This thesis addresses the design, analysis, and experimental validation of a high-directivity and high coupling microwave directional coupler. The motivating application is in broadband signal routing between cores of multi-core processors, where the delay of simple wire interconnects introduces unacceptable latency. The performance goals include scalability with frequency, a coupling coefficient of 3 dB, directivity larger than 40 dB, high return loss, low insertion loss below 3 dB at the center frequency, and small footprint. The approach to this problem taken in the thesis is a combination of edge and broad-side coupling in a multi-layer, multi-conductor microstrip coupled-line system. The two coupling mechanisms between neighboring pairs of coupled lines, along with appropriate end interconnections, allow for reduced size and design that achieves equal propagation velocities for the different modes supported by the five-conductor guiding structure that contribute to coupling. To validate the approach, a coupler designed for operation at 1 GHz is demonstrated to have a isolation of -22 dB with a coupling coefficient of 3\,dB and a return loss of -20 dB. The coupler is implemented on a FR-408 substrate with a permittivity of 3.66 and 1.17mm and 0.17mm thicknesses, and a total area of 12.65 cm^2. Three metalization layers are used in the design, with edge and broad-side coupled pairs of lines on the top two layers and diagonal end interconnects between the top and bottom lines. The coupler design is then scaled to 3 GHz by shortening the coupled-line length, and established -24 dB isolation, coupling of 3 dB, return loss of -20 dB, and has a total area of 6.9 cm^2. The analysis of the coupler shows that full-wave electromagnetic modeling agrees well with measurements and is necessary during the design process, while circuit analysis with built-in coupled-line models shows poorer agreement with experimental data. A tolerance analysis shows that the coupler performance is most sensitive to milling precision and separation between coupled-lines. Based on the measured and simulated results, it is shown that this type of coupler can be further scaled to higher frequencies and on-chip implementations for signal distribution in multi-core processors, or any other application where a number of components need to be interconnected with low latency and no reflection.

In this work we investigate adaptive techniques which can be used to improve the performance of highly scalable video compression schemes under resolution scaling. We propose novel content adaptive methods for motion compensated 3D discrete wavelet transformation (MC 3D-DWT) of video. The proposed methods overcome problems of ghosting and non-aligned aliasing artifacts, which can arise in regions of motion model failure, when the video is reconstructed at reduced temporal or spatial resolutions. We also study schemes which facilitate simultaneous scaling of compressed video bitstreams based on both constant bit-rate and constant distortion criteria, using simple and generic scaling operations. In regions where the motion model fails, the motion compensated temporal discrete wavelet transform (MC TDWT) causes ghosting artifacts under frame-rate scaling, due to temporal lowpass filtering along invalid motion trajectories. To avoid ghosting artifacts, we adaptively select between different lowpass filters, based on a local estimate of the motion modelling accuracy. Experimental results indicate that the proposed adaptive transform substantially removes ghosting artifacts while also preserving the high compression efficiency of the original MC TDWT. We also study the impact of various MC 3D-DWT structures on spatial scalability. Investigating the interaction between spatial aliasing, scalability and energy compaction shows that the t+2D structure essentially has higher compression efficiency. However, where the motion model fails, structures of this form cause non-aligned aliasing artifacts under spatial scaling. We propose novel adaptive schemes to continuously adapt the structure of MC 3D-DWT based on information available within the compressed bitstream. Experimental results indicate that the proposed adaptive structure preserves the high compression efficiency of the t+2D structure while also avoiding the appearance of non-aligned aliasing artifacts under spatial scaling. To provide simultaneous rate and distortion scaling, we study ???layered substream structure. Scaling based on distortion generates variable bit-rate traffic which satisfies the desired average bit-rate and is consistent with the requirements of leaky-bucket traffic models. We propose a novel method which also satisfies constraints on instantaneous bit-rate. This method overcomes the weakness of previous methods with small leaky-bucket buffer sizes. Simulation results indicate promising performance with both MC 3D-DWT interframe and JPEG2000 intraframe compression.

