Die Arbeit stellt neu entwickelte Röntgenbeugungsmethoden vor, mit deren Hilfe der Verzerrungszustand des Kristallgitters von Halbleiter-Wafern und -Bauteilen im Detail charakterisiert werden kann. Hierzu werden die aussergewöhnlichen Eigenschaften der an modernen Synchrotrons wie der ESRF (Grenoble) verfügbaren Röntgenstrahlung genutzt. <br>Im ersten Teil der Arbeit werden Röntgen-Diffraktometrie und -Topographie zu einer Untersuchungsmethode kombiniert, mit der die makroskopische Krümmung von Halbleiter-Wafern ebenso wie ihre mikroskopische Defektstruktur abgebildet werden kann. Der zweite Teil ist der Untersuchung von epitaktisch gewachsenen und geätzten Oberflächengittern mit Abmessungen im Submikrometer-Bereich gewidmet. Die unterschiedlichen Gitterkonstanten der beteiligten Halbleitermaterialien führen zu einem inhomogenen Verzerrungsfeld in der Probe, das sich im Röntgenbild durch eine charakteristische Verformung des Beugungsmusters in der Umgebung der Bragg-Reflexe äussert. Die Analyse der experimentell gemessenen Beugungsmuster geschieht mit Hilfe eines neu entwickelten Simulationsverfahrens, das Elastizitätstheorie und eine semi-kinematische Röntgenbeugungstheorie miteinander verbindet. Durch quantitativen Vergleich der Simulationsergebnisse mit den Messdaten kann auf den genauen Verlauf des Verzerrungsfeldes in den Proben zurückgeschlossen werden. Dieses Verfahren wird erfolgreich auf verschiedene Halbleiter-Probensysteme angewendet, und schliesslich auch auf die Untersuchung von akustischen Oberflächenwellen in Halbleiterkristallen übertragen. / This thesis presents newly developed X-ray methods which can be used to characterize in detail the state of distortion of the crystal lattice in semiconductor wafers, devices and nanostructures. The methods use the extraordinary properties of the X-rays available from modern synchrotron sources such as the ESRF (Grenoble). <br>In the first part of the thesis, X-ray diffractometry and X-ray topography are combined into a new method, called X-ray rocking curve imaging, which allows to image the macroscopic curvature of semiconductor wafers as well as the underlying microscopic defect structure. The second part of the thesis deals with the investigation of epitaxially grown and subsequently etched semiconductor gratings with lateral periods below the micrometer. The lattice mismatch between the different materials used in heteroepitaxy leads to a non-uniform strain field in the sample, which is reflected in a characteristic distortion of the X-ray diffraction pattern around each Bragg peak. The experimental data are evaluated with the help of a newly developed simulation procedure which combines elasticity theory with a semi-kinematical theory of X-ray diffraction. From a quantitative comparison of measured and simulated data the detailed shape of the strain field in the samples can be deduced. This procedure is used successfully for the structural characterization of different types of semiconductor gratings, and is finally applied also to the investigation of surface acoustic waves in crystals.

Noise exposure is one of the most prevalent causes of irreversible occupational disease in Sweden and in many other countries. In hearing conservation programs, aimed at preventing noise-induced hearing loss, audiometry is an important instrument to highlight the risks and to assess the effectiveness of the program. A hazardous working environment and persons affected by it can be identified by monitoring the hearing thresholds of individual employees or groups of employees over time. However, in order to evaluate the prevalence of occupational noise-induced hearing loss, relevant reference data of unexposed subjects is needed. The first part of this dissertation concerns the changes in hearing thresholds over three decades in two occupational environments with high noise levels in the province of Östergötland, Sweden: the mechanical and the wood processing industries. The results show a positive trend, with improving median hearing thresholds from the 1970s into the 1990s. However, the hearing loss present also in the best period, during the 1990s, was probably greater than if the occupational noise exposure had not occurred. This study made clear the need for a valid reference data base, representing the statistical distribution of hearing threshold levels in a population not exposed to occupational noise but otherwise comparable to the group under study. In the second part of the dissertation, reference data for hearing threshold levels in women and men aged from 20 to 79 years are presented, based on measurements of 603 randomly selected individuals in Östergötland. A mathematical model is introduced, based on the hyperbolic tangent function, describing the hearing threshold levels as functions of age. The results show an age-related gender difference, with poorer hearing for men in age groups above 50 years. The prevalence of different degree of hearing loss and tinnitus is described for the same population in the third part of the dissertation. The overall prevalence of mild, moderate, severe or profound hearing loss was 20.9% collectively for women and 25.0% collectively for men. Tinnitus was reported by 8.9% of the women and 17.6% of the men. Approximately 2.4% of the subjects under study had been provided with hearing aids. However, about 7.7% were estimated to potentially benefit from hearing aids as estimated from their degree of hearing loss. Noise-induced hearing loss primarily causes damage to the outer hair cells of the inner ear. The fourth and last part of the dissertation evaluates the outer hair cell function, using otoacoustic emission measurements (OAE). Prevalence results from three different measuring techniques are presented: spontaneous otoacoustic emissions (SOAE), transient evoked otoacoustic emissions (TEOAE) and distortion product otoacoustic emissions (DPOAE). Gender and age effects on the recorded emission levels were also investigated. Women showed higher emission levels compared to men and for both women and men the emission levels decreased with increasing age. The results from the OAE recordings were shown to be somewhat affected by the state of the middle ear. The study included tympanometry, and the relation of the outcome ofthis test to the otoacoustic emissions is described, where high middle ear compliance resulted in low emission level. Reference data for the tympanometric measurements are also presented. The results of this project form an essential part of the important work against noiseinduced hearing loss, which needs continuous monitoring. The reference data presented here will provide a valid and reliable data base for the future assessment of hearing tests performed by occupational health centres in Sweden. This data base will in turn prove useful for comparison studies for Sweden as a responsible fellow EU member country setting high standards for work force safety. The statistical distribution of hearing threshold levels as a function of age for men and women in tabulated form is available on the Swedish Work Environment Authority (Arbetsmiljöverket) web site: http://www.av.se/publikationer/bocker/fysiskt/h293.shtm.

