Engineering Commons^™

Applied Deep Learning: Case Studies In Computer Vision And Natural Language Processing, Md Reshad Ul Hoque

Electrical & Computer Engineering Theses & Dissertations

Deep learning has proved to be successful for many computer vision and natural language processing applications. In this dissertation, three studies have been conducted to show the efficacy of deep learning models for computer vision and natural language processing. In the first study, an efficient deep learning model was proposed for seagrass scar detection in multispectral images which produced robust, accurate scars mappings. In the second study, an arithmetic deep learning model was developed to fuse multi-spectral images collected at different times with different resolutions to generate high-resolution images for downstream tasks including change detection, object detection, and land cover …

Emotion Detection Using An Ensemble Model Trained With Physiological Signals And Inferred Arousal-Valence States, Matthew Nathanael Gray

Electrical & Computer Engineering Theses & Dissertations

Affective computing is an exciting and transformative field that is gaining in popularity among psychologists, statisticians, and computer scientists. The ability of a machine to infer human emotion and mood, i.e. affective states, has the potential to greatly improve human-machine interaction in our increasingly digital world. In this work, an ensemble model methodology for detecting human emotions across multiple subjects is outlined. The Continuously Annotated Signals of Emotion (CASE) dataset, which is a dataset of physiological signals labeled with discrete emotions from video stimuli as well as subject-reported continuous emotions, arousal and valence, from the circumplex model, is used for …

Development Of High Quantum Efficiency Strained Superlattice Spin Polarized Photocathodes Via Metal Organic Chemical Vapor Deposition, Benjamin Belfore

Electrical & Computer Engineering Theses & Dissertations

Spin polarized photocathodes are necessary to examine parity violations and other fundamental phenomena in the field of high energy physics. To create these devices, expensive and complicated growth processes are necessary. While integral to accelerator physics, spin polarized electrons could have other exciting applications in materials science and other fields of physics. In order to explore these other applications feasibly, the relative supply of spin polarized photocathodes with a high rate of both polarization and photoemission needs to be increased. One such way to increase this supply is to develop the means to grow them faster and at a larger …

Recrystallization Of Cu(In,Ga)Se2 Semiconductor Thin Films Via Metal Halides Treatment, Deewakar Poudel

Electrical & Computer Engineering Theses & Dissertations

The advancement of low-cost, highly efficient solar cell devices is a major technological challenge demanding suitable materials and fabrication processes. Polycrystalline Cu(In,Ga)Se_{2 (}CIGS) appear to be one of the most promising semiconductor in thin film photovoltaic technology due to its bandgap tunability, high absorption coefficient, and tendency to produce high efficiency solar cells. High-quality CIGS materials fabricated via a three-stage co-evaporation process can convert primary materials into devices with power conversion efficiency above 23%. Increasing the deposition rate and decreasing the deposition temperature, while maintaining high efficiency, is the major concern for the CIGS solar cells to compete …

Collaborative Robotics Strategies For Handling Non-Repetitive Micro-Drilling Tasks Characterized By Low Structural Mechanical Impedance, Xiangyu Wang

Mechanical & Aerospace Engineering Theses & Dissertations

Mechanical micro-drilling finds widespread use in diverse applications ranging from advanced manufacturing to medical surgery. This dissertation aims to develop techniques that allow programming of robots to perform effective micro-drilling tasks. Accomplishing this goal is faced with several challenges. Micro-drills suffer from frequent breakage caused from variations in drill process parameters. Micro-drilling tasks afford extremely low feed rates and almost zero tolerance for any feed rate variations. The accompanying robot programming task is made difficult as mathematical models that capture the micro-drilling process complexities and sensitive variations in micro-drill parameters are highly difficult to obtain. Therefore, an experimental approach is …

Advanced Topologies Of High Step-Up Dc-Dc Converters For Renewable Energy Applications, Ramin Rahimi

