζει η ταχύτητα του ανέμου. / In the present diploma thesis a study of the effect of a wind turbine system interconnection with medium voltage grid is conducted, in terms of harmonic injection of the turbine to the grid. The study includes the modeling of each part of a medium voltage grid with actual data taken from the Public Power Corporation of Greece. Furthermore, a synchronous wind turbine with permanent magnets is modeled along with all the electronic power converters that are required for the interconnection with the medium voltage grid. All the individual models are connected together to form the overall model, including the medium voltage transmission lines, transformers, loads and their compensation capacitors, along with the wind turbine and the power converters. The wind turbine is connected in a bus of the medium voltage grid. Then, the harmonic content in both the bus of the wind turbine and in other buses of the grid is extracted and compared with the standard IEC 61000-3-6 that sets the limits for harmonic injection in medium voltage grid. Initially, the advantages of use of wind energy are mentioned compared to other conventional forms of energy. Furthermore, the problems caused by the presence of harmonics in the grid are discussed. A theoretical analysis of each section of the medium voltage grid is conducted. For each section, a model in Matlab / Simulink is also created, based on data from the Public Power Corporation of Greece and on equations related to each section. Then, every part of the system connecting the wind turbine to the medium voltage grid is analyzed and modeled in Matlab/Simulink. This system consists of a synchronous wind turbine with permanent magnets, DC-to-DC converter, three-phase inverter, filter and transformer. A closed loop control for each converter is also analyzed and modeled. Finally, the aforementioned models are built up into one single model. By varying the load factor of the grid transformers, the power factor and the number of wind turbines connected to the same bus, the harmonic content of various buses is extracted and checked if it complies with the standard IEC 61000-3-6. Moreover, the harmonic content of the voltage is calculated, when the system of wind turbine changes position in the present medium voltage grid, as well as when the wind speed is changed.

In this thesis we discuss possible ways to give quantitative measurement for two spaces not being quasi-isometric. From this quantitative point of view, we reconsider the definition of quasi-isometries and propose a notion of ''quasi-isometric distortion growth'' between two metric spaces. We revise our article [32] where an optimal upper-bound for Morse Lemma is given, together with the dual variant which we call Anti-Morse Lemma, and their applications.Next, we focus on lower bounds on quasi-isometric distortion growth for hyperbolic metric spaces. In this class, $L^p$-cohomology spaces provides useful quasi-isometry invariants and Poincaré constants of balls are their quantitative incarnation. We study how Poincaré constants are transported by quasi-isometries. For this, we introduce the notion of a cross-kernel. We calculate Poincaré constants for locally homogeneous metrics of the form $dt^2+\sum_ie^{2\mu_it}dx_i^2$, and give a lower bound on quasi-isometric distortion growth among such spaces.This allows us to give examples of different quasi-isometric distortion growths, including a sublinear one (logarithmic).

Registruojant (fotografuojant) objektus, dėl netinkamos vaizdo kameros padėties ar nepakankamos judėjimo erdvės gaunami vaizdo iškraipymai. Tokie vaizdų iškraipymai vadinami sukimosi ir perspektyvos iškraipymais. Šio darbo tikslas – ištirti kaip tiksliai, panaudojant vaizdo kameros posvyrio kampų reikšmes ir atstumą iki registruojamo objekto, galima atstatyti iškreiptą vaizdą. Vaizdo kameros posvyrio kampų (krypavimo ir nuolydžio) matavimui, pritaikomas trijų ašių analoginis akcelerometras, o atstumui matuoti – ultragarsinis atstumo jutiklis. Pasuktų vaizdų atstatymui naudojami sukonstruoto vaizdo kameros posvyrio kampų matuoklio duomenys. Kaip alternatyvi priemonė, vaizdo pasukimo kampui nustatyti, naudojama „radon“ transformacija. Tyrimai parodė, kad vaizdų atstatymui geriau naudoti vaizdo kameros posvyrio kampų matuokliu išmatuotus krypavimo kampus, nes „radon“ transformacija negalima nustatyti sudėtingų vaizdų pasukimo kampo. Vaizdo perspektyvos iškraipymų atstatymui, pagal “pinhole” (mažos skylutės) kameros modelį, kuris dažniausiai naudojamas robotų regoje, buvo sudarytas matematinis modelis. Šis modelis, lyginant su vaizdų atstatymo pagal keturis objekto kampus metodu, tinka įvairesnių formų iškreiptiems vaizdams atstatyti. Vaizdų atstatymui pasitelktas MathWorks, Image Processing paketas. Atstatytų vaizdų palyginimui su originaliu, skaičuojamas dvimatis koreliacijos koeficientas tarp dviejų to paties dydžio vaizdo matricų. / Image distortions may appear because of inappropriate position of the camera or because of an insufficient space movement while taking pictures of objects. These kinds of image distortions are called rotation and perspective distortions. The aim of the work is to investigate how the camera tilt meanings and distance to the object could help to restore the image distortions. To measure camera tilt angles (roll and pitch angles) the three axis analogical accelerometer is used, and to measure the distance - ultrasonic range finder. The measurements of designed camera tilt angles device data is used in order to restore rotated images. As an alternative method for setting roll angle, the “radon” transformation is used. The researches have showed that it is more useful to use the measurements of camera roll angles, because “radon” transformation is not capable to identify rotation angles of elaborate images. To set image perspective distortion rectification according to “pinhole” camera model, which is used in robot vision, the mathematical model was made. This model compared with image restoration according to four object corners method is more suitable for various forms of image distortions. To restore image MathWorks, Image Processing batch was used. The comparison between original and restorted image was found computing the two-dimensional correlation coefficient between image matrices of the same size.

