Browsing by Subject "Electrical engineering"

Now showing 1 - 20 of 48

Acquisition of relative vehicle trajectories to facilitate freeway merging using DSRC based V2V communication
(2017-11) Peng, Zhiyuan
For the anticipated benefits of connected vehicle technology, Intelligent Transportation Systems Joint Program Office (ITSJPO) of the US Department of Transportation continues to emphasize the need for having Dedicated Short Range Communication (DSRC) based vehicle to vehicle (V2V) and/or vehicle to infrastructure (V2I) communication to enhance driver safety and traffic mobility. To take full advantage of connected vehicle technology in most safety applications, precise vehicle positioning information is neeeded in addition to V2V communication. Although, there are many techniques including vision or sensor based systems and differential GPS receivers, which can obtain precise absolute position of a vehicle at the expense of cost and complexity, some critical safety applications such as merge assist or lane change assist systems require only relative positions of surrounding vehicles with lane level resolution so a given vehicle can differentiate the vehicles on its own lane from the vehicles on adjacent lanes. We have adopted a simple approach to acquire accurate relative trajectories of surrounding vehicles using standard GPS receviers and DSRC based V2V communication. Using this approach, we have conducted field tests to successfully acquire relative trajectories of vehicles travelling on multiple lanes towards a merging junction with an accuracy of ±0.5m. The achieved accuracy level in relative trajectory was sufficient to differentiate vehicles travelling on adjacent lanes of a multiple-lane freeway.
Allocation policy analysis for cache coherence protocols for STT-MRAM-based caches
(2014-10) Nandkar, Pushkar Shridhar
Spintronic devices have demonstrated promising results to replace the traditional CMOS devices in Last Level Caches. Recent research have focussed on STT-CMOS hybrid caches and presented techniques to reduce leakage power and achieve performance benefit due to larger caches size that can be accommodated in the same footprint. Instead of using such hybrid caches, we use in-place STT-MRAM replacements for the complete cache hierarchy and show that we can achieve increased performance due to larger caches and significant power benefits due to decreased leakage. Further, we study different cache coherence protocols and with different allocation policies. Our preliminary results show that Non-inclusive protocols save write dynamic energy mostly due to reduced number of line fills compared to an inclusive protocol. We study the complete parsec benchmark suite and discuss the best allocation policy for each benchmark while considering the energy-delay trade off.
Analysis of the bump problem in BSIM3 using NOR gate circuit and implementation of techniques in order to overcome them
(2014-12) Sankaralingam, Subramaniam
In this paper we will be analyze the bump problem in BSIM3 using "Killer" NOR gate circuit. We refer to the NOR gate circuit as "Killer" gate because it kills our simulation results. This problem is witnessed in all the models employing quasi-static approximation. Quasi-static approximation and Non quasi-static approximation will be explained in detail to give better insight into the bump problem. Finally some techniques will be proposed and analyzed with the help of waveforms in order to overcome the problem.
