Electrocardiogram (ECG) signal processing has been the subject of intense research in the past years, due to its strategic place in the detection of several cardiac pathologies. However, ECG signal is frequently corrupted with different types of noises such as 60Hz power line interference, baseline drift, electrode movement and motion artifact, etc. In this thesis, a hybrid two-stage model based on the combination of wavelet decomposition and artificial neural network is proposed for ECG noise reduction based on excellent localization features: wavelet transform and the adaptive learning ability of neural network. Results from the simulations validate the effectiveness of this proposed method. Simulation results on actual ECG signals from MIT-BIH arrhythmia database [30] show this approach yields improvement over the un-filtered signal in terms of signal-to-noise ratio (SNR).

We propose a method of adaptive image denoising using a support vector machine (SVM) classifier to select between multiple well-performing contemporary denoising algorithms for each pixel of a noisy image. We begin by proposing a simple method for realistically generating noisy images, and also describe a number of novel and pre-existing features based on seam energy, local colour, and saliency which are used as classifier inputs. Our SVM strategic image denoising (SVMSID) results demonstrate better image quality than either candidate denoising algorithm for images of moderate noise level, as measured using the perceptually-based quaternion structural similarity image metric (QSSIM). We also demonstrate a modified training point selection method to improve robustness across many noise levels, and propose various extensions to SVMSID for further exploration. / Thesis / Master of Applied Science (MASc)

No description available.

No description available.

No description available.

As technology moves forward it has a tendency to consume more and more power that needs to be cooled by bigger and louder fans, this is especially true for the new generation of 5G radio equipment. This Master thesis is a collaboration with Ericsson and attempts to construct a shroud for containing a number of 5G radio units whilst attenuating the fan noise of the units as effectively as possible. In this project are air ducts used and at the ends silencers are created utilizing the Cremer impedance; the optimal wall impedance for damping an acoustic mode of a propagating wave. To predict the result, a simplified model in an acoustic FEM program was also explored and compared to the sound level of the constructed shroud. The finished shroud successfully reduces the noise of the radio units by 13 dB(A) while causing an increase in temperature of between 2.8°C to 5.9°C. This result was deemed to be a success and the Cremer impedance approach of reducing noise is therefore advised for future development. / Allt eftersom tekniken går framåt tenderar den att också förbruka mer och mer energi som i sin tur måste kylas av kraftigare och mer högljudda fläktar, detta fenomen är särskilt påtagligt när det kommer till den senaste generationens radioutrustning för 5G. Detta examensarbete är ett samarbete mellan KTH och Ericsson med avsikt att skapa en kåpa som är designad för att innesluta ett bestämt antal 5G radiomoduler. Denna kåpa ska i så stor utsträckning som möjligt dämpa det fläktinducerade bullret. I det här projektet nyttjas kanaler med ljuddämpare vid ändarna som dämpar ljudet med hjälp av Cremerimpedans, dvs: den väggimpedans som optimalt dämpar en akustisk mod. För att kunna förutspå resultatet skapades en förenklad akustisk modell i ett FEM program. Resultatet från denna modell jämförs sedan med ljudeffektnivån från slutmätningen av den färdiga kåpan. Resultatet från slutmätningen visar att kåpan lyckas sänka radioenheternas totala ljudeffektnivå med 13 dB(A) samtidigt som en temperaturökning på mellan 2.8°C och 5.9°C erhålls. Det här resultatet bedöms vara en framgång, vilket leder till slutsatsen att ljuddämpning med användning av Cremerimpedans rekommenderas för vidare arbete.

More than one billion vehicle tires reach the end of their useful service life annually. Less than a quarter of rubber waste is reused or recycled in some way. Interest has grown in working to discover means by which to incorporate rubber tire waste into construction materials. This study sought to delve into the use of micronized rubber powder (MRP) as an acoustic agent within particleboard and concrete. In addition, work was conducted to characterize the effect that MRP has on the strength and flexural properties of concrete. Furthermore, research sought to provide insight into how pine biomass, a forest products industry waste, would interact with MRP in concrete as it relates to strength and acoustic properties. As expected, particleboard that contained MRP resulted in lower strength but higher flexibility. Acoustic testing revealed that there was minimal sound absorption improvement at some frequencies and less absorption at low and high frequencies. Sound transmission loss was slightly improved by the addition of MRP to the particleboard. Adding pine biomass and MRP to concrete yielded much lower compressive strength as compared to plain concrete. Visual inspection of the sound absorption coefficient curves over the full range of test frequencies identified limited, if any, advantage for the addition of MRP or biomass. Some ranges of frequencies offered minimal improvement. There appeared to be no appreciable sound absorption advantage to adding MRP, pine biomass, or the combination of the two into concrete mix proportions. Modulus of rigidity was decreased as compared to plain concrete when MRP, pine biomass, or a combination of both were incorporated into the concrete mixture as volume replacement for aggregate. Visual observation revealed that flexural failure for the MRP or pine beams were less sudden and less catastrophic than the plain concrete samples. Modulus of elasticity was decreased as compared to plain concrete when MRP, pine biomass, or a combination of both were incorporated into the concrete mixture as volume replacement for aggregate. The more flexible and ductile concrete produced with MRP and biomass provides a combination of properties that serve to lessen the propagation of cracks throughout the specimen.