This doctoral thesis is the result of a research effort performed in two fields of speech technology, i.e., speech recognition and mispronunciation detection. Although the two areas are clearly distinguishable, the proposed approaches share a common hypothesis based on psychoacoustic processing of speech signals. The conjecture implies that the human auditory periphery provides a relatively good separation of different sound classes. Hence, it is possible to use recent findings from psychoacoustic perception together with mathematical and computational tools to model the auditory sensitivities to small speech signal changes. The performance of an automatic speech recognition system strongly depends on the representation used for the front-end. If the extracted features do not include all relevant information, the performance of the classification stage is inherently suboptimal. The work described in Papers A, B and C is motivated by the fact that humans perform better at speech recognition than machines, particularly for noisy environments. The goal is to make use of knowledge of human perception in the selection and optimization of speech features for speech recognition. These papers show that maximizing the similarity of the Euclidean geometry of the features to the geometry of the perceptual domain is a powerful tool to select or optimize features. Experiments with a practical speech recognizer confirm the validity of the principle. It is also shown an approach to improve mel frequency cepstrum coefficients (MFCCs) through offline optimization. The method has three advantages: i) it is computationally inexpensive, ii) it does not use the auditory model directly, thus avoiding its computational cost, and iii) importantly, it provides better recognition performance than traditional MFCCs for both clean and noisy conditions. The second task concerns automatic pronunciation error detection. The research, described in Papers D, E and F, is motivated by the observation that almost all native speakers perceive, relatively easily, the acoustic characteristics of their own language when it is produced by speakers of the language. Small variations within a phoneme category, sometimes different for various phonemes, do not change significantly the perception of the language’s own sounds. Several methods are introduced based on similarity measures of the Euclidean space spanned by the acoustic representations of the speech signal and the Euclidean space spanned by an auditory model output, to identify the problematic phonemes for a given speaker. The methods are tested for groups of speakers from different languages and evaluated according to a theoretical linguistic study showing that they can capture many of the problematic phonemes that speakers from each language mispronounce. Finally, a listening test on the same dataset verifies the validity of these methods. / <p>QC 20120914</p> / European Union FP6-034362 research project ACORNS / Computer-Animated language Teachers (CALATea)