Automated segmentation and pathology detection in ophthalmic images
(2014-07) Roy Chowdhury, Sohini
Computer-aided medical diagnostic system design is an emerging inter-disciplinary technology that assists medical practitioners for providing quick and accurate diagnosis and prognosis of pathology. Since manual assessments of digital medical images can be both ambiguous and time-consuming, computer-aided image analysis and evaluation systems can be beneficial for baseline diagnosis, screening and disease prioritization tasks. This thesis presents automated algorithms for detecting ophthalmic pathologies pertaining to the human retina that may lead to acquired blindness in the absence of timely diagnosis and treatment. Multi-modal automated segmentation and detection algorithms for diabetic manifestations such as Diabetic Retinopathy and Diabetic Macular Edema are presented. Also, segmentation algorithms are presented that can be useful for automated detection of Glaucoma, Macular Degeneration and Vein Occlusions. These algorithms are robust to normal and pathological images and incur low computationally complexity.First, we present a novel blood vessel segmentation algorithm using fundus images that extracts the major blood vessels by applying high-pass filtering and morphological transforms followed by addition of fine vessel pixels that are classified by a Gaussian Mixture Model (GMM) classifier. The proposed algorithm achieves more than 95% vessel segmentation accuracy on three publicly available data sets. Next, we present an iterative blood vessel segmentation algorithm that initially estimates the major blood vessels, followed by iterative addition of fine blood vessel segments till a novel stopping criterion terminates the iterative vessel addition process. This iterative algorithm is specifically robust to thresholds since it achieves 95.35% vessel segmentation accuracy with 0.9638 area under ROC curve (AUC) on abnormal retinal images from the publicly available STARE data set.We propose a novel rule-based automated optic disc (OD) segmentation algorithm that detects the OD boundary and the location of vessel origin (VO) pixel. This algorithm initially detects OD candidate regions at the intersection of the bright regions and the blood vessels in a fundus image subjected to certain structural constraints, followed by the estimation of a best fit ellipse around the convex hull that combines all the detected OD candidate regions. The centroid of the blood vessels within the segmented OD boundary is detected as the VO pixel location. The proposed algorithm results in an average of 80% overlap score on images from five public data sets.We present a novel computer-aided screening system (DREAM) that analyzes fundus images with varying illumination and fields of view, and generates a severity grade for non-proliferative diabetic retinopathy (NPDR) using machine learning. Initially, the blood vessel regions and the OD region are detected and masked as the fundus image background. Abnormal foreground regions corresponding to bright and red retinopathy lesions are then detected. A novel two-step hierarchical classification approach is proposed where the non-lesions or false positives are rejected in the first step. In the second step, the bright lesions are classified as hard exudates and cotton wool spots, and the red lesions are classified as hemorrhages and micro-aneurysms. Finally, the number of lesions detected per image is combined to generate a severity grade. The DReAM system achieves 100% sensitivity, 53.16% specificity and 0.904 AUC on a publicly available MESSIDOR data set with 1200 images. Additionally, we propose algorithms that detect post-operative laser scars and fibrosed tissues and neovascularization in fundus images. The proposed algorithm achieves 94.74% sensitivity and 92.11% specificity for screening normal images in the STARE data set from the images with proliferative diabetic retinopathy (PDR). Finally, we present a novel automated system that segments six sub-retinal thickness maps from optical coherence tomography (OCT) image stacks of healthy patients and patients with diabetic macular edema (DME). First, each image in the OCT stack is denoised using a Wiener Deconvolution algorithm that estimates the speckle noise variance using a Fourier-domain based structural error. Next, the denoised images are subjected to an iterative multi-resolution high-pass filtering algorithm that detects seven sub-retinal surfaces in six iterative steps. The thicknesses of each sub-retinal layer for all scans from a particular OCT stack are then combined to generate sub-retinal thickness maps. Using the proposed system the average inner sub-retinal layer thickness in abnormal images is estimated as 275 um (r = 0.92) with an average error of 9.3 um, while the average thickness of the outer segments in abnormal images is estimated as 57.4 um (r = 0.74) with an average error of 3.5 um. Further analysis of the thickness maps from abnormal OCT image stacks demonstrates irregular plateau regions in the inner nuclear layer (INL) and outer nuclear layer (ONL), whose area can be estimated with r = 0.99 by the proposed segmentation system.
