Engineering Commons^™

Mosaic: A Prune-And-Assemble Approach For Efficient Model Pruning In Privacy-Preserving Deep Learning, Yifei Cai, Qiao Zhang, Rui Ning, Chunsheng Xin, Hongyi Wu

Computer Science Faculty Publications

To enable common users to capitalize on the power of deep learning, Machine Learning as a Service (MLaaS) has been proposed in the literature, which opens powerful deep learning models of service providers to the public. To protect the data privacy of end users, as well as the model privacy of the server, several state-of-the-art privacy-preserving MLaaS frameworks have also been proposed. Nevertheless, despite the exquisite design of these frameworks to enhance computation efficiency, the computational cost remains expensive for practical applications. To improve the computation efficiency of deep learning (DL) models, model pruning has been adopted as a strategic …

Speculative Anisotropic Mesh Adaptation On Shared Memory For Cfd Applications, Christos Tsolakis, Nikos Chrisochoides

Computer Science Faculty Publications

Efficient and robust anisotropic mesh adaptation is crucial for Computational Fluid Dynamics (CFD) simulations. The CFD Vision 2030 Study highlights the pressing need for this technology, particularly for simulations targeting supercomputers. This work applies a fine-grained speculative approach to anisotropic mesh operations. Our implementation exhibits more than 90% parallel efficiency on a multi-core node. Additionally, we evaluate our method within an adaptive pipeline for a spectrum of publicly available test-cases that includes both analytically derived and error-based fields. For all test-cases, our results are in accordance with published results in the literature. Support for CAD-based data is introduced, and its …

The Transformative Integration Of Artificial Intelligence With Cmmc And Nist 800-171 For Advanced Risk Management And Compliance, Mia Lunati

Cybersecurity Undergraduate Research Showcase

This paper explores the transformative potential of integrating Artificial Intelligence (AI) with established cybersecurity frameworks such as the Cybersecurity Maturity Model Certification (CMMC) and the National Institute of Standards and Technology (NIST) Special Publication 800-171. The thesis argues that the relationship between AI and these frameworks has the capacity to transform risk management in cybersecurity, where it could serve as a critical element in threat mitigation. In addition to addressing AI’s capabilities, this paper acknowledges the risks and limitations of these systems, highlighting the need for extensive research and monitoring when relying on AI. One must understand boundaries when integrating …

Integrating Ai Into Uavs, Huong Quach

Cybersecurity Undergraduate Research Showcase

This research project explores the application of Deep Learning (DL) techniques, specifically Convolutional Neural Networks (CNNs), to develop a smoke detection algorithm for deployment on mobile platforms, such as drones and self-driving vehicles. The project focuses on enhancing the decision-making capabilities of these platforms in emergency response situations. The methodology involves three phases: algorithm development, algorithm implementation, and testing and optimization. The developed CNN model, based on ResNet50 architecture, is trained on a dataset of fire, smoke, and neutral images obtained from the web. The algorithm is implemented on the Jetson Nano platform to provide responsive support for first responders. …

Potential Security Vulnerabilities In Raspberry Pi Devices With Mitigation Strategies, Briana Tolleson

Cybersecurity Undergraduate Research Showcase

For this research project I used a Raspberry Pi device and conducted online research to investigate potential security vulnerabilities along with mitigation strategies. I configured the Raspberry Pi by using the proper peripherals such as an HDMI cord, a microUSB adapter that provided 5V and at least 700mA of current, a TV monitor, PiSwitch, SD Card, keyboard, and mouse. I installed the Rasbian operating system (OS). The process to install the Rasbian took about 10 minutes to boot starting at 21:08 on 10/27/2023 and ending at 21:18. 1,513 megabytes (MB) was written to the SD card running at (2.5 MB/sec). …

Knowing Just Enough To Be Dangerous: The Sociological Effects Of Censoring Public Ai, David Hopkins

Cybersecurity Undergraduate Research Showcase

This paper will present the capabilities and security concerns of public AI, also called generative AI, and look at the societal and sociological effects of implementing regulations of this technology.