In this work we investigate adaptive techniques which can be used to improve the performance of highly scalable video compression schemes under resolution scaling. We propose novel content adaptive methods for motion compensated 3D discrete wavelet transformation (MC 3D-DWT) of video. The proposed methods overcome problems of ghosting and non-aligned aliasing artifacts, which can arise in regions of motion model failure, when the video is reconstructed at reduced temporal or spatial resolutions. We also study schemes which facilitate simultaneous scaling of compressed video bitstreams based on both constant bit-rate and constant distortion criteria, using simple and generic scaling operations. In regions where the motion model fails, the motion compensated temporal discrete wavelet transform (MC TDWT) causes ghosting artifacts under frame-rate scaling, due to temporal lowpass filtering along invalid motion trajectories. To avoid ghosting artifacts, we adaptively select between different lowpass filters, based on a local estimate of the motion modelling accuracy. Experimental results indicate that the proposed adaptive transform substantially removes ghosting artifacts while also preserving the high compression efficiency of the original MC TDWT. We also study the impact of various MC 3D-DWT structures on spatial scalability. Investigating the interaction between spatial aliasing, scalability and energy compaction shows that the t+2D structure essentially has higher compression efficiency. However, where the motion model fails, structures of this form cause non-aligned aliasing artifacts under spatial scaling. We propose novel adaptive schemes to continuously adapt the structure of MC 3D-DWT based on information available within the compressed bitstream. Experimental results indicate that the proposed adaptive structure preserves the high compression efficiency of the t+2D structure while also avoiding the appearance of non-aligned aliasing artifacts under spatial scaling. To provide simultaneous rate and distortion scaling, we study ???layered substream structure. Scaling based on distortion generates variable bit-rate traffic which satisfies the desired average bit-rate and is consistent with the requirements of leaky-bucket traffic models. We propose a novel method which also satisfies constraints on instantaneous bit-rate. This method overcomes the weakness of previous methods with small leaky-bucket buffer sizes. Simulation results indicate promising performance with both MC 3D-DWT interframe and JPEG2000 intraframe compression.

An optomechanical system consists of an optical cavity mode coupled to a mode of a mechanical oscillator. Depending on the configuration of the system, the optomechanical interaction can be used to drive or cool the mechanical mode, coherently swap the optical and mechanical states, or create entanglement. A multimode optomechanical system consists of many optical (mechanical) modes coupled to a mechanical (optical) mode. With the tools of the optomechanical interaction, multimode optomechanical systems provide a rich platform to study new physics and technologies. A central challenge in optomechanical systems is to mitigate the effects of the thermal environment, which remains significant even at cryogenic temperatures, for mechanical oscillators typically used in optomechanical systems. The central theme of this thesis is to study how the properties of multimode optomechanical systems can be used for such mitigation of thermal noise. The most straightforward extension of an optomechanical system to a multimode system is to have a single optical mode couple to two mechanical modes, or a single mechanical mode couple to two optical modes. In this thesis, we study both types of multimode system. In each case, we study the formation of a dark mode, an eigenstate of the three-mode system that is of particular interest. When the system is in a dark state, the two modes of similar character (optical or mechanical) interact with each other through the mode of dissimilar character, but due to interference, the interaction becomes decoupled from the properties of the dissimilar mode. Another interesting application of the three-mode system is two-mode optical entanglement, generated through mechanical motion. Such entanglement tends to be sensitive to thermal noise. We propose a new method for generating two-mode optical entanglement in the three-mode system that is robust against the thermal environment of the mechanical mode. Finally, we propose a novel, scalable architecture for a quantum computer. The architecture makes use of the concepts developed earlier in the thesis, and applies them to a system that on the surface looks quite different from the standard optomechanical system, but is formally equivalent. This dissertation includes previously published and unpublished coauthored material.

