In addition to the previous pipelined floating-point CORDIC design, three different architectures supporting both CORDIC rotation mode and vectoring mode are proposed in this thesis. Detailed analysis and comparison of these architectures are addressed in order to choose the best architecture with minimized area cost and computation latency given the required bit accuracy. Based on the comparison, we have chosen the best architecture and implemented an IEEE single precision floating-point CORDIC processor. The mathematical analysis of the computation errors is done to minimize the bit width of the composing arithmetic components during implementation. The comparison results of different architectures also serve as a general guideline for the design of floating-point sine/cosine units. Finally, we study the application of the floating-point CORDIC to 3D graphics acceleration.

Thèse : Mécanique : Université, Faculté des sciences de Paris : 1839. / Titre provenant de l'écran-titre.

Free-floating carsharing systems are among the newest types of carsharing programs. They allow one-way rentals and have no set “homes” or docks for the carsharing vehicles; instead, users are permitted to drive the vehicles anywhere within the operating zone and leave the vehicle in a legal parking space. Compared to traditional carsharing operations, which require the user to bring the vehicle back to its assigned parking space before being able to end the rental, free-floating carsharing allows much greater spontaneity and flexibility for the user. However, it leads to additional operational challenges for the program. This dissertation provides methodologies for some of these challenges facing both free-floating and traditional carsharing programs. First, it analyzes cities with carsharing to determine what characteristics increase the likelihood of the city supporting a successful carsharing program; high overall population, small household sizes, high transit use, and high levels of government employment all make the city a likely carsharing contender. Second, in terms of membership prediction, several modeling alternatives exist. All of the options find that the operating area is of key importance, with other factors (including household size, household densities, and proportion of the population between ages 20 and 39) of varying importance depending on the modeling technique. Third, carsharing trip frequencies and mode share are of value to both carsharing and metropolitan planning organizations, and this dissertation provides innovative techniques to determine the number of trips taken and the share of total travel completed with carsharing (both free-floating and traditional). Fourth and finally, an original methodology for optimizing the vehicle allocation issue for free-floating carsharing organizations is provided. The methodology takes a user input for the total number of vehicles and returns the allocations across multiple demand periods that will maximize revenue, taking into account the cost of reallocating vehicles between demand periods. / text

No description available.

Thesis (Ph. D.)--University of Adelaide, Dept. of Applied Mathematics, 1991. / Includes bibliographical references (leaves 124-127).

Thesis (M. Phil.)--Hong Kong University of Science and Technology, 2002. / Includes bibliographical references (leaves 72). Also available in electronic version. Access restricted to campus users.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1980. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

This thesis examines the attachment of the floating charge to property in Scots law. The work is divided into two main parts. The first part focuses on how the charge interacts with property in a general sense. The second part considers attachment and its consequences in relation to the regimes for particular types of property. It is contended that the floating charge does not directly affect property prior to attachment. And even upon attachment its status as a “real right” is questionable. This is primarily because the charge is patrimonially limited by its enforcement mechanisms. It can only be enforced through a liquidator, receiver or administrator, and the powers of these parties are seemingly confined to property in the chargor’s estate. The thesis also demonstrates that ownership is a useful tool for examining what is required for a charge to attach to property. The most suitable approach is for ownership by the chargor to be both necessary and sufficient for attachment, but this is not the case under the present law, at least for certain property and transactions. Other currently prevailing views regarding the charge’s attachment are also challenged. This includes the belief that the charge attaches as if it is the relevant form of security for the property in question. Instead, it is suggested that the charge should be considered to attach as a “sui generis” fixed security. In addition, there are a range of difficulties that arise when the charge’s attachment and ranking are considered alongside the rules of transfer and security for specific property types. Uncertainty in the background law, and failure to take account of this when the charge was introduced, and subsequently, has meant that the charge may not operate effectively when, for example, property has been transferred for security purposes. All of this is explored in detail in the second part of the thesis.

TCAD simulations have been performed using SILVACO ATLAS 2D device simulator for a Zero-Capacitor Random Access Memory, a new generation memory cell which is being researched as an alternative for DRAM memory cells in order to get rid of the bulky storage capacitor. In our study we have taken into consideration, a Dual Gate - ZRAM (DGZRAM) as it helps reduce drain-induced barrier lowering and hence leakage, while having better control of the charge in the substrate, The states are written into the device using impact ionization to generate a large number of holes in the substrate, which alter the threshold voltage of the device. The effect of the gate oxide thickness and substrate body thickness are being taken into consideration to increase the change in the threshold voltage and thereby the noise margin. A DGZRAM structure with a Quantum Well introduced into the substrate via a SiGe layer was also simulated. The quantum well introduces a hole storage pocket in the substrate. Comparisons in terms of noise margin, have been made for both the devices which show that the structure with the quantum well in the substrate performs better than the bulk structure. The effect of impact ionization on the electron and hole concentrations have been shown for both the devices. Simulations have been performed taking into consideration gate electrodes with different work functions and it has been observed that while n-polysilicon has a detrimental impact in MOSFETs due to high off-state leakage current, it can be used to obtain low power memory cells. Parameters such as the quantum well doping, composition of Ge in the quantum well, channel length of the device, SiGe layer thickness and its position with respect to the top gate have been varied to obtain the optimum noise margin for the device.

No description available.