No description available.

De nos jours certaines applications vidéo de par des limitations de mémoire et de capacités de calculs requièrent un système où l'encodage d'une séquence vidéo doit être le plus simple possible. Le codage vidéo distribué nouveau paradigme de la compression vidéo semble prometteur pour ce genre de demande. Il permet une compression peu complexe de séquences animées. Son principe repose sur les travaux de D. Slepian et J. K. Wolf [1] ainsi que ceux de A. D. Wyner et J. Ziv [2] menés dans les années 70. La présente thèse énonce trois différentes contributions relatives au codage vidéo conventionnel et distribué. La première est un nouveau procédé de filtrage vidéo basé sur l'utilisation d'arbres de décision. Afin de réduire les artefacts visuels issus d'une trop forte compression, les arbres de décision identifient et appliquent les meilleurs coefficients de filtrage en fonction des données à filtrer. La seconde contribution concerne un schéma de codage vidéo distribué où l'on évite de transmettre au décodeur 1 information relative aux blocs de l'image les moins significatifs. De cette manière, on engendre une diminution importante du débit binaire et une réduction de la complexité de décodage. Ce schéma de codage repose sur une séparation en couche des coefficients de transformée en cosinus discret de l'image. Les coefficients DC sont les premiers à être transmis au décodeur. De là, ils sont analysés par ce dernier afin de trouver les blocs les plus significatifs de l'image. Le décodeur indique alors à l'encodeur, via un canal de retour, quels sont les coefficients AC des blocs à transmettre. Enfin, la dernière contribution consiste en une méthode de représentation binaire adaptative des images dans les procédés de codage vidéo distribué. Cette représentation permet de réduire efficacement le débit binaire et tenant compte uniquement de l'information source la plus pertinente. Ici encore cette méthode repose sur une utilisation plus efficace d 'un canal de retour.

Deep Vein Thrombosis, a potentially fatal event, occurs when a blood clot forms within the deep veins of the body. This most frequently manifests in the lower extremities. The goal of this research was to build an inexpensive device that could apply therapeutic compressive pressure to the lower leg to aid in the prevention of deep vein thrombosis using only mechanical input from the user. Several different prototypes were designed and built with varying degrees of success. Characterization of the final prototype required calibration of pressure and force measurement sensors. Additionally, a mathematical model was developed in order to predict how changes in the design of the device, as well as differing sizes and shapes of lower legs, would impact the amount of applied pressure. The predictions of this mathematical model were found to be substantially larger when compared against empirical data. However, there is evidence to indicate that the final prototype could be minimally altered to apply ample therapeutic pressure. / Bioengineering

The purpose of this study was to analyze the anisotropic behaviour of layered soil at small strain level. This thesis comprised experimental, numerical, and theoretical aspects and provided an insight to the concept of the ‘equivalent’ homogeneous cross-anisotropic material indicating that the layered soil can be replaced by an equivalent transversely isotropic material under certain conditions. In the experimental program, this study extended Vaid’s work (1971) to conduct K0-compression tests in a rigid triaxial cell with a flexible lateral boundary for the determination of the K0-value of normally consolidated soil. The error induced by the compliance of the cell-water system that always existed in the previous study was successively eliminated by attaching a compliance correction system (i.e., GDS controller) to the triaxial cell. Three stress path tests (i.e., the K0-compression test, the plane strain compression test, and the hydrostatic pressure compression test) were conducted consecutively in a rigid triaxial cell for a layered soil specimen to determine the elastic cross-anisotropic properties of the equivalent homogeneous material. The applicability of the proposed approach was demonstrated by conducting the tests on two types of soil (i.e., the homogeneous soil and the layered soil). A three-parameter constitutive equation for describing the soil’s cross-anisotropic elastic behaviour was modified in a basic FEM program. Based on the material properties determined by the experimental study as an input, numerical simulations (i.e., the numerical K0-test) using FEM were conducted to compare numerical results with the test results. In the theoretical part, this study made use of the Reuss and the Voigt approximations and proposed a simple, yet physically meaningful, approach to determine the equivalent cross-anisotropic elastic properties of a multilayered medium. To simplify the exposition, a multilayered medium with two constituent materials that were both isotropic was examined. / Thesis / Master of Applied Science (MASc)