Characterization and loss modeling of silicon carbide based power electronic converters
(2015-04) Ravi, Lakshmi
Silicon Carbide (SiC) based power Metal Oxide Semiconductor Field Effect Transistors (MOSFETs) are great candidates for high-voltage, high-frequency and high-temperature power switching applications because of their favorable material properties when compared with Silicon (Si) power MOSFETs. In this thesis, the design, characterization, and modeling of a power electronic converter based around SiC MOSFETs is investigated. The test converter circuit is designed to be general enough that it can represent a half bridge converter, a DC chopper circuit or an output phase of an inverter for flexibility in testing. A practical characterization procedure is proposed which takes a circuit-level approach, as opposed to a device-level approach, using only the actual power electronic circuit under study and no additional test circuitry. Therefore this study takes into account the inherent parasitic impedances associated with the test circuit and its influence on the SiC devices' high-speed switching behavior. The hardware setup is operated at frequencies up to 200 kHz and efficiencies up to approximately 99% were recorded.Based on the characterization data and analysis, a model is constructed using MATLAB (a mathematical modeling software) for predicting converter and gate driver losses at different load currents, DC bus voltages, and operating temperatures (for both a DC-DC synchronous buck converter and a DC-AC three phase, two-level Voltage Source Inverter). Good agreements are obtained between the model outputs and experimental results. Possible future extensions to the work are discussed.
Charge Carrier Transport and Strain in Graphene Grown on Nitrogen-Seeded Silicon Carbide
(2017-10) Torrey, Ethan R.
The interest in graphene as a possible basis for new, faster, smaller and more flexible electronics is tempered by its lack of a band-gap. In recent years, several methods by which a gap might be created have been proposed and explored. The work presented here is a part of that exploration. In this case, the specific gap-inducing mechanism under study is a method of engineered strain. Graphene can be grown on silicon carbide. By pre-treating the silicon carbide in a process that leaves small amounts of nitrogen on its surface, the subsequently grown graphene is made to wrinkle. By controlling the wrinkling, i.e. the strain in the graphene layer, it may be possible to induce a band-gap. Indeed, Angle-resolved photoemission spectroscopy and scanning tunneling spectroscopy results provide experimental support for this theory. At the same time, optical absorption measurements appear to contradict it. The primary focus of this dissertation is strain and transport measurements taken on devices fabricated from this type of graphene, with the expectation that these would aid in resolving the apparent contradiction in previous results.In the course of this work, a new tri-layer method of gate oxide deposition, using reactive electron beam deposition and plasma-assisted atomic layer deposition, was developed. Also, a method of enhanced Raman spectroscopy was developed for graphene-on-silicon-carbide devices. These methods were applied to a set of samples of graphene grown on nitrogen-seeded silicon carbide (NG) with the concentration of nitrogen varying between samples. In this dissertation, several transport characteristics are shown to exhibit a monotonic dependence upon the nitrogen concentration. These include changes in strain, broadening of the longitudinal resistivity peak, an offset between that peak and the zero-crossing of Hall conductivity, and a thermally activated n-doping mechanism, all measured with respect to an applied gate voltage. In addition, more complicated changes in temperature dependence and B-field dependence of the longitudinal resistivity are observed. These results, along with the surprising decrease in resistivity with the addition of nitrogen, are explained in the context of weak localization effects, increased transport by charge puddle-mediated tunneling, and edge states. While the presence of a band-gap could not be demonstrated conclusively in this, the first report of charge transport in this material, the results are in keeping with the presence of a band-gap short-circuited by edge states.
Control and communication for a secure and reconfigurable power distribution system.
(2011-11) Giacomoni, Anthony Michael
A major transformation is taking place throughout the electric power industry to overlay existing electric infrastructure with advanced sensing, communications, and control system technologies. This transformation to a smart grid promises to enhance system efficiency, increase system reliability, support the electrification of transportation, and provide customers with greater control over their electricity consumption. Upgrading control and communication systems for the end-to-end electric power grid, however, will present many new security challenges that must be dealt with before extensive deployment and implementation of these technologies can begin. In this dissertation, a comprehensive systems approach is taken to minimize and prevent cyber-physical disturbances to electric power distribution systems using sensing, communications, and control system technologies. To accomplish this task, an intelligent distributed secure control (IDSC) architecture is presented and validated in silico for distribution systems to provide greater adaptive protection, with the ability to proactively reconfigure, and rapidly respond to disturbances. Detailed descriptions of functionalities at each layer of the architecture as well as the whole system are provided. To compare the performance of the IDSC architecture with that of other control architectures, an original simulation methodology is developed. The simulation model integrates aspects of cyber-physical security, dynamic price and demand response, sensing, communications, intermittent distributed energy resources (DERs), and dynamic optimization and reconfiguration. Applying this comprehensive systems approach, performance results for the IEEE 123 node test feeder are simulated and analyzed. The results show the trade-offs between system reliability, operational constraints, and costs for several control architectures and optimization algorithms. Additional simulation results are also provided. In particular, the advantages of an IDSC architecture are highlighted when an intermittent DER is present on the system.