Doctoral Dissertations

"This research is focused on developing several advanced topologies of high step-up DC-DC converters to connect low-voltage renewable energy (RE) sources, such as photovoltaic (PV) panels and fuel cells (FCs), into a high-voltage DC bus in renewable energy applications. The proposed converters are based on the combinations of various voltage-boosting (VB) techniques, including interleaved and quadratic structures, switched-capacitor (SC)-based voltage multiplier (VM) cells, and magnetically coupled inductor (CI) and built-in-transformer (BIT). The proposed converters offer outstanding features, including high voltage gain with low or medium duty cycle, a small number of components, low current and voltage stresses on the components, …

Machine Learning Applications In Plant Identification, Wireless Channel Estimation, And Gain Estimation For Multi-User Software-Defined Radio, Viraj K. Gajjar

Doctoral Dissertations

"This work applies machine learning (ML) techniques to selected computer vision and digital communication problems. Machine learning algorithms can be trained to perform a specific task without explicit programming. This research applies ML to the problems of: plant identification from images of leaves, channel state information (CSI) estimation for wireless multiple-input-multiple-output (MIMO) systems, and gain estimation for a multi-user software-defined radio (SDR) application.

In the first task, two methods for plant species identification from leaf images are developed. One of the methods uses hand-crafted features extracted from leaf images to train a support vector machine classifier. The other method combines …

Comparative Study And Design Optimization Of A Dual-Mechanical-Port Electric Machine For Hybrid Electric Vehicle Applications, Hao Chen, Ayman M. El-Refaie, Yuefei Zuo, Shun Cai, Libing Cao, Christopher H. T. Lee

Electrical and Computer Engineering Faculty Research and Publications

A new dual-mechanical-port (DMP) electric machine for hybrid electric vehicle applications, particularly in the power-split continuously variable transmission systems, is proposed in this paper. In order to comprehensively and quantitatively evaluate the pros and cons of the proposed machine, a comparative study of four DMP electric machines with different topologies is conducted. These four investigated DMP electric machines include a conventional DMP machine, a DMP machine with spoke-type permanent magnets, a DMP machine with reluctance rotor, and a DMP machine with open slots which is the proposed machine in this paper. Even though these four machines have similar topologies, they …

A Computational Approach To Evaluating Curricular Alignment To The United Nations Sustainable Development Goals, Philippe Lemarchand, Mick Mckeever, Cormac Macmahon, Philip Owende

Articles

The United Nations (UN) considers universities to be key actors in the pursuit of the Sustainable Development Goals (SDGs). Yet, efforts to evaluate the embeddedness of the SDGs in university curricula tend to rely on manual analyses of curriculum documents for keywords contained in sustainability lexica, with little consideration for the diverse contexts of such keywords. The efficacy of these efforts, relying on expert co-elicitation in both subject-matter contexts and sustainability, suffers from drawbacks associated with keyword searches, such as limited coverage of key concepts, difficulty in extracting intended meaning and potential for greenwashing through “keyword stuffing.” This paper presents …

Lifetime Maximization In Underwater Wireless Communication Networks, Kazi Yasin Islam, Iftekhar Ahmad, Daryoush Habibi, Jiong Jin, Muhammad Waqas

Research outputs 2022 to 2026

The rise in demand for underwater wireless communication networks (UWCN) has been driven by the emergence of new applications including unmanned underwater vehicles, deep-sea exploration, maritime and underwater archaeology research, and diver communications. For various applications, underwater network lifetime must be long, since unlike terrestrial sensor networks, it is a major job to change/recharge node batteries in underwater environments. In this paper, we introduce a solution where critical nodes in a UWCN are periodically recharged by small renewable energy sources. Further, a caching mechanism is introduced to relieve critical nodes of heavy workload when their residual energies run low. By …

Enhanced Supervised Descent Learning Technique For Electromagnetic Inverse Scattering Problems By The Deep Convolutional Neural Networks, He Ming Yao, Rui Guo, Maokun Li, Lijun Jiang, Michael Kwok Po Ng