The objectives of the thesis were to find the mechanisms that govern compression failures in paperboard and to find the link between manufacturing process and paperboard properties. The thesis also investigates two different test methods and evaluates how suitable they are for paperboard grades. The materials are several commercial board grades and a set of hand-formed dynamic sheets that are made to mimic the construction of commercial paperboard. The method consists of mounting a stereomicroscope on a short-span compression tester and recording the compression failure on video, long-span compression testing and standard properties testing. The observed failure modes of paperboard under compression were classified into four categories depending on the appearance of the failures. Initiation of failure takes place where the structure is weakest and fiber buckling happens after the initiation, which consists of breaking of fiber-fiber bonds or fiber wall delamination. The compression strength is correlated to density and operations and raw materials that increase the density also increases the compression strength. Short-span compression and Long-span compression are not suitable for testing all kinds of papers; the clamps in short-span give bulky specimens an initial geometrical shape that can affect the given value of compression strength. Long-span compression is only suitable for a limited range of papers, one problem with too thin papers are low wavelength buckling.

The internal vertebral venous plexus (IVVP) is an extensive vascular network recently implicated in various human and canine spinal disorders. Nevertheless, little recent information is available regarding normal anatomy of canine IVVP and its role in acute spinal injuries. The objectives of the study were; (1) to describe the normal IVVP morphology in the canine cervical region using transverse anatomy sections and computed tomography (CT), (2) to develop a technique for CT examination of the IVVP in vivo, (3) to analyze the quantitative characteristics of the IVVP, and (4) to assess the effect of acute experimental spinal cord compression on IVVP morphology. In the first experiment, CT of the cervical vertebral canal was performed in 6, normal, adult mixed-breed dogs. After dogs were euthanized, a gelatin and iothalamate mixture was injected into the right external jugular vein. Cadavers were then frozen to â 8°C, sliced into transverse sections, and compared with CT images. Vascular components such as the IVVP, interarcuate veins, intervertebral veins, and vertebral veins were accurately depicted on CT images. In the second experiment, CT venography was performed using a biphasic IV injection of iodinated contrast medium. Dimensions of the IVVP and other vertebral canal components were calculated for the C3-C7 vertebral region. Sagittal diameters of the IVVP ranged from 0.6 mm to 3.2 mm. The IVVP area occupied 30.61% of the cervical vertebral epidural space area. When C3-C7 segments were considered as a group, IVVP area dimensions were significantly correlated (r > 0.7, p < 0.0001) with vertebral canal area and dural sac area. In the last experiment, acute spinal cord compression (ASCC) was induced and maintained for 10 minutes using an angioplasty balloon catheter device over the C3/4 vertebral region in 6 dogs. Dogs were evaluated prior to, during, and after compression using digital subtraction venography (DSV) and CT venography. Results showed that ASCC produced a significant change in diameter of the IVVP at the site of compression. This effect persisted during the post-compression period. In conclusion, findings indicate that CT venography and DSV accurately depict the IVVP in dogs, and that significant changes of the IVVP morphology occur under ASCC conditions. / Ph. D.