Design and implementation of a dual-mode ultrasound array driver.
(2011-11) Casper, Andrew Jacob
Ultrasound has a long history as an important medical diagnostic tool. Its non-ionizing nature and relative low cost has enabled this technology to gain widespread acceptance and use in hospitals worldwide. It's currently used to create images of many internal body structures allowing for rapid assessment by a physician. While it is in the diagnostic imaging context that ultrasound is most commonly used, it is however, not the only medical use of ultrasound. Ultrasound is also capable of non-invasively targeting organs and tissue for therapeutic benefits. These benefits can range from non-invasive drug delivery to tissue cauterization. Recent advances in piezocomposite transducer technology have allowed for a new generation of array transducers that are capable of both delivering ultrasound therapy, and imaging with the same device. These transducers, referred to as Dual-Mode Ultrasound Arrays (DMUAs), use the same elements for therapy and imaging, allowing for absolute registration between therapeutic and imaging coordinates. Therefore, realtime DMUA imaging provides unique form of feedback to the physician allowing her/him to identify and quantify the exposure to any obstacles in the path of the therapeutic beam. This feedback provides the basis for realtime resynthesis of the therapeutic beam to minimize the exposure to these obstacles while maximizing the exposure at the target. The advantages of the DMUA approach to image-guided surgery can be realized only with drivers that fully integrate the imaging and therapy functions in a seamless manner. This thesis describes the design and implementation of two real-time DMUA drivers. The first system was an enhancement of a previous design that allowed for the basic features of the DMUA system to be demonstrated. The second was a new design that allowed for a wider range of operation and the implementation of microsequencer to precisely control imaging and therapy sequences.
Design of rail-to-rail operational amplifier using XFAB 0.35µm process
(2014-08) Date, Namrata Anand
Operational amplifier is an integral part of analog circuits. With the advent of portable device technology, power consumption has become an area of concern.To reduce power consumption and improve battery life performance, supply voltages are scaled. This leads to a reduced input common mode range for an operational amplifier. A rail-to-rail amplifier composed of complementary CMOS differential pairs is employed to obtain a wider input common mode range. However, the complementary differential pairs, when operated in parallel lead to a large variation in the total transconductance thereby making the circuit highly unstable. A simple and novel technique to remedy this is to use a diode connected NMOS to shift the transition region of the NMOS differential pair. This circuit is simple and consumes less power. XFAB 0.35µm process technology is used in the design of the circuit. Cadence SPECTRE simulator is used for all the simulations.