Electrical and Computer Engineering Faculty Research & Creative Works

This work proposes a novel deep learning (DL) framework to solve the electromagnetic inverse scattering (EMIS) problems. The proposed framework integrates the complex-valued deep convolutional neural network (DConvNet) into the supervised descent method (SDM) to realize both off-line training and on-line 'imaging' prediction for EMIS. The offline training consists of two parts: 1) DConvNet training: the training dataset is created, and the proposed DConvNet is trained to realize the EM forward process and 2) SDM training: the trained DConvNet is integrated into the SDM framework, and the average descent directions between the initial prediction and the true label of SDM …

Optimization Of High-Speed Channel For Signal Integrity With Deep Genetic Algorithm, Huan Huan Zhang, Zhao Sheng Xue, Xin Yi Liu, Ping Li, Lijun Jiang, Guang Ming Shi

Electrical and Computer Engineering Faculty Research & Creative Works

A deep genetic algorithm (GA) is proposed to optimize the high-speed channel for signal integrity. In the traditional genetic algorithm-based high-speed channel optimization method, the eye height and eye width of the eye diagram are obtained by eye diagram simulation based on the full-wave algorithm, which is computationally expensive. In this letter, a deep neural network (DNN) is trained to predict the eye diagram information corresponding to a set of given design parameters of the high-speed channel. This DNN is embedded into the genetic algorithm to carry out the evaluation operation, which can greatly accelerate the evaluation process. A high-speed …

On The Sensitivity Of Microwave Fabry-Perot Interferometers For Displacement Detection, Chen Zhu, Rex E. Gerald, Jie Huang

Electrical and Computer Engineering Faculty Research & Creative Works

Microwave Interferometers Have Been Attractive Devices for High-Precision Measurements of Displacement-Related Physical Quantities in a Variety of Scientific and Engineering Applications. However, Due to the Long Operating Wavelength, the Displacement Measurement Resolution of Microwave Interferometers is Limited to the Micrometer Scale. in This Article, We Propose and Demonstrate a Novel Technology based on Microwave Interferometry for Displacement Measurements with Potential Picometer Resolution. an Open-Ended (OE) Transmission Line Resonator based on a Homemade Hollow Coaxial Cable is Explored in Proof-Of-Concept Numerical and Experimental Investigations. the Technique Involves Two Stages of Displacement Sensitivity Amplification: One is the Open End-Based Phase Sensitivity Amplification, …

Analytical Method For Joint Optimization Of Ffe And Dfe Equalizations For Multi-Level Signals, Nana Dikhaminjia, Mikheil Tsiklauri, Zurab Kiguradze, Jiayi He, Arun Chada, Bhyrav Mutnury, James L. Drewniak

Electrical and Computer Engineering Faculty Research & Creative Works

Channel equalization is the efficient method for recovering distorted signal and correspondingly reducing bit error rate (BER). Different type of equalizations, like feed forward equalization (FFE) and decision feedback equalization (DFE) are canceling channel effect and recovering channel response. Separate optimization of tap coefficients for FFE and DFE does not give optimal result. In this case FFE and DFE tap coefficients are found separately and they are not collaborating. Therefore, the final equalization result is not global optimal. In the present paper new analytical method for finding best tap coefficients for FFE and DFE joint equalization is introduced. The proposed …

Transient Response Of Esd Protection Devices For A High-Speed I/O Interface, Jianchi Zhou, Yang Xu, Sergej Bub, Steffen Holland, Javad Soleiman Meiguni, David Pommerenke, Daryl G. Beetner

Electrical and Computer Engineering Faculty Research & Creative Works

System-efficient electrostatic discharge (ESD) design (SEED) models of a diode and transient voltage suppressor (TVS) were developed to study their transient response in a high-speed input/output interface. Previously reported SEED models were improved to strengthen their convergence stability and facilitate accurate predictions over a wide range of conditions. These improvements were required to accurately capture the race conditions between the TVS and on-chip diode, where the diode's turn on may prevent turn on of the TVS. Simulations and measurements were performed to demonstrate the impact of the ESD pulse's rise time on race conditions. During a race, results showed the …