Clay consolidation has generally been considered from a macro-scale perspective by measuring the macro-scale compression of a clay soil over time. Clay particles in consolidation tests experience shear and normal forces at the inter-particle level due to force applied to the soil at the macro-scale. These shear and normal forces cause the particles to slide at the micro-scale and produce macro-scale changes in soil volume and shape. By considering the inter-particle interactions at the micro-scale, the shear force - normal force - velocity relationship can be described by the Rate Process Theory (RPT). This research investigated the use of the RPT for analyzing sliding at individual clay particle contacts during secondary compression to describe macro-scale clay behavior. The novel micro-scale friction experiments conducted in this research demonstrated that an Atomic Force Microscope (AFM) can be used to obtain coefficient of friction (μ) measurements for montmorillonite. This method allows for the measurements to be performed over spatial scales of a few microns, can be done under dry conditions or a wide range of aqueous solutions, and requires no calibration beyond making a few microscopic measurements of the probe. Control tests of silica on mica (μ = 0.29 ± 0.02) agree with literature values where limits indicate one standard deviation.μ values for wet and dry sodium montmorillonite were determined to be 0.20 ± 0.03 and 0.72 ± 0.03, respectively. The micro-scale AFM and macro-scale triaxial shear, ring shear, and direct shear experimental data ofμ as a function of sliding velocity were found to match well with those calculated using common RPT parameter values. The activation energy for the macro-scale triaxial shear and corresponding micro-scale friction regime experiments fall within the expected range for pure montmorillonite of 84–109 kJ/mol. Additionally, the micro-scale and macro-scale experimental results fall within the expected range for the number of bonds per unit of normal force of 10^7–10^9 bonds/N. A discrete element method (DEM) model was developed to calculate thin, disk-shaped clay particle movement in three dimensions during compression using the RPT as a contact model. The DEM compression results were compared to macro-scale consolidation experiments conducted on the same reference clay as the micro-scale AFM experiments. The influences on the compression of the number of bonds at each clay contact per unit of normal contact force and the activation energy were quantified. Increasing the activation energy decreased the compression, as expected. Similarly, increasing the number of bonds per unit of normal force at the contacts decreased the compression, as expected. Realistic clay fabrics with varying particle sizes, particle size distributions, and aspect ratios led to a compression model with behavior similar to the macro-scale laboratory compression tests. This research provides evidence of the close correspondence between macro-scale and micro-scaleμ measurements and contributes to multi-disciplinary understanding of factors that control friction between clay particles and deformation of clay masses. The results from this work can be applied to a wide range of time-dependent phenomena such as clay secondary compression, shear deformation, and fault dynamics behavior. / Ph. D.

The Byrne (1991) model was developed to predict excess pore water pressure for saturated sands under cyclic loading. However, the model can also be used to predict seismic compression in dry or partially saturated clean sands, which is the focus of this research. The original Byrne (1991) model has two primary limitations. One limitation is that calibration coefficients for the model have only been developed for clean sand, while seismic compression is a concern for a variety of soil types in engineering practice. Another limitation is that the existing calibration coefficients are solely correlated with soil relative density. This is in contrast to findings from studies performed over the last two decades that show various environmental and compositional factors, in addition to relative density, influence seismic compression behavior. To overcome these shortcomings and others the model was transformed to allow it to be implemented in "simplified" and "non-simplified" manners and systematic model calibration procedures were developed by means of MATLAB code. Both "simplified" and "non-simplified" variants of the model are used to analyze a site in Japan impacted by the 2007, Mw6.6 Niigata-ken Chuetsu-oki earthquake. The results from the analyses are in general accord with the post-earthquake field observations and highlight the utility and versatility of the models. / Master of Science / Earthquake shaking can cause compression of volume in soil, which may induce damage to various infrastructures. This phenomenon is known as seismic compression. Byrne (1991) proposed one model that can be used to evaluate the magnitude of seismic compression. However, this model has two significant limitations. One limitation is its coefficient expression is suitable for merely one soil type, while seismic compression is a concern for a variety of soil types in engineering practice. Another limitation is that the existing model coefficients are only correlated with soil density. This is in contrast to findings from research conducted over the last two decades that show many other environmental and compositional factors, in addition to soil density, affect the magnitude of seismic compression. To overcome these shortcomings and others the model was modified and calibrated, where mathematical transformations were performed for the model to allow it to be implemented in “simplified” and “non-simplified” calculation manners. Also, systematic model modification procedures were established by means of codes written by one software called MATLAB. Both the “simplified” and “nonsimplified” calculation methods of the model were used to analyze a site in Japan impacted by an earthquake occurred in 2007, named Niigata-ken Chuetsu-oki Earthquake. The results from the analyses are in general accord with the records obtained after the earthquake and highlight the utility and versatility of the modified models.

