Modern processor architectures suffer from an ever increasing gap between processor and memory performance. The current memory-register model attempts to hide this gap by a system of cache memory. Line Associative Registers(LARs) are proposed as a new system to avoid the memory gap by pre-fetching and associative updating of both instructions and data. This thesis presents a fully LAR-based architecture, targeting a previously developed instruction set architecture. This architecture features an execution pipeline supporting SWAR operations, and a memory system supporting the associative behavior of LARs and lazy writeback to memory.

The health and productive output of dairy cows can be closely correlated to individual cow feed intake. Being able to monitor feed intake on a daily basis is beneficial dairy farm management. Each cow can be addressed individually with minimal time required from those working with the animals. This is essential as time management is closely tied to resource management in a dairy operation. Anything that can save time and resources and increase profitability and herd health is a paramount advantage in dairy farming. This study examined the use of machine vision structured light illumination three-dimensional scanning of cow feed to determine the volume and weight of feed in a bin before and after feeding dairy cow. Calibration and control tests were conducted to determine the effectiveness and capability of implementing such a machine vision feed scanning system. Such a system is ideal as it does not obstruct workflow or cow feeding behavior. This is an improvement over existing systems as the system in this research study can be implemented into existing farm operations with minimal effort and costs.

The gradually widening speed disparity of between CPU and memory has become an overwhelming bottleneck for the development of Chip Multiprocessor (CMP) systems. In addition, increasing penalties caused by frequent on-chip memory accesses have raised critical challenges in delivering high memory access performance with tight power and latency budgets. To overcome the daunting memory wall and energy wall issues, this thesis focuses on proposing a new heterogeneous scratchpad memory architecture which is configured from SRAM, MRAM, and Z-RAM. Based on this architecture, we propose two algorithms, a dynamic programming and a genetic algorithm, to perform data allocation to different memory units, therefore reducing memory access cost in terms of power consumption and latency. Extensive and intensive experiments are performed to show the merits of the heterogeneous scratchpad architecture over the traditional pure memory system and the effectiveness of the proposed algorithms.

Body mounted germanium substrate solar cell arrays form the faces of many small satellite designs to provide the primary power source on orbit. High efficiency solar cells are made affordable for university satellite programs as triangular devices trimmed from wafer scale solar cells. The smaller cells allow array designs to pack tightly around antenna mounts and payload instruments, giving the board design flexibility. One objective of this work is to investigate the reliability of solar cells attached to FR-4 printed circuit boards. FR-4 circuit boards have significantly higher thermal expansion coefficients and lower thermal conductivities than germanium. This thermal expansion coefficient mismatch between the FR-4 board and the components causes concern for the power system in terms of failures seen by the solar cells. These failures are most likely to occur with a longer orbital lifetime and an extended exposure to harsh environments. This work compares various methods of attaching solar cells to printed circuit boards, using solder paste alone and with a silicone adhesive, and considering the application of these adhesives by comparing the solder joints when printed by screen versus a stencil. An environmental test plan was used to compare the survivability and performance of the solar arrays.

Structured Light Illumination (SLI) is the process where spatially varied patterns are projected onto a 3-D surface and based on the distortion by the surface topology, phase information can be calculated and a 3D model constructed. Phase Measuring Profilometry (PMP) is a particular type of SLI that requires three or more patterns temporarily multiplexed. High speed PMP attempts to scan moving objects whose motion is small so as to have little impact on the 3-D model. Given that practically all machine vision cameras and high speed cameras employ a Field Programmable Gate Array (FPGA) interface directly to the image sensors, the opportunity exists to do the processing on camera. This thesis focuses on the design, implementation, testing, and evaluation of a camera-projector system to implement a PMP dual-frequency scheme for 3-D shape measurement on a single FPGA chip. The processor architecture is implemented and tested using the Xilinx Spartan 3 FPGA chip on an Opal Kelly development board. The hardware is described using VHDL and Verilog Hardware Description Languages (HDLs).

Fault detection and location algorithms have allowed for the power industry to alter the power grid from the traditional model to becoming a smart grid. This thesis implements an already established algorithm for detecting faults, as well as an impedance-based algorithm for detecting where on the line the fault has occurred and develops a smart algorithm for future HDL conversion using Simulink. Using the algorithms, the ways in which this implementation can be used to create a smarter grid are the fundamental basis for this research. Simulink was used to create a two-bus power system, create environment variables, and then Matlab was used to program the algorithm such that it could be FPGA-implementable, where the ways in which one can retrieve the data from a power line has been theorized. This novel approach to creating a smarter grid was theorized and created such that real-world applications may be further implemented in the future.