Framing Automation Trust: How Initial Information About Automated Driving Systems Influences Swift Trust In Automation And Trust Repair For Human Automation Collaboration, Scott Anthony Mishler

Psychology Theses & Dissertations

The study examines how trust in automation is influenced by initial framing of information before interaction and how later active calibration methods can further influence trust repair or dampening after an automation error in a three-experiment study. As more human drivers begin to use automated driving systems (ADSs) for the first time, their initial understanding of the system can influence their trust leading to a miscalibration of trust. Prior studies have investigated how trust develops through interactions with an automated system, but few have looked at integrating swift trust and framing to calibrate trust before interaction and investigate further active …

Framework For Implementing Advanced Radar Plotting Aid Capability For Small Maritime Vessels, Jason Stark Harris

Electrical & Computer Engineering Theses & Dissertations

Every year in the United States many people are killed or injured when maritime vessels collide with other vessels or fixed objects. According to the United States Coast Guard, the top contributing factors to these collisions are operator inattention, operator inexperience and an improper lookout. Larger commercial vessels are required to have RADAR systems which support Automatic RADAR Plotting Aid (ARPA) which can automatically detect collisions and alert an operator to change course. These systems can be very expensive which put them out of reach of the average recreational boater. It is however possible to implement a low cost ARPA …

Machine Learning Approach To Activity Categorization In Young Adults Using Biomechanical Metrics, Nathan Q. C. Holland

Mechanical & Aerospace Engineering Theses & Dissertations

Inactive adults often have decreased musculoskeletal health and increased risk factors for chronic diseases. However, there is limited data linking biomechanical measurements of generally healthy young adults to their physical activity levels assessed through questionnaires. Commonly used data collection methods in biomechanics for assessing musculoskeletal health include but are not limited to muscle quality (measured as echo intensity when using ultrasound), isokinetic (i.e., dynamic) muscle strength, muscle activations, and functional movement assessments using motion capture systems. These assessments can be time consuming for both data collection and processing. Therefore, understanding if all biomechanical assessments are necessary to classify the activity …

Lithium Extraction From Aqueous Solution Using Magnesium Doped Lithium Ion-Sieve Composite, Ujjwal Pokharel

Civil & Environmental Engineering Theses & Dissertations

This study focused on the development of a magnesium doped ion-sieve composite to extract lithium from an aqueous solution using adsorption process. The extraction process will provide a sustainable alternative to the currently practiced solar evaporation/concentration method which is very slow (takes 24 months) and water intensive. Currently, most of the lithium in the world comes from the mining of lithium or from the evaporative extraction process from the brine. Additionally, the Salton Sea area geothermal brines are recognized as potentially important domestic sources of lithium. Lithium concentration in geothermal brines from the Salton Sea area is reported to be …

An Effective Model For Dynamic Properties Of Local Soils And Their Influence On Seismic Response Of A Typical Reinforced Concrete Building, Kaveh Zehtab

Civil & Environmental Engineering Theses & Dissertations

This dissertation presents an effective model to determine the dynamic properties of local soils and their impact on the seismic response of a typical mid-height reinforced concrete building. The results of the study suggest that the use of global models and empirical relationships for the dynamic properties of soils may not consider the effects of unique stress conditions on the cyclic response of the foundation soil. The study also investigates the seismic response of a site in central Adapazarı, Turkey, where four- to seven-story buildings suffered significant damage during the 1999 Kocaeli earthquake. A new model is presented that combines …

Fabrication Of Smooth Sac305 Thin Films Via Magnetron Sputtering And Evaluations Of Microstructure, Creep, And Electrical Resistivity, Manish Ojha

Mechanical & Aerospace Engineering Theses & Dissertations

SAC305 (96.5%Sn-3%Ag-0.5%Cu) is the leading alternative to the traditional Sn-Pb solder eutectic alloy owing to its low melting temperature, better compatibility with other components, and excellent mechanical/structural properties. In the realm of modern electronics, where devices are increasingly miniaturized, the design and characterization of thin solder joints become paramount. The orientation and size of the grains within the solder can influence its ability to withstand mechanical stresses. However, research on SAC thin films remains sparse, and these films present unique challenges and characteristics compared to their bulk counterparts, influenced by factors like interfaces, stresses, thickness, microstructure, and the nature of …

Faster, Cheaper, And Better Cfd: A Case For Machine Learning To Augment Reynolds-Averaged Navier-Stokes, John Peter Romano Ii

Mechanical & Aerospace Engineering Theses & Dissertations

In recent years, the field of machine learning (ML) has made significant advances, particularly through applying deep learning (DL) algorithms and artificial intelligence (AI). The literature shows several ways that ML may enhance the power of computational fluid dynamics (CFD) to improve its solution accuracy, reduce the needed computational resources and reduce overall simulation cost. ML techniques have also expanded the understanding of underlying flow physics and improved data capture from experimental fluid dynamics.