Carbonaceous Compressions are a widespread preservational style for fossils, yet their taphonomy remains poorly understood. Previous studies focusing on the taphonomy of carbonaceous compressions have primarily looked at exceptionally preserved faunas in plane view. The precious nature of these fossils leaves destructive techniques of analysis out of the question, but these techniques are necessary if the taphonomy of carbonaceous compressions is to be deciphered. This study analyzes Neoproterozoic carbonaceous compressions from the Yangtze Gorges area in order to address this issue. Chuaria fossils from the Jiulongwan, Sixi, and Sifangtan sections of the Doushantuo Formation and Vendotaenia fossils from the Wuhe and Miaohe sections of the Denying Formation are microchemically analyzed in both plane view and cross section in order to gain a greater understanding of the makeup of carbonaceous compressions. Results confirm and elaborate on previous studies. Likely clay coats are detected on some Chuaria specimens, while they are absent on less thermally mature specimens. Evidence for sulfate reduction in association with carbonaceous compressions is found. Sulfur enrichment, rather than clay coats, is found in association with Vendotaenia fossils. These observations lead to the hypothesis that while organic remains require a very precise set of taphonomic conditions in order to be preserved as carbonaceous compressions, there may be more than one set of conditions that allow for preservation. More studies of a greater taxonomic and taphonomic range of carbonaceous compressions are needed, however, if the mechanisms which control this preservational pathway are to be fully understood. / Master of Science

Hardware Acceleration is the most sought technique in chip design to achieve better performance and power efficiency for critical functions that may be in-efficiently handled from traditional OS/software. As technology started advancing with 7nm products already in the market which can provide better power and performance consuming low area, the latency-critical functions that were handled by software traditionally now started moving as acceleration units in the chip. This thesis describes the accelerator architecture, implementation, and prototype for one of such functions namely "Locality-Aware memory compression" which is part of the "OS-controlled memory compression" scheme that has been actively deployed in today's OSes. In brief, OS-controlled memory compression is a new memory management feature that transparently, dramatically, and adaptively increases effective main memory capacity on-demand as software-level memory usage increases beyond physical memory system capacity. OS-controlled memory compression has been adopted across almost all OSes (e.g., Linux, Windows, macOS, AIX) and almost all classes of computing systems (e.g., smartphones, PCs, data centers, and cloud). The OS-controlled memory compression scheme is Locality Aware. But still under OS-controlled memory compression today, applications experience long-latency page faults when accessing compressed memory. To solve this per- performance bottle-neck, acceleration technique has been proposed to manage "Locality Aware Memory compression" within hardware thereby enabling applications to access their OS- compressed memory directly. This Accelerator is referred to as HALK throughout this work, which stands for "Hardware-accelerated Locality-aware Memory Compression". The literal mean- ing of the word HALK in English is 'a hidden place'. As such, this accelerator is neither exposed to the OS nor to the running applications. It is hidden entirely in the memory con- troller hardware and incurs minimal hardware cost. This thesis work explores developing FPGA design prototype and gives the proof of concept for the functionality of HALK by running non-trivial micro-benchmarks. This work also provides and analyses power, performance, and area of HALK for ASIC designs (at technology node of 7nm) and selected FPGA Prototype design. / Master of Science / Memory capacity has become a scarce resource across many digital computing systems spanning from smartphones to large-scale cloud systems. The slowing improvement of memory capacity per dollar further worsens this problem. To address this, almost all industry-standard OSes like Linux, Windows, macOS, etc implement Memory compression to store more data in the same space. This is handled with software in today's systems which is very inefficient and suffers long latency thus degrading the user responsiveness. Hardware is always faster in performing computations compared to software. So, a solution that is implemented in hardware with the low area and low cost is always preferred as it can provide better performance and power efficiency. In the hardware world, such modules that perform specifically targeted software functions are called accelerators. This thesis shows the work on developing such a hardware accelerator to handle ``Locality Aware Memory Compression" so as to allow the applications to directly access compressed data without OS intervention thereby improving the overall performance of the system. The proposed accelerator is locality aware which means least recently allocated uncompressed page would be picked for compression to free up more space on-demand and most recently allocated page is put into an uncompressed format.