The study investigated the feasibility of including DPOAEs in the annual medical surveillance test battery for the identification of NIHL in a group of employees in a manufacturing industry in KwaZulu-Natal. Feasibility was investigated by exploring the sensitivity, specificity and predictive efficiency of DPOAEs, the ability of DPOAEs to detect subtle noise-induced cochlea changes, the test-retest reliability of DPOAEs and lastly, the duration of time taken to conduct the DPOAE test bilaterally. A cross-sectional and repeated measures within-in participant design was utilized in the study. A purposive convenience sampling technique was used, as well as a stratified sampling approach in order to realize objective two of the study. The study consisted of 60 participants, which were further stratified into four test groups, i.e. Group A: 0-3 years, Group B: 3.1-6 years and Group C: 6.1-9 years and Group D: 9.1-13 years of working within the beverage manufacturing industry. A high sensitivity and negative predictive value was reported in the current study, suggesting that DPOAEs may be able to identify those who present with subtle cochlea changes as a result of exposure to occupational noise. The sensitivity of DPOAEs was 100% at 1, 2, 4, 6 and 8kHz in the right ear and at 4 and 6kHz in the left ear. The specificity of DPOAEs in the current study ranged between 55%-97% across the frequency range in the right ear and 49%-88% in the left ear. A negative predictive value of 100% was obtained bilaterally across the frequency range, except at 8kHz in the left ear. Visual inspection of the DPgram in the current study revealed a bilateral reduction in DPOAE amplitudes for all test groups in the high frequency region of the DP-Gram, namely, 5477Hz and 7303Hz, in the absence of a statistically significant difference (p>0.05). A greater frequency range appears to be affected in this group of workers, indicating that the type of noise, namely, impulse noise, may result in cochlea changes. Corresponding changes on the pure tone audiogram were not observed, however, noise notch configurations were observed for the groups with a longer history of noise exposure. This was not seen bilaterally as is typically expected with NIHL. Good test-retest reliability across the frequency range obtained in the current study further indicates the feasibility of including DPOAEs in the annual medical surveillance test battery. Additionally, the current study calculated an average of 86 seconds (1 minute 26 seconds) to conduct the DPOAE test bilaterally, confirming that DPOAEs are a quick test to administer. The findings of this study suggest that DPOAEs may be used to monitor early subtle noise-induced cochlea changes for workers exposed to noise in the beverage manufacturing industry as part of the annual medical surveillance test battery. / Thesis (M.Comm.Path.)-University of KwaZulu-Natal, Westville, 2013.

Data are valuable assets to any organizations or individuals. Data are sources of useful information which is a big part of decision making. All sectors have potential to benefit from having information. Commerce, health, and research are some of the fields that have benefited from data. On the other hand, the availability of the data makes it easy for anyone to exploit the data, which in many cases are private confidential data. It is necessary to preserve the confidentiality of the data. We study two categories of privacy: Data Value Hiding and Data Pattern Hiding. Privacy is a huge concern but equally important is the concern of data utility. Data should avoid privacy breach yet be usable. Although these two objectives are contradictory and achieving both at the same time is challenging, having knowledge of the purpose and the manner in which it will be utilized helps. In this research, we focus on some particular situations for clustering and classification problems and strive to balance the utility and privacy of the data. In the first part of this dissertation, we propose Nonnegative Matrix Factorization (NMF) based techniques that accommodate constraints defined explicitly into the update rules. These constraints determine how the factorization takes place leading to the favorable results. These methods are designed to make alterations on the matrices such that user-specified cluster properties are introduced. These methods can be used to preserve data value as well as data pattern. As NMF and K-means are proven to be equivalent, NMF is an ideal choice for pattern hiding for clustering problems. In addition to the NMF based methods, we propose methods that take into account the data structures and the attribute properties for the classification problems. We separate the work into two different parts: linear classifiers and nonlinear classifiers. We propose two different solutions based on the classifiers. We study the effect of distortion on the utility of data. We propose three distortion measurement metrics which demonstrate better characteristics than the traditional metrics. The effectiveness of the measures is examined on different benchmark datasets. The result shows that the methods have the desirable properties such as invariance to translation, rotation, and scaling.

