Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2006. / A real time attitude and heading reference system (AHRS) was successfully implemented for use on an airship. The AHRS was tested on board a small airship (blimp) with real data supplied from the inertial measurement unit and GPS receiver. The inertial measurement unit was built with lower grade sensors, resulting in significant reductions in component cost. To ensure accurate navigation results, the high rate inertial measurements were complemented with low rate GPS velocity and position updates in an extended Kalman filter configuration. A study was made of various Kalman filter configurations, especially the possibility of splitting a big Kalman filter into smaller Kalman filters. It was found that the best trade-off between accuracy and processing power was achieved by having two smaller Kalman filters running in sequence. The first extended Kalman filter estimates the attitude of the airship, while the second extended Kalman filter estimates the velocity and position of the airship. The two smaller Kalman filters were implemented on an onboard computer to provide real time estimates of the attitude, velocity and position of the airship.

Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2006. / Digital audio data storage mediums have long been used within the consumer market. Today, because of the advancement of processor clock speeds and increased MOSFET switching capabilities, digital audio data formats can be directly amplified using power electronic inverters. These amplifiers known as Class-D have an advantage over there analogue counterparts because of their high efficiency. This thesis deals with the signal processing algorithms necessary to convert the digital audio data obtained from the source to a digital pulse width modulated signal which controls a full bridge inverter for audio amplification. These algorithms address difficulties experienced in the past which prevented high fidelity digital pulse width modulators to be implemented. The signal processing algorithms are divided into modular blocks, each of which are defined in theory, designed and simulated in Matlab® and then implemented within VHDL firmware. These firmware blocks are then used to realize a Class-D audio amplifier.

Thesis (MScEng (Electrical and Electronic Engineering))--Stellenbosch University, 2008. / This thesis is aimed at testing commercially available CMOS amplifier ICs at 4 K. Super Conducting Electronics (SCE) will also be used to amplify RSFQ signals for easier detection by CMOS technology and better signal-to-noise ratios. The SCE comprises of a Suzuki stack amplifier, a 250 μA JTL and a DC-to-SFQ converter. The Suzuki stack amplifier is simulated in WRSPICE. It is able to amplify an SFQ signal synchronised with an external clock signal. The amplified signal can then be detected by a normal commercially available CMOS amplifier IC. To keep the noise in the signal to a minimum, the commercial amplifier must be be situated as close as possible to the SCE. The amplifier must therefore be able to operate at 4 K. Ten different amplifier ICs were tested and three was found that worked down to 4 K.

Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2007. / A major challenge in the design and operation of High Temperature Superconducting (HTS) devices is the fabrication of reproducible Josephson junctions with good IcRn products. One objective of this thesis was to fabricate successfully HTS step-edge junctions. This objective necessitated a critical evaluation of the available facilities to provide much needed improvements. These improvements included a newly optimised photolithography process, the incorporation of a three-gridded extraction system into the in-house argon ion mill as well as alterations to the Pulsed Laser Deposition (PLD) system to improve thin film quality. These process modifications finally allowed for the fabrication of novel PrBa2Cu3O7−δ buffered step-edge junctions. These junctions were tested for dc and ac Josephson effects and displayed IcRn products of 1.5 mV at 55 K as well as well-defined Shapiro steps. A second objective was to introduce a high quality thin film deposition system that could produce smooth superconducting films for use in filters and multilayer technology. An Inverted Cylindrical Magnetron system was built and optimised to grow YBa2Cu3O7−δ thin films on MgO (001) substrates. A complete optimisation process of these films are presented by utilising several growth and electrical characterisation methods such as XRD, RBS and AFM.

Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2007. / As secondary objective an introduction to the quantification, theory and measurement of phase noise is presented to make this field of study more accessible for the novice to the field. Available phase noise theory is evaluated at the hand of its application to the design of a low phase noise quartz crystal oscillator. A low phase noise crystal oscillator was designed by application of the presented theory. This oscillator was constructed and measured yielding phase noise low enough to compare favourably with commercially available ultra-low phase noise crystal oscillators. Within the sensitivity of the phase noise measurement equipment good agreement between the theoretically predicted and the measured phase noise was achieved.