In this thesis we study a scalar uniform and non-uniform quantization of random processes (or signals) in average case setting. Quantization (or discretization) of a signal is a standard task in all nalog/digital devices (e.g., digital recorders, remote sensors etc.). We evaluate the necessary memory capacity (or quantization rate) needed for quantized process realizations by exploiting the correlation structure of the model random process. The thesis consists of an introductory survey of the subject and related theory followed by four included papers (A-D). In Paper A we develop a quantization coding method when quantization levels crossings by a process realization are used for its coding. Asymptotical behavior of mean quantization rate is investigated in terms of the correlation structure of the original process. For uniform and non-uniform quantization, we assume that the quantization cellwidth tends to zero and the number of quantization levels tends to infinity, respectively. In Papers B and C we focus on an additive noise model for a quantized random process. Stochastic structures of asymptotic quantization errors are derived for some bounded and unbounded non-uniform quantizers when the number of quantization levels tends to infinity. The obtained results can be applied, for instance, to some optimization design problems for quantization levels. Random signals are quantized at sampling points with further compression. In Paper D the concern is statistical inference for run-length encoding (RLE) method, one of the compression techniques, applied to quantized stationary Gaussian sequences. This compression method is widely used, for instance, in digital signal and image processing. First, we deal with mean RLE quantization rates for various probabilistic models. For a time series with unknown stochastic structure, we investigate asymptotic properties (e.g., asymptotic normality) of two estimates for the mean RLE quantization rate based on an observed sample when the sample size tends to infinity. These results can be used in communication theory, signal processing, coding, and compression applications. Some examples and numerical experiments demonstrating applications of the obtained results for synthetic and real data are presented.

Intratympanic injection of various otoprotectants through the round window membrane (RWM) might become available in the near future as an alternative to the currently available medical and surgical methods used to treat several inner ear diseases. The most common outcome of such diseases is sensorineural hearing loss (SNHL). Two examples of  these otoprotectants are Edaravone and Brain-Derived Neurotrophic Factor (BDNF), both of which have already proved effective against  noise-induced hair cell loss, barotrauma  and ototoxicity caused by cisplatin. In four different studies we used two electrophysiological methods, auditory brainstem response (ABR) and distortion product otoacoustic emission (DPOAE), to study the effects of tobramycin and Pseudomonas aeruginosa exotoxin A (PaExoA) on the inner ears of 129 male Sprague-Dawley rats. In two investigations, not only the otoprotective effects of Edaravone on tobramycin-induced ABR threshold shifts and PaExoA-induced DPOAE  threshold changes, were studied but even different application times, in order to establish in which interval it was still possible to achieve effective otoprotection.We found that Edaravone gave otoprotection from tobramycin when injected simultaneously or within 7 days, but it had only a limited effect on the changes in DPOAE thresholds caused by PaExoA when injected 1, 2, or 4 hours after the exotoxin. The effect of BDNF on PaExoA-induced ABR threshold shifts was investigated in two studies, where different doses of intratympanically injected PaExoA were used and where BDNF was applied simultaneously, 12 or 72 hours efter exotoxin instillation. We found that BDNF had an otoprotective effect on SNHL induced by different doses PaExoA when injected simultaneously or with no more than 12 hours delay.

Cette étude présente des contributions en traitement statistique du signal avec des applications biomédicales. La thèse est divisée en deux parties. La première partie traite de la détection des hotspots à l'interface des protéines. Les hotspots sont les résidus dont les contributions énergétiques sont les plus importantes dans l'interaction entre protéines. Les forêts aléatoires (Random Forests) sont utilisées pour la classification. Une nouvelle famille de descripteurs de hotspot est également introduite. Ces descripteurs sont basés seulement sur la séquence primaire unidimensionnelle d'acides aminés constituant la protéine. Aucune information sur la structure tridimensionnelle de la protéine ou le complexe n'est nécessaire. Ces descripteurs, capitalisant les caractéristiques fréquentielle des protéines, nous permettent de savoir la façon dont la séquence primaire d'une protéine peut déterminer sa structure tridimensionnelle et sa fonction. Dans la deuxième partie, le RDT (Random Distortion Testing), un test robuste d'hypothèse, est considéré. Son application en détection du signal a montré que le RDT peut résister aux imperfections du modèle d'observation. Nous avons également proposé une extension séquentielle du RDT. Cette extension s'appelle le RDT Séquentiel. Trois problèmes classiques de détection d¿écart/distorsion du signal sont reformulés et résolus dans le cadre du RDT. En utilisant le RDT et le RDT Séquentiel, nous étudions la détection d'AutoPEEP (auto-Positive End Expiratory Pressure), une anomalie fréquente en ventilation mécanique. C'est la première étude de ce type dans la littérature. L'extension à la détection d'autres types d'asynchronie est également étudiée et discutée. Ces détecteurs d'AutoPEEP et d'asynchronies sont les éléments principaux de la plateforme de suivi de manière automatique et continue l'interface patient-ventilateur en ventilation mécanique.