Dans les dernières décennies, la réalité virtuelle a connu un essor fulgurant dans de nombreuses disciplines. Elle permet via des paradigmes d’interaction et d’immersion de plonger l’utilisateur au cœur d’un environnement artificiel crée numériquement. Ces paradigmes s’appuient sur l’utilisation des interfaces sensori-motrices bien spécifiques qui permettent à l’utilisateur d’interagir et accomplir des tâches particulières dans l’environnement virtuel. Néanmoins, des nombreux problèmes, d’origine technologique, sont souvent présents et peuvent pénaliser la qualité de l’interaction ainsi que le degré d’immersion de l’utilisateur dans l’environnement virtuel.L’objectif de cette thèse est de proposer une procédure complète visant à guider l’utilisateur à calibrer une interface sensori-motrice spécifique et par conséquent tenter de pallier à certains défauts technologiques. L’originalité de la thèse réside dans l’utilisation d’une approche qui combine deux domaines de recherche qui ne s’associent que très rarement : celui du traitement de données et celui de la réalité virtuelle. Cette approche servira de cadre théorique et technique pour la conception d’une procédure de calibration complète permettant de garantir une interaction continue et précise dans l’environnement virtuel.Afin de contrebalancer les défauts et limites techniques, le travail a été conduit sur plusieurs fronts : acquisition, traitement de données et validation. La première phase est marquée par l’utilisation d’un protocole innovant dans la mesure où il repose sur les techniques de réalité virtuelle pour récolter les données de calibration. Dans la deuxième phase, deux techniques de calibration ont été proposées pour améliorer la précision absolue de l’interface de réalité virtuelle. Les deux techniques se distinguent par leurs qualités d’approximateurs universels ainsi que par leurs capacités à estimer les sorties du système concerné à partir des entrées sans connaître à priori son modèle mathématique. Dans la dernière phase, deux prototypes d’applications de réalité virtuelle ont été développées pour s’assurer de la pertinence de notre approche. / Over last decades, virtual reality has been widely used in many disciplines. It is able to plunge the user at the heart of an artificial environment created digitally through interaction and immersion paradigms. These paradigms are based on the use of very specific interfaces that help user to interact and performspecific tasks in the virtual environment. Nevertheless, many technical problems are often present and may penalize the quality of that interaction and may break user immersion in the virtual environment.The goal of this thesis is to build a comprehensive procedure to guide the user to calibrate a virtual reality interface and therefore attempt to overcome some technological shortcomings. The originality of the thesis is the use of an approach that combines two areas of research that will combine very rarely, that of data processing and the virtual reality. This approach will provide theoretical and technical framework for the design of a comprehensive calibration procedure to ensure continuous and precise interaction in the virtual environment.To overcome problems described above, the work was conducted on several fronts :data acquisition, processing and validation. The first step is by the use of a new protocol insofar as it is based on virtual reality techniques to collect calibration data. In second step, two calibration methods have been proposed to improve the absolute accuracy of the virtual reality interface. Both methods are universal approximators as well as their ability to estimate the outputs of the involved system from inputs even the model of the system being calibrated remains unknown. In the last step, two virtual reality applications prototypes were developed in order to assess the relevance of our approach.

This dissertation devotes itself to algorithmic approaches to the problem of scalable multicast with network coding. Several original contributions can be concluded as follows. We have proved that the scalable multicast problem is NP-hard, even with the ability to perform network coding at the network nodes. Several approximations are derived based on different heuristics, and systematic approaches have been devised to solve those problems. We showed that those traditional routing methods reduce to a special case in the new network coding context. Two important frameworks usually found in traditional scalable multicast solutions, i.e. layered multicast and rainbow multicast, are studied and extended to the network coding scenario. Solutions based on these two frameworks are also presented and compared. Suprisingly, these two distinctive approaches in the traditional sense become connected and share a similar essence of data mixing in the light of network coding. Cases are presented where these two approaches become equivalent and achieve the same Performance. We have made significant advances in constructing good solutions to the scalable multicast problem by solving various optimization problems formulated in our approaches. In the layered multicast framework, we started with a straight-forward extension of the traditional layered multicast to the network coding context. The proposed method features an intra-layer network coding technique which is applied on different optimized multicast graphs. Later on, we further improved this method by introducing the inter-layer network coding concept. By allowing network coding among data from different data layers, more leverage is gained when optimizing the network flow, thus higher performance is achieved. In the rainbow multicast framework, we choose uneven erasure protection (UEP) technique as the practical way of constructing balanced MDC, and optimize this MDC design using the max-flow information of receivers. After the MDC design is finalized, a single linear network broadcast code is employed to deliver MDC encoded data to receivers while satisfying the individual max-flow of all the receivers. Although this rainbow multicast based solution may sacrifice the performance in some cases, it greatly simplifies the rate allocation problem raised in the layered multicast framework. The use of one single network code also makes the network codes construction process a lot clearer. Extensive amount of simulation is performed and the results show that network coding based scalable multicast solutions can significantly outperform those traditional routing based solutions. In addition to the imaginary linear objective function used in the simulation, the practical convex objective function and real video data are also used to verify the effectiveness of the proposed solutions. The role of different parameters in the proposed approaches are analyzed, which gives us more guidelines on how to fine-tune the system. / Thesis / Doctor of Philosophy (PhD)