Raman spectroscopy is an effective optical analysis of the biochemically specific characterization of tissues without contrast agents or exogenous dyes. Applications of Raman spectroscopy include analysis and biomarker investigation, disease diagnosis and surgical guidance. One major challenge in Raman spectroscopy is removing inherent fluorescence background present in samples to acquire Raman signatures. In some tissues, like liver, kidney and darkly pigment skin, the auto-fluorescence background is strong enough to overwhelm the Raman peaks in conventional Near-Infrared (NIR) Raman systems. Recent publications have shown that using Raman systems with excitation sources with wavelengths beyond 830 nm and short-wave infrared (SWIR) InGaAs Array detectors resulted in dramatically reduced auto-fluorescence. The unique characteristics of Raman signals collected from SWIR systems versus NIR Raman systems requires inspection of the suitability of spectral pre-processing techniques. This thesis focused on the development of spectral processing techniques at three different steps; 1) detector background & noise reduction; 2) Auto-fluorescence background subtraction; 3) detection of outlier measurements to assist statistical classification. Detector background and noise reduction was compared between two different techniques, and a direct subtraction method resulted in better performance to reduce fixed pattern noise unique to InGaAs arrays. For the aim 2, three different algorithms for fluorescence background removal were developed, and a modified polynomial fitting method was found to be most appropriate for the low signal-to-noise (SNR) spectra. Finally, local outlier factor(LOF), a multivariate statistical outlier metric, was implemented in a two-stage fashion, and shown to be effective at identifying raw measurement errors and Raman spectra outliers. The overall outcome of this thesis was the evaluation of spectral processing techniques for SWIR Raman spectroscopy systems, and the development of specific techniques to optimize data quality and best prepare spectra for statistical analysis. / Bioengineering

An innovative application of existing technology is proposed for attenuating the effects of transient phenomena, such as rotor-stator and rotor-strut interactions, linked to noise and fatigue failure in turbomachinery environments. A computational study was designed to assess the potential of Passive Porosity Technology as a mechanism for alleviating interaction effects and radiated noise by reducing the fluctuating forces acting on the vane surfaces. The study involved a typical high bypass fan stator airfoil immersed in a subsonic free field and exposed to the effects of a transversely moving wake. Time histories of the primitive aerodynamic variables obtained from Computational Fluid Dynamics (CFD) calculations were input into an acoustic prediction code to estimate noise levels at a radial distance of ten chords from the stator airfoil. This procedure was performed on the solid airfoil to obtain a baseline, and on approximately fifty porous configurations in order to isolate those that would yield maximum noise reductions without compromising the aerodynamic performance of the stator. It was found that, for a single stator immersed in a subsonic flow field, communication between regions of high pressure differential - made possible by the use of passive porosity - tends to induce a time-dependent oscillatory pattern of small inflow-outflow regions near the stator leading edge (LE), which is well established before wake effects come into play. The oscillatory pattern starts at the LE, and travels downstream on both suction and pressure sides of the airfoil. The amplitude of the oscillations seemed to be proportional to the extension of the porous patch on the pressure side. Regardless of this effect, which may not have occurred if the airfoil were placed within a stator cascade, communication between regions of high pressure differential is necessary to significantly alter the noise radiation pattern of the stator airfoil. Whether those changes result in noise abatement or enhancement depends primarily on the placement and extension of the porous patches. For most viable configurations, porosity reduced loading noise but increased thickness noise. Variations in nominal porosity were of secondary importance. In general, the best aerodynamic performers (i.e., those configurations that were able to reduce unsteady lift without severely altering the lift and/or drag characteristics of the solid airfoil) were also the best acoustic performers. As a result of using passive surface porosity, overall peak radiated noise was reduced by approximately 1.0 dB. This reduction increased to about 2.5 dB when the effects of loading noise alone were considered. / Ph. D.

Currently in the United States, Active Noise Reduction (ANR) headphones cannot be tested and labeled as hearing protection devices (HPDs) due to inherent limitations with the existing psychophysical headphone testing standard, real-ear attenuation at threshold (REAT). This research focused on the use of a standard, for physical, microphone-in-real-ear testing, (MIRE, ANSI S12.42-1995), to determine if MIRE may be appropriately used to measure the total attenuation (i.e., passive + active) of ANR headphones. The REAT " Method B, Subject-Fit protocol," ANSI S12.6-1997(R2002), was also used to assess passive attenuation (and used for comparison with the MIRE data), as this is the current standard for passive Headphone attenuation testing. The MIRE protocol currently does not specify a standardized location for measurement microphone placement. Prior research is mixed as to the potential benefits and shortcomings of placing the measurement microphone outside versus inside the ear canal. This study captured and compared acoustic spectral data at three different microphone locations: in concha, in ear canal-shallow depth, and in ear canal-deep depth (with a probe tube microphone positioned near the tympanic membrane), using human test participants and five ANR headphones of differing design. Results indicate that the MIRE protocol may be used to supplant the REAT protocol for the measurement of passive attenuation, although differences were observed at the lowest-tested frequency of 125 Hz. Microphone placement analysis revealed no significant difference among the three locations specified, with a noted caveat for the probe tube microphone location at the highest tested frequency of 8000 Hz. Overall findings may be useful to standards-making committees for evaluating a viable solution and standardized method for testing and labeling ANR headphones for use as hearing protection devices. Microphone placement results may assist the practitioner in determining where to place measurement microphones to best suit their particular needs when using MIRE. Discussion includes an in-depth interpretation of the data, comparisons within and between each protocol, and recommendations for further avenues to explore based on the data presented. / Ph. D.