Les techniques ultrasonores offrent de nombreux avantages, à la fois par leur utilisation facile et la sécurité du patient. De plus, la recherche, visant à étendre les possibles champs d’applications, est particulièrement active. Cependant, l’accès à des équipements adaptés et supportant des logiciels est conditio sine qua non pour l’expérimentation de nouvelles techniques. Ce projet de thèse traite des problématiques de traitement du signal et d'image dans un contexte d'imagerie médicale et vise à répondre à deux objectifs scientifiques: le premier consiste à contribuer au développement d'une puissante plateforme de recherche (ULA-OP) alors que le second a pour objectif d'introduire et de valider, grâce à cette plateforme, des méthodes de traitement non-standard qui ne pourraient pas être évaluées avec des équipements médicaux commerciaux. ULA-OP, un équipement recherche qui donne accès aux développeurs à une grande liberté de contrôle et de configuration l’ensemble des parties actives du système, de la transmission aux traitements des signaux échographiques. Il offre aussi la possibilité d’accéder aux signaux bruts à n’importe quel niveau de la chaîne de réception. Durant cette thèse, les capacités du système ont été améliorées en implémentant des outils logiciels comme des simulateurs de champ acoustique (propagation linéaire et non-linéaire), et en développant des programmes de génération de signaux post-échographique. L’ULA-OP a été crucial pour développer et tester différentes techniques non-standard telles qu’un schéma adaptatif de formation de voie et une méthode d’imagerie Doppler couleur/vecteur, qui seront détaillés dans le manuscrit. En particulier, une nouvelle méthode a été développée pour des applications d’élastographie quasi-statique. Cette méthode, basée sur un algorithme d’estimation du mouvement dans le domaine fréquentiel et combinée à une méthode d’imagerie haute fréquence, a permis d’améliorer la qualité des élastogrammes obtenus. Cette nouvelle méthode a d’abord été testée in-vitro par des traitements hors ligne des signaux reçus et pour ensuite être implémentée en temps réel sur le ULA-OP. Les résultats obtenus montrent que cette technique est performante et que les élastogrammes présentent une qualité supérieure comparée à ceux obtenues avec les méthodes connues de la littérature / Ultrasound techniques offer many advantages, in terms of both ease of realization and patients’ safety. The research aimed at expanding the fields of application, is nowadays particularly active. The availability of suitable hardware and supporting software tools is condicio sine qua non for the experimentation of new techniques. This Ph.D project addresses signal/image processing issues in medical ultrasound and seeks to achieve two major scientific goals: the first is to contribute to the development of a powerful ultrasound research platform (ULA¬OP), while the second is introducing and validating, through this platform, non-standard methods which could not be tested with commercial equipment. ULA-OP is a research system, which gives developers great freedom in terms of management and control of every section, from signal transmission to echo-signal processing; it also offers the possibility to access raw data at any point in the receive chain. During the thesis, the capabilities of the system were improved by creating advanced software tools, such as acoustic field simulators (for linear and nonlinear propagation), and by developing echo-signals post-elaboration programs. ULA-OP was crucial to develop and test various non-standard techniques such as an adaptive beamforming scheme and a color/vector Doppler imaging method, which will be detailed in this thesis. In particular, a novel technique was developed for quasi-static elastography applications. This technique, based on a frequency domain displacement estimation algorithm, combined with a high-frame-rate averaging method, aims at improving the quality of the elastograms. The new method was first tested in-vitro by offline processing the received signals, and then it was implemented in real-time on ULA-OP. The results show that this technique is effective and that the obtained elastograms present higher quality compared with those obtained with standard algorithms

Made available in DSpace on 2014-07-29T15:08:20Z (GMT). No. of bitstreams: 1 Dissertacao_Hugo_Silva.pdf: 1463961 bytes, checksum: 3a6b5b3e5e7db58659b1bb58d6d4cd66 (MD5) Previous issue date: 2009-03-13 / This work is concerned with the estimation of Direction-Of-Arrival (DOA) angles of plane waves impinging on a sensor array. Among all methods of estimation found in litera-ture, MODEX (MODE with eXtra roots) outstands for its performance and computational complexity. However, recently, a method called SEAD (SEArch of Direction by differential spectrum) was proposed. It has shown better estimation performance against noise than MODEX has. However, its computational complexity is prohibitive for real-time applications. In order to reduce it s computational complexity, a new estimate selection procedure on SEAD is proposed, that yields to significantly less candidate angles than before. Additionally, the introduction of iterative refinements on estimates has allowed improving resolution as well as complexity reduction / Este trabalho aborda o problema de estimação de ângulos de incidência (DOA) de ondas planas sobre um arranjo de sensores. Dentre os vários métodos encontrados na literatu-ra, o método MODEX (MODE with eXtra roots) se destaca por seu desempenho e complexi-dade computacional. Recentemente, foi proposto o método SEAD (SEArch of Direction by differential spectrum), que apresenta desempenho de estimação mais robusto que o MODEX aos efeitos do ruído, contudo, possui complexidade computacional proibitiva para aplicações de tempo-real. Na busca por reduzir esta complexidade computacional, este trabalho apresenta uma nova proposta de seleção de estimativas para o SEAD que gera uma quantidade signifi-cativamente menor de ângulos candidatos. Adicionalmente, a inserção de um processo iterati-vo de refinamento de soluções permitiu que a resolução do estimador fosse aprimorada em relação àquela inicialmente ajustada, além de reduzir a complexidade computacional da sele-ção

During the launch sequence of the United Launch Alliance Delta IV launch vehicle, large amounts of pure hydrogen are introduced into the launch table and ignited by Radial-Outward-Firing-Igniters (ROFIs). This ignition results in a significant flame, or plume, that rises upwards out of the launch table due to buoyancy. The presence of the plume causes increased and unwanted heat loads on the surface of the vehicle. A proposed solution is to add a series of fans and structures to the existing launch table configuration that are designed to inject ambient air in the immediate vicinity of the launch vehicle's nozzles to suppress the plume rise. In addition to the air injection, secondary fan systems can be added around the launch table openings to further suppress the hydrogen plume. The proposed air injection solution is validated by computational fluid dynamics simulations that capture the combustion and compressible flow observed during the Delta IV launch sequence. A solution to the hydrogen plume problem will have direct influence on the efficiency of the launch vehicle: lower heat loads result in thinner vehicle insulation and thus allow for a larger payload mass. Current results show that air injection around the launch vehicle nozzles and air suppression around the launch table openings significantly reduces the size of the plume around the launch vehicle prior to liftoff.

No description available.

The aim of the master thesis is to sum up theoretical information about beamforming methods of microphone arrays and to verify their functionality. At the beginning of this work there are simulated different varietes of linear uniform and nonuniform microphone arrays and circular arrays. The results are verificated by a practical measurement in ideal conditions. Then I will focuse on implementation of the DAS(Delay And Sum), SAB(Sub Array Beamforming), CDB(Constant Directivity Beamforming), CDB-CA(CDB-Circular Arrays) beamformer including theoretical and practical verification of the functionality in ideal conditions. At the end of this thesis are all beamforming methods compared with each other at SNR(signal to Noise Ratio) and directivity parameters.

The goal of this master’s thesis is to explore the possibilities of multichannel localization of acoustic signal sources and their following application on a real signal localization and separation, using Beamforming methods. During this thesis two beamforming methods were selected, namely Delay and Sum a Constant Directivity Beamforming - Circular Arrays, and were applicated on real environment signals using two microphone arrays’ geometries ULA (Uniform linear array) and UCA (Uniform Circular array).