Thesis (MScIng)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: In this study the phoneme labels derived from a phoneme recogniser are used for phonetic speaker recognition. The time-dependencies among phonemes are modelled by using hidden Markov models (HMMs) for the speaker models. Experiments are done using firstorder and second-order HMMs and various smoothing techniques are examined to address the problem of data scarcity. The use of word labels for lexical speaker recognition is also investigated. Single word frequencies are counted and the use of various word selections as feature sets are investigated. During April 2004, the University of Stellenbosch, in collaboration with Spescom DataVoice, participated in an international speaker verification competition presented by the National Institute of Standards and Technology (NIST). The University of Stellenbosch submitted phonetic and lexical (non-acoustic) speaker recognition systems and a fused system (the primary system) that fuses the acoustic system of Spescom DataVoice with the non-acoustic systems of the University of Stellenbosch. The results were evaluated by means of a cost model. Based on the cost model, the primary system obtained second and third position in the two categories that were submitted. / AFRIKAANSE OPSOMMING: Hierdie projek maak gebruik van foneem-etikette wat geklassifiseer word deur ’n foneemherkenner en daarna gebruik word vir fonetiese sprekerherkenning. Die tyd-afhanklikhede tussen foneme word gemodelleer deur gebruik te maak van verskuilde Markov modelle (HMMs) as sprekermodelle. Daar word ge¨eksperimenteer met eerste-orde en tweede-orde HMMs en verskeie vergladdingstegnieke word ondersoek om dataskaarsheid aan te spreek. Die gebruik van woord-etikette vir sprekerherkenning word ook ondersoek. Enkelwoordfrekwensies word getel en daar word ge¨eksperimenteer met verskeie woordseleksies as kenmerke vir sprekerherkenning. Gedurende April 2004 het die Universiteit van Stellenbosch in samewerking met Spescom DataVoice deelgeneem aan ’n internasionale sprekerverifikasie kompetisie wat deur die National Institute of Standards and Technology (NIST) aangebied is. Die Universiteit van Stellenbosch het ingeskryf vir ’n fonetiese en ’n woordgebaseerde (nie-akoestiese) sprekerherkenningstelsel, asook ’n saamgesmelte stelsel wat as primˆere stelsel dien. Die saamgesmelte stelsel is ’n kombinasie van Spescom DataVoice se akoestiese stelsel en die twee nie-akoestiese stelsels van die Universiteit van Stellenbosch. Die resultate is ge¨evalueer deur gebruik te maak van ’n koste-model. Op grond van die koste-model het die primˆere stelsel tweede en derde plek behaal in die twee kategorie¨e waaraan deelgeneem is.