Huber Kalman Filter For Wi-Fi Based Vehicle Driver's Respiration Detection, Yang Yang, Yunlong Luo, Alex Qi, Ge Shi, Miao Miao, Jun Fan, Jianhua Ma, Yihong Qi

Electrical and Computer Engineering Faculty Research & Creative Works

The use of breath detection in vehicles can reduce the number of vehicular accidents caused by drivers in poor physical condition. Prior studies of contactless respiration detection mainly targeted a static person. However, there are emerging applications to sense a driver, with emphasis on contactless methods. For example, being able to detect a driver's respiration while driving by using a vehicular Wi-Fi system can significantly enhance driving safety. The sensing system can be mounted on the back of the driver's seat, and it can sense the tiny chest displacement of the driver via Wi-Fi signals. The body displacement and car …

A Spice-Compatible Model To Simulate Rfi-Induced Buzz Noise Problem In A Camera, Wei Zhang, Shengxuan Xia, Xin Fang, Xu Wang, Takashi Enomoto, Hideki Shumiya, Kenji Araki, Chulsoon Hwang

Electrical and Computer Engineering Faculty Research & Creative Works

This article proposed a SPICE-compatible model to fast simulate the buzz noise problem in a camera device. Based on the reciprocity theorem, the proposed SPICE-compatible model consists of cascaded scattering (S) parameters extracted from the field coupling. It can provide an accurate estimation of the buzz noise transfer function (TFtotal) between the radio-frequency antenna and the audio system within an average error of 1 dB. Besides, the proposed model allows for coupling decomposition by separating the audio system into different regions, fast buzz noise mitigation by adding filtering circuits, and fast antenna evaluation/selection by characterizing the TFtotal with different antenna …

Experimental Evaluation Of A 63.3:1 Dual-Stage Coaxial Magnetic Gear, Hossein Baninajar, Sina Modaresahmadi, H. Y. Wong, Jonathan Bird, W. Williams, B. Dechant

Electrical and Computer Engineering Faculty Publications and Presentations

This paper presents the construction and testing results for a 63.3:1 dual-stage coaxial magnetic gear for use in a marine hydrokinetic generator demonstrator application. The dual-stage magnetic gear is composed of series-connected coaxial magnetic gear. The stage-2 magnetic gear utilizes Halbach magnet arrays on both rotors and has a 9.5:1 gear ratio. The stage-1 magnetic gear utilizes a Halbach rotor on the outer rotor and a flux concentration inner rotor. The stage-1 magnetic gear has a 6.67:1 gear ratio and at the peak torque of 1220Nm the stage-1 MG was shown to be capable of operating with a 268 N·m/L …

Fall Prediction Based On Instrumented Measures Of Gait And Turning In Daily Life In People With Multiple Sclerosis, Ishu Arpan, Vrutangkumar Shah, James Mcnames, Graham Harker, Patricia Carlson-Kuhta, Rebecca I. Spain, Mahmoud El-Gohary, Martina Mancini, Fay Horak

Electrical and Computer Engineering Faculty Publications and Presentations

This study investigates the potential of passive monitoring of gait and turning in daily life in people with multiple sclerosis (PwMS) to identify those at future risk of falls. Seven days of passive monitoring of gait and turning were carried out in a pilot study of 26 PwMS in home settings using wearable inertial sensors. The retrospective fall history was collected at the baseline. After gait and turning data collection in daily life, PwMS were followed biweekly for a year and were classified as fallers if they experienced >1 fall. The ability of short-term passive monitoring of gait and turning, …

Warp-Aware Adaptive Energy Efficiency Calibration For Multi-Gpu Systems, Zhuowei Wang, Xiaoyu Song, Lianglun Cheng, Hai Wan, Wuqing Zhao, Tao Wang