Detection of behavioral markers using wearable wireless markers sensors
(2014-08) Min, Cheol-Hong
In this thesis, we propose new methods and systems to detect pseudo periodic behavioral patterns of people with physical and mental disabilities. We have defined the infrastructure of a non-intrusive, cost effective and user friendly system to assist patients with behavioral problems which may be due to cognitive impairments due to mental disorders such as Autism Spectrum Disorder (ASD), Traumatic Brain Injuries (TBI) or Alzheimer's disease. This dissertation starts by describing advances in sensors and monitoring systems. The system is structured to monitor motions or motor patterns by obtaining sensor data from an in-home and wearable wireless sensor system, and to give reminders and feedback to users and assist therapists and caregivers. We then present the system's applications and results in detecting and classifying the behavioral patterns in activities of daily living (ADLs). Unlike previous work which issues a reminder like an alarm clock, our system minimally intervenes with the user only when needed by detecting, classifying and monitoring the tasks. The system is flexible and can easily adapt to subject variability with minimal trainings, and the same algorithm can be used to adapt to new ADLs. To better assist cognitively impaired patients, the system detects incompletion and interrupted activities of the subject and issues a reminder/feedback in an intelligent manner. Our system uses three different sensor platforms to monitor and detect abnormal state to better assist the patients with right guidance at the right time. To achieve the goals defined above, development of signal processing methods based on Gaussian Mixture Model (GMM) and Sequential Classifier from Time Domain and Frequency Domain features are discussed. Data fusion to optimally select, combine and manage sensors from different platforms which possess various characteristics and sampling frequencies to collect data is addressed. A key contribution is the selection of a subset of sensors to be monitored and processed at any given time to reduce computation load and limit providing feedback to the patient only when needed. Sensor data fusion methods address how to combine the information obtained from selected sensors in an intelligent for analysis and classification. We explore automatic extraction of features across sensors in the time-frequency plane. We also investigate several behavior recognition strategies for comparison purposes. Algorithm to detect novel patterns is proposed. The novel pattern detection algorithm to find patterns unknown to the system at the time of training is critical as behavioral patterns change or new patterns are developed. An on-line unsupervised learning method to detect and track novel patterns by analyzing features from Higher Order Statistics is also proposed. The proposed algorithm is tested over 60 hours of data collected across 20 subjects and 4 autistic patients with classification accuracy of 94.6%. Finally, a different sensing platform was investigated to enhance the wearability and comfort level of the user for long term monitoring. We showed that using an array of stitched stretch sensors on every-day wear is feasible and demonstrated its potential for activity detection. We also showed that using a combination of different platforms to complement sensing modalities can be beneficial to improving the classification accuracy of the system. We show that the proposed combination of Gaussian Mixture Model with a sequential classifier is efficient and allows potential for real-time application of the activity detection system. This thesis establishes that despite the similarities in the activities it is possible to accurately detect and classify the specific behavioral patterns. The results are compared with the previously developed methods and show that the proposed method can detect the activities with high accuracy and also allows novel event detection to adapt to the behavioral patterns to a user.
Development and evaluation of a place and route flow using Virtuoso-GXL layout tool suite.
(2011-05) Vinod, Arvind
In this work, the automatic placement and routing tools from Cadence Virtuoso(R)-GXL layout tool suite have been used to create a flow to allow automatic placement and routing of small to medium sized blocks using schematic or verilog gate-level netlist as input, for aiding the design of testchips. The flow has also been tried on a larger design and timing results are compared with results from a commercial P&R tool-Cadence Encounter(R) Digital Implementation System.
DiCAD design methodology
(2012-05) Andren, Brian Wayne
This paper presents a method for obtaining a fine tuning range digital controlled oscillator using a digital controlled artificial dielectric. Additionally, a design methodology is also provided in order to provide step-by-step instructions for the design of fine-tuning range digital controlled oscillators using a digital controlled artificial dielectric.
Dynamics and control of Newtonian and viscoelastic fluids
(2014-09) Lieu, Binh K.
Transition to turbulence represents one of the most intriguing natural phenomena. Flows that are smooth and ordered may become complex and disordered as the flow strength increases. This process is known as transition to turbulence. In this dissertation, we develop theoretical and computational tools for analysis and control of transition and turbulence in shear flows of Newtonian, such as air and water, and complex viscoelastic fluids, such as polymers and molten plastics.Part I of the dissertation is devoted to the design and verification of sensor-free and feedback-based strategies for controlling the onset of turbulence in channel flows of Newtonian fluids. We use high fidelity simulations of the nonlinear flow dynamics to demonstrate the effectiveness of our model-based approach to flow control design.In Part II, we utilize systems theoretic tools to study transition and turbulence in channel flows of viscoelastic fluids. For flows with strong elastic forces, we demonstrate that flow fluctuations can experience significant amplification even in the absence of inertia. We use our theoretical developments to uncover the underlying physical mechanism that leads to this high amplification. For turbulent flows with polymer additives, we develop a model-based method for analyzing the influence of polymers on drag reduction. We demonstrate that our approach predicts drag reducing trends observed in full-scale numerical simulations.In Part III, we develop mathematical framework and computational tools for calculating frequency responses of spatially distributed systems. Using state-of-the-art automatic spectral collocation techniques and new integral formulation, we show that our approach yields more reliable and accurate solutions than currently available methods.