Thesis (MSCIng)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: Bi-polar Si transistors optimized for pulse conditions is still the most popular choice as amplification element in the final stages of solid-state radar amplifiers in L and S band. With the radar market being small, the design data for these devices is normally fairly limited and it is up to the designers to thoroughly characterize them for their designs. This is normally done through loadpull experiments. Professional automated load-pull equipment is very expensive especially at the higher power levels. In spite of being automated and under computer control, load-pull exercises still is very time consuming and as such expensive. For small companies that only occasionally need to design such amplifiers it is not economically viable to acquire such equipment and different strategies have to be found to stay competitive. This report investigates such a strategy and its implementation. A procedure to quickly and accurately characterize such devices was developed and two amplifiers were designed and build with this procedure and compared to their traditional counterparts for verification. The results were very promising and with a bit more work, the technique can likely be used to characterize these devices for design work outside of the parameters designated by the manufacturers. / AFRIKAANSE OPSOMMING: Bipolere Silikon transistors wat vir werking onder gepulsde toestande geoptimiseer is, is nog steeds die mees gewilde keuse as versterkingselement in die finale stadiums van vastetoestand radar versterkers in die L en S bande. Met die radar mark wat geredelik klein is, is die ontwerp inligting vir hierdie elemente gewoonlik redelik karig en is dit die taak van die ontwerpers om die elemente te karakteriseer vir hulle ontwerp doeleindes. Dit word normaalweg gedoen deur lastrek eksperimente. Geoutomatiseerde lastrek toerusting is baie duur, veral as dit onder hoë drywingstoestande moet werk. Al is die toerusting geoutomatiseer en onder rekenaar beheer, is lastrek oefeninge nog steeds baie tydrowend en daarom dan ook baie duur. Vir klein maatskappye wat net nou en dan nodig het om sulke versterkers te ontwerp is dit gewoon nie ekonomies regverdigbaar om sulke toerusting aan te skaf nie, en ander strategië moet gevind word om ekonomies kompeterend te bly. Hierdie verslag ondersoek so 'n strategie en die implimentering daarvan. n Prosedure om gepulsde bipolere transistore vinnig en akkuraat te karakteriseer is ontwikkel en twee versterkers is met die prosedure ontwerp en gebou. Die versterkers is geverifieer deur hulle met hulle tradisionele eweknië te vergelyk. Die resultate lyk baie belowend en met n bietjie meer werk kan die metode waarskynlik ook gebruik word om die transistors buite die toepassings gebied, soos deur die vervaardigers aangedui, te gebruik.

Thesis (MScIng)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: Speech enhancement is the process of removing background noise from speech signals. The equivalent process for images is known as image denoising. While the Fourier transform is widely used for speech enhancement, image denoising typically uses the wavelet transform. Research on wavelet-based speech enhancement has only recently emerged, yet it shows promising results compared to Fourier-based methods. This research is enhanced by the availability of new wavelet denoising algorithms based on the statistical modelling of wavelet coefficients, such as the hidden Markov tree. The aim of this research project is to investigate wavelet-based speech enhancement from a statistical perspective. Current Fourier-based speech enhancement and its evaluation process are described, and a framework is created for wavelet-based speech enhancement. Several wavelet denoising algorithms are investigated, and it is found that the algorithms based on the statistical properties of speech in the wavelet domain outperform the classical and more heuristic denoising techniques. The choice of wavelet influences the quality of the enhanced speech and the effect of this choice is therefore examined. The introduction of a noise floor parameter also improves the perceptual quality of the wavelet-based enhanced speech, by masking annoying residual artifacts. The performance of wavelet-based speech enhancement is similar to that of the more widely used Fourier methods at low noise levels, with a slight difference in the residual artifact. At high noise levels, however, the Fourier methods are superior. / AFRIKAANSE OPSOMMING: Spraaksuiwering is die proses waardeur agtergrondgeraas uit spraakseine verwyder word. Die ekwivalente proses vir beelde word beeldsuiwering genoem. Terwyl spraaksuiwering in die algemeen in die Fourier-domein gedoen word, gebruik beeldsuiwering tipies die golfietransform. Navorsing oor golfie-gebaseerde spraaksuiwering het eers onlangs verskyn, en dit toon reeds belowende resultate in vergelyking met Fourier-gebaseerde metodes. Hierdie navorsingsveld word aangehelp deur die beskikbaarheid van nuwe golfie-gebaseerde suiweringstegnieke wat die golfie-ko¨effisi¨ente statisties modelleer, soos die verskuilde Markovboom. Die doel van hierdie navorsingsprojek is om golfie-gebaseerde spraaksuiwering vanuit ‘n statistiese oogpunt te bestudeer. Huidige Fourier-gebaseerde spraaksuiweringsmetodes asook die evalueringsproses vir sulke algoritmes word bespreek, en ‘n raamwerk word geskep vir golfie-gebaseerde spraaksuiwering. Verskeie golfie-gebaseerde algoritmes word ondersoek, en daar word gevind dat die metodes wat die statistiese eienskappe van spraak in die golfie-gebied gebruik, beter vaar as die klassieke en meer heuristiese metodes. Die keuse van golfie be¨ınvloed die kwaliteit van die gesuiwerde spraak, en die effek van hierdie keuse word dus ondersoek. Die gebruik van ‘n ruisvloer parameter verhoog ook die kwaliteit van die golfie-gesuiwerde spraak, deur steurende residuele artifakte te verberg. Die golfie-metodes vaar omtrent dieselfde as die klassieke Fourier-metodes by lae ruisvlakke, met ’n klein verskil in residuele artifakte. By ho¨e ruisvlakke vaar die Fouriermetodes egter steeds beter.