Electrical and Computer Engineering Faculty Publications and Presentations

Massive GPU acceleration processors have been used in high-performance computing systems. The Dennard-scaling has led to power and thermal constraints limiting the performance of such systems. The demand for both increased performance and energy-efficiency is highly desired. This paper presents a multi-layer low-power optimisation method for warps and tasks parallelisms. We present a dynamic frequency regulation scheme for performance parameters in terms of load balance and load imbalance. The method monitors the energy parameters in runtime and adjusts adaptively the voltage level to ensure the performance efficiency with energy reduction. The experimental results show that the multi-layer low-power optimisation with …

Optimized Three Deep Learning Models Based-Pso Hyperparameters For Beijing Pm2.5 Prediction, Andri Pranolo, Yingchi Mao, Aji Prasetya Wibawa, Agung Bella Putra Utama, Felix Andika Dwiyanto

Knowledge Engineering and Data Science

Deep learning is a machine learning approach that produces excellent performance in various applications, including natural language processing, image identification, and forecasting. Deep learning network performance depends on the hyperparameter settings. This research attempts to optimize the deep learning architecture of Long short term memory (LSTM), Convolutional neural network (CNN), and Multilayer perceptron (MLP) for forecasting tasks using Particle swarm optimization (PSO), a swarm intelligence-based metaheuristic optimization methodology: Proposed M-1 (PSO-LSTM), M-2 (PSO-CNN), and M-3 (PSO-MLP). Beijing PM2.5 datasets was analyzed to measure the performance of the proposed models. PM2.5 as a target variable was affected by dew point, pressure, …

Comparison Of Machine Learning Techniques For Activities Of Daily Living Classification With Electromyographic Data, Sergio A. Salinas, Mohamed Elgalhud, Luke Tambakis, Sanket Salunke, Kshitija Patel, Kenneth Mcisaac, Katarina Grolinger, Ana Luisa Trejos, Abdelkader Ouda, Hamada Ghenniwa

Electrical and Computer Engineering Publications

Advances in data science and wearable robotic devices present an opportunity to improve rehabilitation outcomes. Some of these devices incorporate electromyography (EMG) electrodes that sense physiological patient activity, making it possible to develop rehabilitation systems able to assess the patient’s progress when performing activities of daily living (ADLs). However, additional research is needed to improve the ability to interpret EMG signals. To address this issue, an off-line classification approach for the 26 upper-limb ADLs included in the KIN-MUS UJI dataset is presented in this paper. The ADLs were performed by 22 subjects, while seven EMG signals were recorded from their …

Designing Harvesting And Hauling Cost Models For Energy Cane Production For Biorefineries, Prabodh Illukpitiya, Firuz Yuldashev, Kabirat Nasiru

Agricultural and Environmental Sciences Faculty Research

The harvesting and hauling operations of bioenergy feedstock is an important area in biofuel production. Production costs can be minimized by maintaining optimal machinery units for these operations. The objective of this study is to design an optimal harvesting unit for bioenergy refinery and estimate harvesting and hauling costs of energy cane. A biorefinery with the annual capacity of processing twenty-five million imp. gallons of ethanol were considered. Given the efficiency of harvesting, a two-row soldier system was considered. Considering the year-round supply of energy cane to the refinery, the optimal machinery unit was designed, and the combined operation costs …

Computational Models To Detect Radiation In Urban Environments: An Application Of Signal Processing Techniques And Neural Networks To Radiation Data Analysis, Jose Nicolas Gachancipa

Beyond: Undergraduate Research Journal

Radioactive sources, such as uranium-235, are nuclides that emit ionizing radiation, and which can be used to build nuclear weapons. In public areas, the presence of a radioactive nuclide can present a risk to the population, and therefore, it is imperative that threats are identified by radiological search and response teams in a timely and effective manner. In urban environments, such as densely populated cities, radioactive sources may be more difficult to detect, since background radiation produced by surrounding objects and structures (e.g., buildings, cars) can hinder the effective detection of unnatural radioactive material. This article presents a computational model …

Failure Detection In Deep Neural Networks For Medical Imaging, Sabeen Ahmed, Dimah Dera, Saud Ul Hassan, Nidhal Bouaynaya, Ghulam Rasool