The effect of copy number variation on human phenotypes
(2012-09) Alsagabi, Majid Ibrahim
The human DNA copy number variation (DCV) has been proven to be correlated to abnormal traits and features in human beings. The genomic hybridization experiment is a powerful biological tool to measure the level of the DNA copy number in thousands or millions of genomic sites simultaneously. The experiment is subject to large amounts of noise and a high level of uncertainty about the biological meaning of its measurements.The existing methods to detect the DCV are based on the two-channel approach which consists of test and reference samples. Most of the methods are ill conditioned for large data sets because of their complexity and sophisticated approaches. Furthermore, they fall short of achieving an acceptable sensitivity or they generate large amounts of false calls. The first part of this thesis explores the existing methods and presents four new models to simplify the solution. The four models are based on Band-Pass Wavelet Transform, Uncovered Markov Model, the Uniformly Most Powerful Test, and the Maximum Likelihood Estimator. The four models achieve the highest sensitivity, lowest false alarm rate, and the least complexity of all models.The second part of the thesis presents a novel model for DCV detection using a single-channel approach. The model is based on the concept of sensor networks which can be used to analyze the DNA samples from one or two channels. The model comprises three normalization techniques to remove the non-biological bias from the measurements. Then, it estimates the true distribution of the normal measurements by isolating their distribution from the heterogeneous mixture. The complexity of calculating the probability of the average error is overcome by using the saddle-point approximation and the log-lattice design. The accuracy of the saddle-point approximation is proven for both the two-channel and the single-channel approaches in homogenous and non-homogenous environments. The analysis includes both simulated and real-world datasets and it explores the recurrent DCV in large populations using the International Hapmap Project Datasets. The end of the second part of the thesis demonstrates the stationarity of the hybridization experiment and shows its impact on reducing the complexity of the analysis.The third part of the thesis investigates patterns of the DNA copy number variations. The human genetic network is a quite complex system where hundreds, or even thousands, of DNA segments interact internally with each other directly or indirectly to control all the body's functions. A bottom-up subspace-clustering algorithm is presented to reveal the biological signature of two studied phenotypes: Autism, and the lethal castration-resistant prostate cancer.
Error correction and opportunistic scheduling protocols in random wireless networks
(2014-12) Rajanna, Amogh
Retransmission with error correction capability and opportunistic user scheduling are two of the cross layer protocols that hold promise to substantially improve the performance of wireless networks.In this thesis, we do a performance analysis of Hybrid Automatic Repeat reQuest (HARQ), a joint error correction and retransmission protocol, and downlink multiuser diversity opportunistic scheduling in both single hop and multihop wireless adhoc networks (WANETs).In the first part of the thesis, we study the performance of rateless codes employed in the physical layer of a WANET. The nodes of the WANET are modeled by a homogeneous space time Poisson point process with Rayleigh fading, constant transmission power per node and pure ALOHA as the channel access protocol. The thesis considers 2 types of receivers, an ideal matched receiver and a practical nonmatched receiver. For such a WANET, the thesis quantifies the rate density and the dynamic variations of packet transmission time by deriving an upper bound to the CCDF of the packet transmission time.The thesis presents a WANET system model in which a packet transmission spans a single coherence time and it is shown that the rate density can be upto $70\%$ of the ergodic rate density. This is good news, because the presented network does not require diversity, and transmits each message within one coherence time. Thus, the presented network nearly achieves the ERD, while requiring significantly shorter delays. From a rate density perspective, the thesis illustrates the advantage of power control in the form of channel thresholding. For both the rate density and the dynamic variations of packet transmission time, the analytical insights are supported by a very good match with the simulation results.In the second part of the thesis, we do a performance analysis of the cooperative HARQ protocol in a wireless adhoc multihop network employing spatial ALOHA. We model the nodes