Nonrigid deformation modeling and estimation from images is a technically challenging task due to its nonlinear, nonconvex and high-dimensional nature. Traditional optimization procedures often rely on good initializations and give locally optimal solutions. On the other hand, learning-based methods that directly model the relationship between deformed images and their parameters either cannot handle complicated forms of mapping, or suffer from the Nyquist Limit and the curse of dimensionality due to high degrees of freedom in the deformation space. In particular, to achieve a worst-case guarantee of ∈ error for a deformation with d degrees of freedom, the sample complexity required is O(1/∈d). In this thesis, a generative model for deformation is established and analyzed using a unified theoretical framework. Based on the framework, three algorithms, Data-Driven Descent, Top-down and Bottom-up Hierarchical Models, are designed and constructed to solve the generative model. Under Lipschitz conditions that rule out unsolvable cases (e.g., deformation of a blank image), all algorithms achieve globally optimal solutions to the specific generative model. The sample complexity of these methods is substantially lower than that of learning-based approaches, which are agnostic to deformation modeling. To achieve global optimality guarantees with lower sample complexity, the structureembedded in the deformation model is exploited. In particular, Data-driven Descentrelates two deformed images that are far away in the parameter space by compositionalstructures of deformation and reduce the sample complexity to O(Cd log 1/∈).Top-down Hierarchical Model factorizes the local deformation into patches once theglobal deformation has been estimated approximately and further reduce the samplecomplexity to O(Cd/1+C2 log 1/∈). Finally, the Bottom-up Hierarchical Model buildsrepresentations that are invariant to local deformation. With the representations, theglobal deformation can be estimated independently of local deformation, reducingthe sample complexity to O((C/∈)d0) (d0 ≪ d). From the analysis, this thesis showsthe connections between approaches that are traditionally considered to be of verydifferent nature. New theoretical conjectures on approaches like Deep Learning, arealso provided. practice, broad applications of the proposed approaches have also been demonstrated to estimate water distortion, air turbulence, cloth deformation and human pose with state-of-the-art results. Some approaches even achieve near real-time performance. Finally, application-dependent physics-based models are built with good performance in document rectification and scene depth recovery in turbulent media.

Analytical chemistry is a field with a vast variety of applications. A robust companion in the field is liquid chromatography, the method used in this thesis, which is an established workhorse and a versatile tool in many different disciplines. It can be used for identification and quantification of interesting compounds generally present in low concentrations, called analytical scale chromatography. It can also be used for isolation and purification of high value compounds, called preparative chromatography. The latter is usually conducted in large scale with high concentrations. With high concentrations it is also possible to determine something called adsorption isotherms. Determination of adsorption isotherms is a useful tool for quite a wide variety of reasons. It can be used for characterisation of chromatographic separation systems, and then gives information on the retention mechanism as well as provides the possibility to study column-column and batch-batch reproducibility. If a protein is immobilised on a solid support, adsorption isotherms can be used for pharmacological characterisation of drug-protein interactions. Moreover, they can be used for the study of unexpected chromatographic phenomena. If the adsorption isotherm is known it is also possible to simulate chromatograms, and subsequently optimise the separation process numerically. The gain of a numerically optimised separation process is higher purity or yield of valuable compounds such as pharmaceuticals or antioxidants, as well as reducing the solvent usage. Taken all together, it saves time, money and the environment. However, the process of the adsorption isotherm determination requires a number of careful experimental considerations and preparations, and these are the main focus of the thesis. Important steps along the way include the choice of separation system and of suitable analytes, preparation of mobile phases and sample solutions, calibration, determination of injection profiles and column void, and of course the adsorption isotherm determination method itself. It is also important to keep track of parameters such as temperature and pH. These issues are discussed in this thesis. At the end, a description of useful methods for processing of the raw adsorption isotherm data is presented, as well as a brief passage on methods for numerical optimisation.

Liquid water formation and flooding in PEM fuel cell gas distribution channels can significantly degrade fuel cell performance by causing substantial pressure drop in the channels and by inhibiting the transport of reactants to the reaction sites at the catalyst layer. A better understanding of the mechanisms of discrete water droplet transport by air flow in such small channels may be developed through the application of quantitative flow visualization techniques. This improved knowledge could contribute to improved gas channel design and higher fuel cell efficiencies. An experimental investigation was undertaken to gain better understanding of the relationships between air velocity in the channel, secondary rotational flows inside a droplet, droplet deformation, and threshold shear, drag, and pressure forces required for droplet removal. Micro-digital-particle-image-velocimetry (micro-DPIV) techniques were used to provide quantitative visualizations of the flow inside the liquid phase for the case of air flow around a droplet adhered to the wall of a 1 mm x 3 mm rectangular gas channel model. The sidewall against which the droplet was adhered was composed of PTFE treated carbon paper to simulate the porous GDL surface of a fuel cell gas channel. Visualization of droplet shape, internal flow patterns and Velocity measurements at the central cross-sectional plane of symmetry in the droplet were obtained for different air flow rates. A variety of rotational secondary flow patterns within the droplet were observed. The nature of these flows depended primarily on the air flow rate. The peak velocities of these secondary flow fields were observed to be around two orders of magnitude below the calculated channel-averaged driving air velocities. The resulting flow fields show in particular that the velocity at the air-droplet interface is finite. The experimental data collected from this study may be used for validation of numerical simulations of such droplet flows. Further study of such flow scenarios using the techniques developed in this experiment, including the general optical distortion correction algorithm developed as part of this work, may provide insight into an improved force balance model for a droplet exposed to an air flow in a gas channel.