Electrical and Electronics Commons^™

Meltpondnet: A Swin Transformer U-Net For Detection Of Melt Ponds On Arctic Sea Ice, Ivan Sudakow, Vijayan K. Asari, Ruixu Liu, Denis Demchev

Electrical and Computer Engineering Faculty Publications

High-resolution aerial photographs of Arctic region are a great source for different sea ice feature recognition, which are crucial to validate, tune, and improve climate models. Melt ponds on the surface of melting Arctic sea ice are of particular interest as they are sensitive and valuable indicators and are proxy to the processes in the Arctic climate system. Manual analysis of this remote sensing data is extremely difficult and time-consuming due to the complex shapes and unpredictable boundaries of the melt ponds, and that leads to the necessity for automatizing the processes. In this study, we propose a robust and …

Circuit Optimization Techniques For Efficient Ex-Situ Training Of Robust Memristor Based Liquid State Machine, Alex Henderson, Christopher Yakopcic, Cory Merkel, Steven Harbour, Tarek M. Taha, Hananel Hazan

Electrical and Computer Engineering Faculty Publications

Spiking neural network hardware offers a high performance, power-efficient and robust platform for the processing of complex data. Many of these systems require supervised learning, which poses a challenge when using gradient-based algorithms due to the discontinuous properties of SNNs. Memristor based hardware can offer gains in portability, power reduction, and throughput efficiency when compared to pure CMOS. This paper proposes a memristor-based spiking liquid state machine (LSM). The inherent dynamics of the LSM permit the use of supervised learning without backpropagation for weight updates. To carry out the design space evaluation of the LSM for optimal hardware performance, several …

Recent Advancements In Electrochemical Conversion Of Carbon Dioxide, Nandan Nag, Amit Kumar, Sumit Sharma, Sandeep Kumar, Amit K. Thakur

Civil & Environmental Engineering Faculty Publications

Electrochemical reduction of carbon dioxide into eco-friendly and clean products is a promising approach to eradicate pollution. Although carbon dioxide emission is inhibited by the advent of renewable sources of energy, it is present in the atmosphere and needs to be cleaned. The reduction of carbon dioxide from atmospheric gases can be accomplished by its adsorption and subsequent transportation to electrolytic chambers, where it is reduced to hydrocarbons, organic acids or carbonates. This review focuses on developing a three compartment electrochemical cell to reduce carbon dioxide used as a catholyte. Various factors affecting the electrochemical reduction of carbon dioxide and …

"Demeter" Soil Monitoring System, Ryan Matthews, Rachel Rummer, Temilolu Fayomi, Alex Fuller

Williams Honors College, Honors Research Projects

The purpose of this project is to develop a soil monitoring system that can remotely sense and relay soil conditions back to a user. The deMETER soil probe, Demeter is the Greek goddess of the harvest, is designed to aid hobbyist gardeners, small-scale farms, and nurseries to monitor their dynamic soil conditions and maximize their harvest. The probe is a self-powered system that can monitor the moisture and essential nutrients of the soil profile to determine which areas should receive water and fertilizer. This would significantly cut water and fertilizer waste. The solution will include an embedded system with sensors …

An Analysis On Adversarial Machine Learning: Methods And Applications, Ali Dabouei

Graduate Theses, Dissertations, and Problem Reports (ETD)

Deep learning has witnessed astonishing advancement in the last decade and revolutionized many fields ranging from computer vision to natural language processing. A prominent field of research that enabled such achievements is adversarial learning, investigating the behavior and functionality of a learning model in presence of an adversary. Adversarial learning consists of two major trends. The first trend analyzes the susceptibility of machine learning models to manipulation in the decision-making process and aims to improve the robustness to such manipulations. The second trend exploits adversarial games between components of the model to enhance the learning process. This dissertation aims to …

Scalable Scientific Data Management On High-Performance Computing Systems, Zhenbo Qiao

Dissertations

As high-performance computing (HPC) is being scaled up to exascale to accommodate new modeling and simulation needs, I/O has continued to be a major bottleneck in the end-to-end scientific processes. To bridge the widening gap between compute and I/O, and enable data to be more efficiently stored and analyzed, simulation outputs need to be refactored, reduced, and appropriately mapped to storage tiers. Also, a major question that the community is striving to answer is how to co-design data storage and complex physics-rich analytics in a way that the time to knowledge can be minimized in post-processing. As HPC storage systems …