Thesis (MScIng)--University of Stellenbosch, 2004. / ENGLISH ABSTRACT: The preliminary development of a FM radio receiver module is discussed. An existing narrowband system operating between 48MHz and 50MHz will be replaced. Digital components were investigated, compared and used with analogue techniques to build a more flexible two-way radio receiver system. A direct digital synthesizer was considered as a replacement for the current synthesized phased lock loop local oscillator and much attention was given to the local oscillator and mixer design, characteristics and measurement procedures. A detailed study of receiver systems was undertaken to determine the specifications needed for every receiver component to achieve satisfactory receiver performance in the end. Receiver characteristics as well as receiver measurement procedures are defined. A software tool was developed to aid the design process, establishing computationally whether the receiver specifications are met prior to the final design. The complete design process, from fundamental specifications through to the developed final receiver module is discussed. A modular design approach was used to guarantee easy manufacturing, substitution and testing. This approach comprises the break-down of the receiver into well defined components that are each matched to 50O. The separate components of the system were designed, measured and characterized to make it possible to replace only a single component instead of the entire system when a part becomes redundant. / AFRIKAANSE OPSOMMING: Die grondslag vir die ontwikkeling van ‘n FM radio ontvanger module word in hierdie dokument gelê. ‘n Bestaande noubandstelsel wat tussen 48MHz and 50MHz ontvang word vervang deur hierdie nuwe stelsel wat aangewend sal kan word in die bestaande tweerigtingradio se omhulsel. Digitale komponente is ondersoek, vergelyk en gebruik saam met analoogtegnieke om ‘n meer buigsame radiostelsel te bewerkstellig. ‘n Direkte digitale sintitiseerder is oorweeg as ‘n vervanging vir die huidige fasesluitlus ossillator met heelwat klem op die oscillator-en mengerontwerp, komponent spesifikasies en metingsprosedures. ‘n Diepgaande studie van ontvangerstelsels is gedoen om te bepaal wat die tipiese spesifikasies vir elke ontvangerstadium is, sodat die finale ontvanger se spesifikasies behaal kan word. Ontvanger eienskappe en meetprosedures word volledig gedefinieer. ‘n Sagtewareprogram is ontwikkel om die ontvanger-ontwerpsproses te vergemaklik deur vooraf te kan vasstel watter ontvangerspesifikasies bereik sal kan word al dan nie. Die volledige ontwerpsproses, vanaf fundamentele spesifikasies tot by die finale ontvanger word omskryf. ‘n Modulere-ontwerp prosedure is gebruik ter versekering van die maklike vervaardiging, vervanging en toetsing van elke komponent. Die radio is tydens ontwerp opgebreek in boublokkies wat elkeen aangepas word na 50O. Elke aparte boublokkie van die ontvangerstelsel is afsonderlik ontwerp, gemeet en volledig gespesifiseer om dit moontlik te maak om slegs een komponent te vervang in plaas van die hele stelsel wanneer ‘n enkele komponent nie meer beskikbaar is nie.

Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2005. / The Gaussian mixture model (GMM) performs very effectively in applications such as speech and speaker recognition. However, evaluation speed is greatly reduced when the GMM has a large number of mixture components. Various techniques improve the evaluation speed by reducing the number of required Gaussian evaluations.