Cadmium sulfide (CdS) and cadmium telluride (CdTe) are two leading semiconductor materials used in the fabrication of thin film solar cells of relatively high power conversion efficiency and low manufacturing cost. In this work, CdS/CdTe solar cells with a varying set of processing parameters and device designs were fabricated and characterized for comparative evaluation. Studies were undertaken to elucidate the effects of (i) each step in fabrication and (ii) parameters like thickness, sheet resistance, light absorptivity solution concentration, inert gas pressure etc. Best results were obtained when the thickness of CdS planar film for the window layer was in the range of 150 nm to 200 nm. Also, CdS nanowires were fabricated for use as the window layer in CdS-CdTe solar cells. Their materials characteristics were studied with scanning electron microscopy (SEM) and X-ray Diffraction (XRD). Spectral absorption measurements on the planar CdS films and nanowire CdS layers were performed and results compared. It was established that the nanowire CdS design was superior because its absorption of sunlight was far less than that of planar CdS film, which would lead to enhanced performance in the CdS-CdTe solar cell through higher short circuit current density and higher open circuit voltage. Diode behavior of CdS-CdTe devices on planar CdS and nanowire CdS was analyzed and compared. KEYWORDS: Thin Film Solar Cell, Nanowire, UV Absorption, Open-circuit Voltage, Close Space Sublimation

Given that data is increasing exponentially everyday, extracting and understanding the information, themes and relationships from large collections of documents is more and more important to researchers in many areas. In this paper, we present a cooperative semantic information processing system to help biomedical researchers understand and discover knowledge in large numbers of titles and abstracts from PubMed query results. Our system is based on a prevalent technique, topic modeling, which is an unsupervised machine learning approach for discovering the set of semantic themes in a large set of documents. In addition, we apply a natural language processing technique to transform the “bag-of-words” assumption of topic models to the “bag-of-important-phrases” assumption and build an interactive visualization tool using a modified, open-source, Topic Browser. In the end, we conduct two experiments to evaluate the approach. The first, evaluates whether the “bag-of-important-phrases” approach is better at identifying semantic themes than the standard “bag-of-words” approach. This is an empirical study in which human subjects evaluate the quality of the resulting topics using a standard “word intrusion test” to determine whether subjects can identify a word (or phrase) that does not belong in the topic. The second is a qualitative empirical study to evaluate how well the system helps biomedical researchers explore a set of documents to discover previously hidden semantic themes and connections. The methodology for this study has been successfully used to evaluate other knowledge-discovery tools in biomedicine.

This thesis presents the design of an attitude determination system for small satellites that automatically corrects for attitude drift. Existing attitude determination systems suffer from attitude drift due to the integration of noisy rate gyro sensors used to measure the change in attitude. This attitude drift leads to a gradual loss in attitude knowledge, as error between the estimated attitude and the actual attitude increases. In this thesis a Kalman filter is used to complete sensor fusion which combines sensor observations with a projected attitude based on the dynamics of the satellite. The system proposed in this thesis also utilizes a novel sensor called the stellar gyro to correct for the drift. The stellar gyro compares star field images taken at different times to determine orientation, and works in the presence of the sun and during eclipse. This device provides a relative attitude fix that can be used to update the attitude estimate provided by the Kalman filter, effectively compensating for drift. Simulink models are developed of the hardware and algorithms to model the effectiveness of the system. The Simulink models show that the attitude determination system is highly accurate, with steady state errors of less than 1 degree.

Systems designed to enhance intelligibility of speech in noise are difficult to evaluate quantitatively because intelligibility is subjective and often requires feedback from large populations for consistent evaluations. Attempts to quantify the evaluation have included related measures such as the Speech Intelligibility Index. These require separating speech and noise signals, which precludes its use on experimental recordings. This thesis develops a procedure using an Intelligibility Ruler (IR) for efficiently quantifying intelligibility. A calibrated Mean Opinion Score (MOS) method is also implemented in order to compare repeatability over a population of 24 subjective listeners. Results showed that subjects using the IR consistently estimated SII values of the test samples with an average standard deviation of 0.0867 between subjects on a scale from zero to one and R2=0.9421. After a calibration procedure from a subset of subjects, the MOS method yielded similar results with an average standard deviation of 0.07620 and R2=0.9181.While results suggest good repeatability of the IR method over a broad range of subjects, the calibrated MOS method is capable of producing results more closely related to actual SII values and is a simpler procedure for human subjects.