Evaluating Deep-Learning Models For Debris-Covered Glacier Mapping, Zhiyuan Xie, Vijayan K. Asari, Umesh K. Haritashya

Electrical and Computer Engineering Faculty Publications

In recent decades, mountain glaciers have experienced the impact of climate change in the form of accelerated glacier retreat and other glacier-related hazards such as mass wasting and glacier lake outburst floods. Since there are wide-ranging societal consequences of glacier retreat and hazards, monitoring these glaciers as accurately and repeatedly as possible is important. However, the accurate glacier boundary, especially the debriscovered glacier (DCG) boundary, which is one of the primary inputs in many glacier analyses, remains a challenge even after many years of research using conventional remote sensing methods or machine-learning methods. The GlacierNet, a deep-learning-based approach, utilized the …

Sigesn Light-Emitting Devices: From Optical To Electrical Injection, Yiyin Zhou

Graduate Theses and Dissertations

Si photonics is a fast-developing technology that impacts many applications such as data centers, 5G, Lidar, and biological/chemical sensing. One of the merits of Si photonics is to integrate electronic and photonic components on a single chip to form a complex functional system that features compact, low-cost, high-performance, and reliability. Among all building blocks, the monolithic integration of lasers on Si encountered substantial challenges. Si and Ge, conventional epitaxial material on Si, are incompetent for light emission due to the indirect bandgap. The current solution compromises the hybrid integration of III-V lasers, which requires growing on separate smaller size substrates …

Interplay Between The Lattice And Spin Degrees Of Freedom In Magnetoelectric And Magnetic Materials, Temuujin Bayaraa

Graduate Theses and Dissertations

This dissertation contains several investigations on the cross-coupling between structural and spin degrees of freedom in multiferroic and ferrimagnetic compounds by means of first-principles calculations and ab-initio-based Monte-Carlo simulations. We start with the reviews of magnetoelectricity, ferrimagnetism, strain engineering, followed by a brief introduction to first-principles computational methods, magnetic effective Hamiltonians, and other techniques that are utilized here. The results section of the dissertation can be divided into two parts. The first half focuses on magnetoelectric effects arising from different sources, while the second half is about the ferrimagnetic nature of materials. In the first part, we examine the epitaxial …

Goes-16 Level 2 Land Surface Temperature Product-Filling For Cloud Masked Data, Guadalupe Ortega

Open Access Theses & Dissertations

The GOES-16 satellite ground segment team develops different levels of environmental products and makes them readily available to the public. The level 2 Land Surface Temperature (LST) product aims to provide a remote measurement of the Earth’s surface temperature every hour. LST is of fundamental importance to many aspects of the geosciences, for example, to study the Urban Heat Islands (UHI) effect. The product however, does not provide values for pixels where the satellite imager’s view is obstructed by clouds, as determined by a cloud mask intermediate product. The objective of this thesis is to estimate LST in the cloud …

Resampling And Super-Resolution Of Hexagonally Sampled Images Using Deep Learning, Dylan Flaute, Russell C. Hardie, Hamed Elwarfalli

Electrical and Computer Engineering Faculty Publications

Super-resolution (SR) aims to increase the resolution of imagery. Applications include security, medical imaging, and object recognition. We propose a deep learning-based SR system that takes a hexagonally sampled low-resolution image as an input and generates a rectangularly sampled SR image as an output. For training and testing, we use a realistic observation model that includes optical degradation from diffraction and sensor degradation from detector integration. Our SR approach first uses non-uniform interpolation to partially upsample the observed hexagonal imagery and convert it to a rectangular grid. We then leverage a state-of-the-art convolutional neural network (CNN) architecture designed for SR …

A Unified Framework Of Deep Learning-Based Facial Expression Recognition System For Diversified Applications, Sanoar Hossain, Saiyed Umer, Vijayan K. Asari, Ranjeet Kumar Rout

Electrical and Computer Engineering Faculty Publications

This work proposes a facial expression recognition system for a diversified field of appli- cations. The purpose of the proposed system is to predict the type of expressions in a human face region. The implementation of the proposed method is fragmented into three components. In the first component, from the given input image, a tree-structured part model has been applied that predicts some landmark points on the input image to detect facial regions. The detected face region was normalized to its fixed size and then down-sampled to its varying sizes such that the advantages, due to the effect of multi-resolution …