Electrical and Computer Engineering Faculty Publications

Deep neural networks (DNNs) have started to find their role in the modern healthcare system. DNNs are being developed for diagnosis, prognosis, treatment planning, and outcome prediction for various diseases. With the increasing number of applications of DNNs in modern healthcare, their trustworthiness and reliability are becoming increasingly important. An essential aspect of trustworthiness is detecting the performance degradation and failure of deployed DNNs in medical settings. The softmax output values produced by DNNs are not a calibrated measure of model confidence. Softmax probability numbers are generally higher than the actual model confidence. The model confidence-accuracy gap further increases for …

Development Of A Configurable Real-Time Event Detection Framework For Power Systems Using Swarm Intelligence Optimization, Umar Farooq

Dissertations and Theses

Modern power systems characterized by complex topologies require accurate situational awareness to maintain an adequate level of reliability. Since they are large and spread over wide geographical areas, occurrence of failures is inevitable in power systems. Various generation and transmission disturbances give rise to a mismatch between generation and demand, which manifest as frequency events. These events can take the form of negligible frequency deviations or more severe emergencies that can precipitate cascading outages, depending on the severity of the disturbance and efficacy of remedial action schema. The impacts of such events have become more critical with recent decline in …

Precision Maritime Localization And Landing With Real-Time Kinematic Gnss, Alexander Jordan, Matthew Kent Rydalch, Tim Mclain, Michael Williamson Tabango

Student Works

This paper presents a highly effective method for UAV precision shipboard localization and landing using Real-time Kinematic Global Navigation Satellite System (RTK GNSS). To assess the feasibility of RTK GNSS for this use case we explicitly exclude vision-based localization techniques which are most often presented in the literature. Instead, the methods presented in this paper use only RTK GNSS with an inertial measurement unit aboard the landing pad to estimate the state of the boat and the relative position of the UAV with respect to the boat. We use a continuous-discrete extended Kalman filter combined with a complementary filter for …

Modeling Environment For Testing A Distributed Energy Resource Management System (Derms) Using Gridapps-D Platform, Shiva Poudel, Sean Keene, Roshan Kini, Sarmad Hanif, Robert B. Bass, Jaime Kolln

Electrical and Computer Engineering Faculty Publications and Presentations

The electric power system is currently undergoing a major transition due to growing numbers of distributed energy resources (DERs) and increased distribution automation. If optimally managed and operated, DERs could provide flexibility and highly valuable grid services such as restoration, peak shaving, voltage regulation, and frequency support to maintain grid reliability. Different applications and enterprises, such as distributed energy resources management systems (DERMS), are being developed for coordinated and optimal operation of DERs. However, to attract sufficient DER participation and achieve the coordinated operation of DERs, systems and components must be interoperable and information exchange must be secure. Along this …

Interactive Planetarium Project, Eric Babcock, Cody Park, Johann Van Hilst

Discovery Day - Prescott

The Interactive Planetarium Project will design and build the software framework for connectivity between the Digistar 6 planetarium projection software and the smartphones of all audience members in the Jim and Linda Lee Planetarium. The goal of this project is to make planetarium shows more participatory, add a feature to our planetarium shows that many other universities do not yet have, and create a framework for future students and faculty to build from. To demonstrate our technology, we will make a real-time competitive trivia game able to support 60 concurrent users (number of expected audience members in the planetarium).

The …

Adaptive Balancing Compensation In Distribution Grid Of Ultra High-Power Manufacturing Plants, Motab Almousa

LSU Doctoral Dissertations

This dissertation presents a method of adaptive compensation, capable of handling power factor and the supply quality improvement under non-sinusoidal conditions. The focus in this dissertation is put on developing compensation for ultra-high power metallurgical plants, meaning on compensation in three-phase, four-wire ultra-high power dynamic, distribution systems. A separate attention in the dissertation is put on the adaptive compensation of ultrahigh power arc furnaces. The dissertation also presents an original method of adaptive compensation of DC current generated by such arc furnaces.

The proposed compensator is developed in the frame of the Currents’ Physical Components (CPC) – based power theory. …