ITC/USA 2007 Conference Proceedings / The Forty-Third Annual International Telemetering Conference and Technical Exhibition / October 22-25, 2007 / Riviera Hotel & Convention Center, Las Vegas, Nevada / Real-time transmission of airborne images to a ground station is highly desirable in many telemetering applications. Such transmission is often through an error prone, time varying wireless channel, possibly under jamming conditions. Hence, a fast, efficient, scalable, and error resilient image compression scheme is vital to realize the full potential of airborne reconnaisance. JPEG2000, the current international standard for image compression, offers most of these features. However, the computational complexity of JPEG2000 limits its use in some applications. Thus, we present a scalable low complexity coder (SLCC) that possesses many desirable features of JPEG2000, yet having high throughput.

"The maintenance and inspection of societal structures and equipment such as skyscrapers, bridges, and ship hulls are important to maintaining a safe lifestyle. Improper maintanance and delayed inspection can lead to catastrophic failure. In lieu of placing humans in potential harm, mobile robotic machining systems can be used to enable remote repair and maintenance within constrictive, hazardous, and inaccessible environments. Due to their intrinsic high mobility and 6-DOF control, hexapod walking robots are a salient solution to mobile machining. However, the static structure of traditional hexapod robots can be rather limiting when attempting to traverse over irregular terrain or manipulating objects. This research realizes a new scalable hexapod robot and analyzes the lateral stable workspace of the robot under different external loading conditions. The scalable design allows the robot to extend its legs which enhances the workspace and improves stability while manuevering through constrictive and irregular terrain. The design incorporates two additional prismatic joints into the legs of the traditional hexapod robot design providing a compact, rigid, and efficient design. The electronic printed circuit boards were designed and assembled in-house. A distributed control architecture was implemented to off-load low-level leg control to dedicated leg controllers. An analysis on the lateral stable workspace of the scalable hexapod robot under different external loading conditions is presented. A dynamic stable workspace criterion is derived. The stable workspace criterion provides a metric for comparing stable workspaces between hexapod robots with different configurations. Multiple simulations and physical experiments were conducted to demonstrate the advantages of a scalability in hexapod designs."

In this thesis, a 14-bit low-power Analog-to-Digital Converter (ADC) is designed for sensor applications. Following on previous work, the ADC is designed to be frequency scalable by 1000 times from 1.67S/s to 1.67kS/s. To reduce power, class AB opamps are used. The design was fabricated in 0.18um CMOS and occupies an area of 0.35mm2. Operating at full-rate as a Delta-Sigma modulator, the ADC achieves 91.8dB peak SNDR while consuming 83uW. In incremental mode, the ADC powers off periodically to achieve frequency scalability, maintaining 84.7dB to 89dB peak SNDR while operating from 1.67S/s to 1.67kS/s.

In this thesis, a 14-bit low-power Analog-to-Digital Converter (ADC) is designed for sensor applications. Following on previous work, the ADC is designed to be frequency scalable by 1000 times from 1.67S/s to 1.67kS/s. To reduce power, class AB opamps are used. The design was fabricated in 0.18um CMOS and occupies an area of 0.35mm2. Operating at full-rate as a Delta-Sigma modulator, the ADC achieves 91.8dB peak SNDR while consuming 83uW. In incremental mode, the ADC powers off periodically to achieve frequency scalability, maintaining 84.7dB to 89dB peak SNDR while operating from 1.67S/s to 1.67kS/s.