Colloidal Quantum Dot (Cqd) Based Mid-Wavelength Infrared Optoelectronics, Shihab Bin Hafiz

Dissertations

Colloidal quantum dot (CQD) photodetectors are a rapidly emerging technology with a potential to significantly impact today’s infrared sensing and imaging technologies. To date, CQD photodetector research is primarily focused on lead-chalcogenide semiconductor CQDs which have spectral response fundamentally limited by the bulk bandgap of the constituent material, confining their applications to near-infrared (NIR, 0.7-1.0 um) and short-wavelength infrared (SWIR, 1-2.5 um) spectral regions. The overall goal of this dissertation is to investigate a new generation of CQD materials and devices that advances the current CQD photodetector research toward the technologically important thermal infrared region of 3-5 ?m, known as …

Algebraic, Computational, And Data-Driven Methods For Control-Theoretic Analysis And Learning Of Ensemble Systems, Wei Miao

McKelvey School of Engineering Graduate Student Theses & Dissertations

In this thesis, we study a class of problems involving a population of dynamical systems under a common control signal, namely, ensemble systems, through both control-theoretic and data-driven perspectives. These problems are stemmed from the growing need to understand and manipulate large collections of dynamical systems in emerging scientific areas such as quantum control, neuroscience, and magnetic resonance imaging. We examine fundamental control-theoretic properties such as ensemble controllability of ensemble systems and ensemble reachability of ensemble states, and propose ensemble control design approaches to devise control signals that steer ensemble systems to desired profiles. We show that these control-theoretic properties …

Continuous-Time And Complex Growth Transforms For Analog Computing And Optimization, Oindrila Chatterjee

McKelvey School of Engineering Graduate Student Theses & Dissertations

Analog computing is a promising and practical candidate for solving complex computational problems involving algebraic and differential equations. At the fundamental level, an analog computing framework can be viewed as a dynamical system that evolves following fundamental physical principles, like energy minimization, to solve a computing task. Additionally, conservation laws, such as conservation of charge, energy, or mass, provide a natural way to couple and constrain spatially separated variables. Taking a cue from these observations, in this dissertation, I have explored a novel dynamical system-based computing framework that exploits naturally occurring analog conservation constraints to solve a variety of optimization …

Machine Learning For Analog/Mixed-Signal Integrated Circuit Design Automation, Weidong Cao

McKelvey School of Engineering Graduate Student Theses & Dissertations

Analog/mixed-signal (AMS) integrated circuits (ICs) play an essential role in electronic systems by processing analog signals and performing data conversion to bridge the analog physical world and our digital information world.Their ubiquitousness powers diverse applications ranging from smart devices and autonomous cars to crucial infrastructures. Despite such critical importance, conventional design strategies of AMS circuits still follow an expensive and time-consuming manual process and are unable to meet the exponentially-growing productivity demands from industry and satisfy the rapidly-changing design specifications from many emerging applications. Design automation of AMS IC is thus the key to tackling these challenges and has been …

Hardware For Quantized Mixed-Precision Deep Neural Networks, Andres Rios

Open Access Theses & Dissertations

Recently, there has been a push to perform deep learning (DL) computations on the edge rather than the cloud due to latency, network connectivity, energy consumption, and privacy issues. However, state-of-the-art deep neural networks (DNNs) require vast amounts of computational power, data, and energyâ??resources that are limited on edge devices. This limitation has brought the need to design domain-specific architectures (DSAs) that implement DL-specific hardware optimizations. Traditionally DNNs have run on 32-bit floating-point numbers; however, a body of research has shown that DNNs are surprisingly robust and do not require all 32 bits. Instead, using quantization, networks can run on …

Fast Magnetic Resonance Image Reconstruction With Deep Learning Using An Efficientnet Encoder, Tahsin Rahman