Abstract In this thesis, a measurement system was set to perform comprehensive measurements on RF power amplifiers. Data obtained from the measurements is then processed mathematically to obtain three dimensional graphs of the basic parameters affected or generated by nonlinearities of the amplifier i.e. gain, efficiency and distortion. Using a class AB amplifier as the DUT, two sets of signals – both swept in power level and frequency - were generated to validate the method, a two-tone signal and a WCDMA signal. The three dimensional plot gives a thorough representation of the behavior of the amplifier in any arbitrary range of spectrum and input level. Sweet spots are consequently easy to detect and analyze. The measurement setup can also yield other three dimensional plots of variations of gain, efficiency or distortion versus frequencies and input levels. Moreover, the measurement tool can be used to plot traditional two dimensional plots such as, input versus gain, frequency versus efficiency etc, making the setup a practical tool for RF amplifiers designers. The test signals were generated by computer then sent to a vector signal generator that generates the actual signals fed to the amplifier. The output of the amplifier is fed to a vector signal analyzer then collected by computer to be handled. MATLAB® was used throughout the entire process. The distortion considered in the case of the two-tone signals is the third order intermodulation distortion (IM3) whereas Adjacent Channel Power Ratio (ACPR) was considered in the case of WCDMA.

In this thesis, a parsimoniously parameterized digital predistorter is derived for linearization of the IQ modulation mismatch and the amplifier imperfection in the signal generator [1]. It is shown that the resulting predistorter is linear in its parameters, and thus they may be estimated by the method of least-squares. Spectrally pure signals are an indispensable requirement when the signal generator is to be used as part of a test bed. Due to the non-linear characteristic of the IQ modulator and power amplifier, distortion will be present at the output of the signal generator. The device under test was the IQ modulation mismatch and power amplifier deficiencies in the signal generator. In [2], the dynamic range of low-cost signal generators are improved by employing model based digital pre-distortion and the designed predistorter seems to give some improvement of the dynamic range of the signal generator. The goal of this project is to implement and verify the theory parts [1] using data program (Matlab) to improve the dynamic range of the signal generator. The design digital pre-distortion that is implemented in software so that the dynamic range of the signal generator output after predistortion is superior to that of the output prior to it. In this project, we have observed numerical problems in the proposed theory and we have found other methods to solve the problem. The polynomial model is commonly used in power amplifier modeling and predistorter design. However, the conventional polynomial model exhibits numerical instabilities when higher order terms are included, we have used the conventional and orthogonal polynomial models. The result shows that the orthogonal polynomial model generally yield better power amplifier modeling accuracy as well as predistortion linearization performance then the conventional polynomial model.

Quality of the castings is affected by several factors which the designer should take into consideration during the product development process. Although residual stress is one of those, it is often not considered in practical computations. Hence residual stresses are one of the forgotten areas in designing of machine parts. This master thesis is focused on the investigation of residual stresses in a high pressure die casted component, with the aim of extending its service life, by taking results from the study as a feedback. The investigation of residual stresses was done on a variety of specimens, cast aluminum flywheel, provided by Husqvarna AB. This flywheel is a component in a product of the same company.In evaluating the residual stresses in the part, two tools-simulation and physical measurement were used. Moreover, comparison with these two methods is also done at an area of interest on the flywheel. The simulation was carried out by using MAGMAhpdc-a module for high pressure die casting process, from the commercial software package MAGMAsoft; while for the physical measurements, the hole drilling method was used, a method believed to be less accurate at low stresses areas. The findings obtained from this study show that the results from both procedures are close, with small deviations observed, which reveals the reliability of the hole drilling method even when the stress levels are low. It is also found that the compressive residual stresses dominate in the component-a preferred phenomenon with regards to residual stress.

H.264, also known as MPEG-4 Part 10: Advanced Video Coding, is the latest MPEG standard for video coding. It provides approximately 50% bit rate savings for equivalent perceptual quality compared to any previous standard. In the same fashion as previous MPEG standards, only the bitstream syntax and the decoder are specified. Hence, coding performance is not only determined by the standard itself but also by the implementation of the encoder. In this report we propose two methods for improving the coding performance while remaining fully compliant to the standard. After transformation and quantization, the transform coefficients are usually entropy coded and embedded in the bitstream. However, some of them might be beneficial to discard if the number of saved bits are sufficiently large. This is usually referred to as coefficient thresholding and is investigated in the scope of H.264 in this report. Lagrangian optimization for video compression has proven to yield substantial improvements in perceived quality and the H.264 Reference Software has been designed around this concept. When performing Lagrangian optimization, lambda is a crucial parameter that determines the tradeoff between rate and distortion. We propose a new method to select lambda and the quantization parameter for non-reference frames in H.264. The two methods are shown to achieve significant improvements. When combined, they reduce the bitrate around 12%, while preserving the video quality in terms of average PSNR. To aid development of H.264, a software tool has been created to visualize the coding process and present statistics. This tool is capable of displaying information such as bit distribution, motion vectors, predicted pictures and motion compensated block sizes.