UMD Theses and Dissertations
Permanent URI for this collectionhttp://hdl.handle.net/1903/3
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a given thesis/dissertation in DRUM.
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
4 results
Search Results
Item Adaptive Play: A Place of Healing & Learning(2017) Coronado, Paula Fuenzalida; Tilghman, James; Architecture; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)For many years the disabled community has been secluded from our every day surroundings due to severe impairments and lack of adaptable environments. This is an issue that has fortunately begun to see progress in the early education systems taking place throughout the United States. In more recent years we have seen an increased involvement of school systems providing inclusion programs at the beginning stages of children’s development. Unfortunately architecture has not fully embraced this issue in order to provide spaces that are mindful of this diversely unique population of children. This thesis will explore architecture as a means to provide a space for children of all disabilities, and without, to interact and learn from one another at an early age in order to create an environment of inclusion within communities.Item Design Tools for Dynamic, Data-Driven, Stream Mining Systems(2015) Sudusinghe, Kishan Palintha; Bhattacharyya, Shuvra S; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The proliferation of sensing devices and cost- and energy-efficient embedded processors has contributed to an increasing interest in adaptive stream mining (ASM) systems. In this class of signal processing systems, knowledge is extracted from data streams in real-time as the data arrives, rather than in a store-now, process later fashion. The evolution of machine learning methods in many application areas has contributed to demands for efficient and accurate information extraction from streams of data arriving at distributed, mobile, and heterogeneous processing nodes. To enhance accuracy, and meet the stringent constraints in which they must be deployed, it is important for ASM systems to be effective in adapting knowledge extraction approaches and processing configurations based on data characteristics and operational conditions. In this thesis, we address these challenges in design and implementation of ASM systems. We develop systematic methods and supporting design tools for ASM systems that integrate (1) foundations of dataflow modeling for high level signal processing system design, and (2) the paradigm on Dynamic Data-Driven Application Systems (DDDAS). More specifically, the contributions of this thesis can be broadly categorized in to three major directions: 1. We develop a new design framework that systematically applies dataflow methodologies for high level signal processing system design, and adaptive stream mining based on dynamic topologies of classifiers. In particular, we introduce a new design environment, called the lightweight dataflow for dynamic data driven application systems environment (LiD4E). LiD4E provides formal semantics, rooted in dataflow principles, for design and implementation of a broad class of stream mining topologies. Using this novel application of dataflow methods, LiD4E facilitates the efficient and reliable mapping and adaptation of classifier topologies into implementations on embedded platforms. 2. We introduce new design methods for data-driven digital signal processing (DSP) systems that are targeted to resource- and energy-constrained embedded environments, such as unmanned areal vehicles (UAVs), mobile communication platforms, and wireless sensor networks. We develop a design and implementation framework for multi-mode, data driven embedded signal processing systems, where application modes with complementary trade-offs are selected, configured, executed, and switched dynamically, in a data-driven manner. We demonstrate the utility of our proposed new design methods on an energy-constrained, multi-mode face detection application. 3. We introduce new methods for multiobjective, system-level optimization that have been incorporated into the LiD4E design tool described previously. More specifically, we develop new methods for integrated modeling and optimization of real-time stream mining constraints, multidimensional stream mining performance (e.g., precision and recall), and energy efficiency. Using a design methodology centered on data-driven control of and coordination between alternative dataflow subsystems for stream mining (classification modes), we develop systematic methods for exploring complex, multidimensional design spaces associated with dynamic stream mining systems, and deriving sets of Pareto-optimal system configurations that can be switched among based on data characteristics and operating constraints.Item Physical properties of lamprey spinal cord regeneration: adaptive vs. maladaptive recovery(2014) Luna Lopez, Carlos; Aranda-Espinoza, Helim J.; Bioengineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Spinal cord injury (SCI) is a physical trauma that can result in paralysis and even death; to date no treatment exists that can successfully promote functional or adaptive recovery. Although humans are unable to regenerate after complete SCI, there are animal models that have been studied for their ability to regrow and reconnect their nerve fibers. From the group of animals that are capable of spinal cord regeneration, in the best studied is the lamprey (Petromyzon Marinus) it has been noted that recovery can be maladaptive. When left to recover at warm temperature (23 ⁰C) most lampreys had adaptive behavior, but at cold temperature (10 ⁰C) most lampreys showed maladaptive behavior. In this thesis we studied the physical factors that influence adaptive and maladaptive recovery in lampreys. In the first part, we analyzed axonal regeneration and blood clot formation at early time points after injury (1-2 weeks). We found that lampreys in cold temperature have a blood clot that could be blocking spinal cord regeneration. In the second part of this work, we analyzed the biomechanical and structural differences between lampreys in warm and cold temperature. We used in vivo X-ray imaging and tensile loading testing of the spinal cord and notochord structures, before and after injury. We found that lampreys at warm temperature are more favorable to create a permissive mechanical and structural environment for regeneration. Lastly, we used those lessons learned previously to enhance regeneration of maladaptive animals. We removed the blood clot at the injury site and created a time frequency analysis to measure the recovery of coordination. We found that lampreys in cold temperature with clot removal had a more adaptive recovery after injury than those without removal. In summary, by using the lamprey we were able to compare the differences between regeneration in warm and cold temperature and found the physical factors that influence maladaptive recovery. Removing one of these factors, in this case the blood clot, successfully enhanced the recovery of coordination. These results have the potential to be translated to higher animals and aid in the creation of successful treatments for SCI.Item Low Power Adaptive Circuits: An Adaptive Log Domain Filter and A Low Power Temperature Insensitive Oscillator Applied in Smart Dust Radio(2010) Zhai, Yiming; Goldsman, Neil; Electrical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)This dissertation focuses on exploring two low power adaptive circuits. One is an adaptive filter at audio frequency for system identification. The other is a temperature insensitive oscillator for low power radio frequency communication. The adaptive filter is presented with integrated learning rules for model reference estimation. The system is a first order low pass filter with two parameters: gain and cut-off frequency. It is implemented using multiple input floating gate transistors to realize online learning of system parameters. Adaptive dynamical system theory is used to derive robust control laws in a system identification task. Simulation results show that convergence is slower using simplified control laws but still occurs within milliseconds. Experimental results confirm that the estimated gain and cut-off frequency track the corresponding parameters of the reference filter. During operation, deterministic errors are introduced by mismatch within the analog circuit implementation. An analysis is presented which attributes the errors to current mirror mismatch. The harmonic distortion of the filter operating in different inversion is analyzed using EKV model numerically. The temperature insensitive oscillator is designed for a low power wireless network. The system is based on a current starved ring oscillator implemented using CMOS transistors instead of LC tank for less chip area and power consumption. The frequency variance with temperature is compensated by the temperature adaptive circuits. Experimental results show that the frequency stability from 5°C to 65°C has been improved 10 times with automatic compensation and at least 1 order less power is consumed than published competitors. This oscillator is applied in a 2.2GHz OOK transmitter and a 2.2GHz phase locked loop based FM receiver. With the increasing needs of compact antenna, possible high data rate and wide unused frequency range of short distance communication, a higher frequency phase locked loop used for BFSK receiver is explored using an LC oscillator for its capability at 20GHz. The success of frequency demodulation is demonstrated in the simulation results that the PLL can lock in 0.5μs with 35MHz lock-in range and 2MHz detection resolution. The model of a phase locked loop used for BFSK receiver is analyzed using Matlab.