Open Access Theses & Dissertations

This thesis aims to develop an efficient, deep network based method for Magnetic Resonance Imaging (MRI) acceleration through undersampled MR image reconstruction. Deep Neural Networks, particularly Deep Convolutional Networks, have been demonstrated to be highly effective in a wide variety of computer vision tasks, including MRI reconstruction. However, modern highly efficient encoder structures, such as the EfficientNet can potentially reduce reconstruction times further while improving reconstruction quality. To that end, we have developed a multi-channel U-Net MRI reconstruction network which uses an EfficientNet encoder and a custom asymmetric. The network was trained and tested using 5x undersampled multi-channel brain MR …

Non-Invasive In-Vitro Glucose Monitoring Using Optical Sensor And Machine Learning Techniques For Diabetes Applications, Maryamsadat Shokrekhodaei

Open Access Theses & Dissertations

Diabetes is a major public health challenge affecting more than 451 million people. Physiological and experimental factors influence the accuracy of non-invasive glucose monitoring, and these need to be addressed before replacing the finger prick method with a non-invasive glucose measurement technique. Also, the suitable employment of machine learning techniques on experimental data can significantly improve the accuracy of glucose predictions.

This work includes the design, development, testing and data analysis of an optical based sensor for glucose measurements. The feasibility of non-invasive measurement of glucose within aqueous solutions that assimilate the composition of human blood plasma is investigated. The …

Socio-Technical Perspective For Electronic Tax Information System In Tanzania, Lucas Ngowi, Ellen Kalinga

Tanzania Journal of Engineering and Technology (TJET)

Socio-technical systems theory has rarely been used by system architects in setting up computing systems. However, the role of socio-technical concepts in computing, which is becoming social in nature, has made the concepts more relevant and commercial. Tax information systems are examples of such systems because they are influenced by external variables such as the political environment, technological trends, and social environment, introducing complexity in their deployment and determining the type of e-services and their delivery to a diverse group of people. It was observed that in Tanzania there is resistance, reluctance and minimal use of electronic tax system because …

Efficient, Dual-Particle Directional Detection System Using A Rotating Scatter Mask, Robert Olesen, Bryan V. Egner, Darren E. Holland, Valerie Martin, James E. Bevins, Larry W. Burggraf, Buckley E. O'Day Iii

AFIT Patents

A directional radiation detection system and an omnidirectional radiation detector. The omnidirectional radiation detector detects radiation comprising at least one of: (i) gamma rays; and (ii) neutron particles. A radiation scatter mask (RSM) of the radiation detection system includes a rotating sleeve received over the omnidirectional radiation detector and rotating about a longitudinal axis. The RSM further includes: (i) a fin extending longitudinally from one side of the rotating sleeve; and (ii) a wall extending from the rotating sleeve and spaced apart from the fin having an upper end distally positioned on the rotating sleeve spaced apart or next to …

Computational Modeling Of Black Phosphorus Terahertz Photoconductive Antennas Using Comsol Multiphysics With Experimental Comparison Against A Commercial Lt-Gaas Emitter, Jose Isaac Santos Batista

Graduate Theses and Dissertations

This thesis presents computational models of terahertz (THz) photoconductive antenna (PCA) emitter using COMSOL Multiphysics commercial package. A comparison of the computer simulated radiated THz signal against that of an experimentally measured signal of commercial reference LT-GaAs emitter is presented. The two-dimensional model (2D) aimed at calculating the photoconductivity of a black phosphorus (BP) PCA at two laser wavelengths of 780 nm and 1560 nm. The 2D model was applied to the BP PCA emitter and the LT-GaAs devices to compare their simulated performance in terms of the photocurrent and radiated THz signal pulse. The results showed better performance of …

Flying Free: A Research Overview Of Deep Learning In Drone Navigation Autonomy, Thomas Lee, Susan Mckeever, Jane Courtney

Articles

With the rise of Deep Learning approaches in computer vision applications, significant strides have been made towards vehicular autonomy. Research activity in autonomous drone navigation has increased rapidly in the past five years, and drones are moving fast towards the ultimate goal of near-complete autonomy. However, while much work in the area focuses on specific tasks in drone navigation, the contribution to the overall goal of autonomy is often not assessed, and a comprehensive overview is needed. In this work, a taxonomy of drone navigation autonomy is established by mapping the definitions of vehicular autonomy levels, as defined by the …

On-Chip Nanoscale Plasmonic Optical Modulators, Abdalrahman Mohamed Nader Abdelhamid