in such a network by a homogeneous 2-D Poisson point process. We study the tradeoff between the transport capacity submetrics inherent in the network by optimizing the transport capacity w.r.t the network design parameters, HARQ coding rate and medium access probability. Using stochastic geometirc approximations, we obtain an analytic expression for the expected progress of opportunistic routing and optimize the capacity approximation by convex optimization. By way of numerical results, we show that the network design parameters obtained by optimizing the analytic approximation of transport capacity closely follows that of Monte Carlo based exact transport capacity optimization. As a result of the analysis, we argue that the optimal HARQ coding rate and medium access probability are independent of the node density in the network.In the final part of the thesis, we do a cost-benefit analysis of multiuser diversity in single antenna broadcast channels. It is well known that the multiuser diversity can be beneficial but there is a significant cost in terms of system resources, bandwidth and power associated with acquiring instantaneous CSI. We work out a cost-benefit analysis of multiuser diversity for 2 types of CSI feedback methods, dedicated feedback and SNR dependent feedback, quantifying how many users should feedback CSI i.e the amount of available multiuser diversity that should be used from a net throughput perspective. Dedicated feedback, in which orthogonal resources are allocated to each user, has significant feedback cost and this limits the amount of available multiuser diversity that can be used. SNR dependent feedback method, in which only users with SNR above a threshold attempt to feedback, has relatively much smaller feedback cost and this allows for all of the available multiuser diversity to be used. Next, we study the effect of single user multiantenna techniques, which reduce the SNR variation, on the number of feedback users. It is seen that a broadcast channel using single user multiantenna techniques should reduce the number of feedback users with the spatial dimension.
Exploration of graded indium gallium nitride heterojunction solar cells for laterally integrated, spectrally-split solar cell arrays.
(2011-12) Krohn, Jennifer Jo
The ternary group III/group V direct bandgap semiconductor alloy system of indium gallium nitride is emerging as a material with great potential for the production of highly efficient, low cost terrestrial photovoltaic devices. Indium gallium nitride alloy is a direct bandgap semiconductor with an energy gap in the range of 0.7 eV to 3.4 eV depending on the alloy composition. Here we explore a unique photovoltaic device design based on indium gallium nitride single-junction pin graded heterojunction cells arranged laterally and illuminated by a spectrally-split solar spectrum. Mathematical models for photon absorption, charge carrier generation, total charge carrier concentrations, and charge carrier flow are derived and suitable software tools are developed, implementing these equations as well as power density and efficiency calculations to simulate the photovoltaic device operation under maximum power point conditions. Efficiencies of the individual cells in the array differ significantly with the largest bandgap pin producing the highest fill factor and energy conversion efficiency. The small bandgap n-doped layer contributes the largest amount of photogenerated electron-hole pairs, and increasing the width of this layer leads to significantly larger efficiencies. Overall, the photovoltaic device yields efficiencies competitive with existing technologies. Modifications to the design can be incorporated into the software to explore additional methods of increasing efficiencies.
Exploring a multiprocessor design space to analyze the impact of using STT-RAM in the memory hierarchy
(2014-09) Borse, Nishant Ashok
Spin-tronic memory is a promising technology and offers advantages due to its non-volatility and higher density. At the same time, based on device properties, there are trade-offs that decide the energy and performance penalty overhead. To decide these trade-offs its it imperative to understand the sensitivity of different parameters in the memory subsystem. In this work, we use a known statistical technique to analyze processor core and memory parameters for their sensitivity towards performance and energy for a Spin-tronic based memory hierarchy. We also study how does the sensitivity of processor core parameters like Re-order buffer, Load Store queue etc. vary when we replace a traditional SRAM memory with the new spin-tronic technology. Further, given a mix of different memory technologies and important processor core parameters, we use find the optimal configuration for delay, energy and area using the method of simulated annealing.