Theses and Dissertations

In this thesis work, techniques for downsizing Optical modulators to nanoscale for the purpose of utilization in on chip communication and sensing applications are explored. Nanoscale optical interconnects can solve the electronics speed limiting transmission lines, in addition to decrease the electronic chips heat dissipation. A major obstacle in the path of achieving this goal is to build optical modulators, which transforms data from the electrical form to the optical form, in a size comparable to the size of the electronics components, while also having low insertion loss, high extinction ratio and bandwidth. Also, lap-on-chip applications used for fast diagnostics, …

Design And Characterization Of Standard Cell Library Using Finfets, Phanindra Datta Sadhu

Master's Theses

The processors and digital circuits designed today contain billions of transistors on a small piece of silicon. As devices are becoming smaller, slimmer, faster, and more efficient, the transistors also have to keep up with the demands and needs of the daily user. Unfortunately, the CMOS technology has reached its limit and cannot be used to scale down due to the transistor's breakdown caused by short channel effects. An alternative solution to this is the FinFET transistor technology, where the gate of the transistor is a three dimensional fin that surrounds the transistor and prevents the breakdown caused by scaling …

Pilltank, Lucas Chang, Hayden Tam, Aaron Teh, Krista Round

Electrical Engineering

Imagine an elderly family member, going through their daily routine of taking their pills. They find their pill box; however, they are having trouble identifying all the pills in there. Is there a name on the tablet? Can they read what it says? Do they just trust that the medication in their box is correct? How can they properly take care of themselves if they can not even confirm that what they are taking is the right medication? To combat this issue that many face, we present PillTank.

To decrease the risk of consuming the wrong medication, PillTank identifies the …

Intelligent And Secure Fog-Aided Internet Of Drones, Jingjing Yao

Dissertations

Internet of drones (IoD), which utilize drones as Internet of Things (IoT) devices, deploys several drones in the air to collect ground information and send them to the IoD gateway for further processing. Computing tasks are usually offloaded to the cloud data center for intensive processing. However, many IoD applications require real-time processing and event response (e.g., disaster response and virtual reality applications). Hence, data processing by the remote cloud may not satisfy the strict latency requirement. Fog computing attaches fog nodes, which are equipped with computing, storage and networking resources, to IoD gateways to assume a substantial amount of …

Aerosol Vapor Synthesis Of Organic Processable Pedot Particles And Measuring Electric Conductivity Using A 3d Printed Probe Station, Yang Lu

McKelvey School of Engineering Graduate Student Theses & Dissertations

Conducting polymers are organic semiconductors characterized by conjugated backbones (alternating single-double bonds) that enable mixed ionic-electronic conductivity. Their polymeric nature, tunable band structure and reversible redox capability have demonstrated fundamental advances in the fields ranging from electrochemical energy storage, sensing, to electro/photo catalysis and neuromorphic engineering. Conjugated backbones, the origin of all the unique physical and chemical properties associated with conducting polymers, prevent their solubility due to high lattice energy which hinders processing. Current solution utilizes a long-chain polymer (PSS) as dopants to render conducting polymer water dispersible (PEDOT:PSS). Nonetheless, PSS is highly acidic and hydrophilic limiting applicability with acid-incompatible …

Direct Drive Solar Panel Control Circuit, Marcorios Bekheit

Physics

A control circuit is built for insulated solar electric cookers (ISEC). Power delivery and temperature safety are the focus. Using a maximum power point tracking (MPPT) algorithm, Arduino Nano, voltage and current sensors, and a buck converter, the solar panel’s output power was maximized for a direct load heat resistor with 3.5Ω for a range of solar intensities. Using a resistance temperature detector, a temperature sensor is built for safety shutoff.

Discrepancy-Based Analysis Of Measurement Sampling Points In Compressive Sensing, Felipe Batista Da Silva

Open Access Theses & Dissertations

Compressive sensing (CS) is a technique in signal processing that under certain conditions allows someone to reconstruct sparse signals from fewer linear measurements. A problem in CS is modeled in terms of an underdetermined linear system, whose associated matrix is previously designed. Then, it is of interest in CS to know what a good sampling defined by the sensing matrix is and how to measure it. In this work, we provided analytical proofs of properties of the metric discrepancy that allow us to propose a fast algorithm for discrepancy calculation. Such metric measures the quality of the sampling measurement points …