Frugal sensing and estimation over wireless networks
(2014-04) Mehanna, Omar
Spectrum sensing and channel estimation are two important examples of background tasks needed for efficient wireless network operations. Channel and spectrum state communication overheads can become a serious burden, unless appropriate sensing and estimation strategies are designed that can do the job well with very limited, judicious feedback. This thesis considers two `frugal' sensing and estimation problems in this regime: crowdsourced power spectrum sensing using a network of low-end sensors broadcasting few bits; and channel estimation and tracking for transmit beamforming in frequency-division duplex (FDD) mode.In the case of spectrum sensing, each sensor is assumed to pass the received signal through a random wideband filter, measure the average power at the output of the filter, and send out a single bit to a fusion center (FC) depending on its measurement. Exploiting linearity with respect to the autocorrelation as well as important non negativity properties in a novel linear programming (LP) formulation, it is shown that adequate power spectrum sensing is possible from few bits, even for dense spectra. The formulation can be viewed as generalizing classical nonparametric spectrum estimation to the case where the data is in the form of inequalities, rather than equalities. Taking into account fading and insufficient sample averaging considerations, a different convex maximum likelihood (ML) formulation is developed, outperforming the LP formulation when the power estimates prior to thresholding are noisy. Assuming availability of a downlink channel that the FC can use to send threshold information, active sensing strategies are developed which quickly narrow down the power spectrum estimate.For the downlink channel tracking problem, the receiver is assumed to send back to the transmitter a coarsely quantized version of the received transmitter-beamformed pilot signal, instead of sending quantized channel information as in codebook-based beamforming. A novel channel tracking approach is proposed that exploits the quantization bits in a maximum a posteriori (MAP) estimation formulation, and closed-form expressions for the channel estimation mean-squared error and the corresponding signal-to-noise ratio are derived under certain conditions.
GPU-based digital hologram reconstruction and particle detection
(2014-12) Taylor, Daniel
Digital holograms, when combined with tracer particles, can be used for examining otherwise-invisible fluid flows. These holograms can be captured with standard digital imaging equipment, however processing them to extract tracer or particle locations is computationally expensive. Exacerbating the issue is that hundreds or thousands of holograms must be reconstructed to analyze a single flow.Presented here is a hologram reconstruction and particle extraction system exploiting the massive parallelism of graphics processing units (GPUs) to reduce the time required to locate the particles in 3D space by orders of magnitude. This system requires no expensive proprietary hardware and runs on standard computers, with the only special requirement being an off-the-shelf Nvidia GPU.
Grid fault ride-through in matrix converters for adjustable speed drives
(2014-12) Orser, David
A novel ride-through approach for matrix converters in adjustable speed drives is presented, utilizing the input filter capacitors as an energy transfer mechanism to support motor flux during grid fault events. The addition of three bi-directional switches is required to isolate the input filter capacitors from the collapsed grid voltages. The additional input switches, a new ride-through vector control strategy, and the post fault reconnection logic are shown to enable ride-through of many cycle faults without the use of an additional energy storage devices. The proposed architecture is verified in theory, simulation, and hardware.The architecture is valid for both indirect and direct matrix converters, provides full bi-directional power flow, and requires no additional reactive components. Additional benefits include reduced in-rush current, reduced transient voltage overshoot at plug-in, reduced damping losses, and potential harvesting of energy from remaining active grid phases.To support the work, a review of power quality assessments is included. Through this review it is shown that the proposed architecture allows matrix converter-based adjustable speed drives to successfully operate in >95% of grid fault events.

University Digital Conservancy

Browse by Subject

Browsing by Subject "Electrical engineering"