This dissertation presents an in-depth literature review and discusses ways the field of fluid dynamics has leveraged ML modeling to date. The author selects and describes …

Studies Of Flowfields And Dynamic Stability Characteristics Of A Quadrotor, Engin Baris

Mechanical & Aerospace Engineering Theses & Dissertations

Electric multirotor air vehicles have become a pervasive technology and research topic in industry, academia, and daily life, and small quadrotors are one of the most preferred designs in the multirotor marketplace. However, the configuration of the quadrotors makes aerodynamic interaction effects one of the key factors of the vehicle performance in both hover and non-axial forward flight conditions.

In the present work, aerodynamic characteristics of the cross-configured small quadrotor in hover, edgewise, and maneuvering flight modes were investigated in detail by performing static and dynamic wind tunnel tests at various RPM levels, wind speeds, pitch and yaw angles, and …

Cheiloscopy Examination And Classification Of Lip Prints With And Without Parafunctional Oral Habits: A Cross-Sectional Observation Study, Emily Smith Regan

Dental Hygiene Theses & Dissertations

Problem: Lip prints are unique and have potential for use as a human identifier. The purpose of this study was to observe possible cheiloscopy differences of individuals with and without parafunctional oral habits. Additionally, inter-rater reliability (IRR) of lip print examiners was observed. Methods: This IRB approved blinded cross-sectional observational study collected lip prints from sixty-six individuals using lipstick and adhesive tape to transfer lip prints to white bond paper for viewing purposes. Each set of included lip prints was divided into quadrants and dichotomized as those with or without an oral parafunctional habit. Each quadrant sample was manually analyzed …

Pulsed Electric Field Ablation: Mechanisms Of Differential Cell Sensitivity And Methods To Mitigate Neuromuscular Excitation, Emily Gudvangen

Biomedical Engineering Theses & Dissertations

Ablation therapies aim to destroy tumors while maintaining minimal damage to surrounding healthy tissue. Pulsed electric field (PEF) ablation targets the cell membrane during the destruction of tissues, which consequently spares extracellular structures such as ducts and blood vessels, leaving the framework for tissue regeneration. However, the utility of PEF is limited by neuromuscular excitation, causing pain and involuntary muscle contraction. To identify PEF protocols that efficiently ablate tissue while minimizing neuromuscular stimulation, various PEF protocols differing in pulse duration, shape, and repetition rate were tested. The ratio of cell death and nerve excitation thresholds, which determines the range of …

Deep-Learning-Based Classification Of Digitally Modulated Signals, John A. Snoap

Electrical & Computer Engineering Theses & Dissertations

This dissertation presents several novel deep-learning (DL)-based approaches for classifying digitally modulated signals, one method of which involves the use of capsule networks (CAPs) together with cyclic cumulant (CC) features of the signals. These were blindly estimated using cyclostationary signal processing (CSP) and were then input into the CAP for training and classification. The classification performance and the generalization abilities of the proposed approach were tested using two distinct datasets that contained the same types of digitally modulated signals but had distinct generation parameters. The results showed that the classification of digitally modulated signals using CAPs and CCs proposed in …

Optics Studies For Multipass Energy Recovery At Cebaf: Er@Cebaf, Isurumali Neththikumara

Physics Theses & Dissertations

Energy recovery linacs (ERLs), focus on recycling the kinetic energy of electron beam for the purpose of accelerating a newly injected beam within the same accelerating structure. The rising developments in the super conducting radio frequency technology, ERL technology has achieved several noteworthy milestones over the past few decades. In year 2003, Jefferson Lab has successfully demonstrated a single pass energy recovery at the CEBAF accelerator. Furthermore, they conducted successful experiments with IR-FEL demo and upgrades, as well as the UV FEL driver. This multi-pass, multi-GeV range energy recovery demonstration proposed to be carried out at CEBAF accelerator at Jefferson …

Investigation Of Nanosecond Pulsed Electric Fields (Nspef) Induced Anti-Cancer Mechanism And Enhanced B16f10 Melanoma Cancer Treatment, Kamal Asadipour

Biomedical Engineering Theses & Dissertations

The use of nanosecond pulsed electric fields (nsPEF) has emerged as a promising area of research with vast implications across various scientific disciplines. The ability to generate ultra-short, high-voltage electric pulses has paved the way for numerous applications, ranging from fundamental investigations of biological phenomena to the development of innovative medical therapies. The aim of this thesis is to highlight the importance of nsPEF in two critical areas: 1) Understanding the impact of subtle postpulse waveforms through a comprehensive analysis of two common pulse generators and 2) using this knowledge to advance melanoma treatment by enhancing the therapeutic effect of …

The Effect Of Through Thickness Reinforcement Angle On The Disbonding Behavior In Skin-Stringer Configuration, Christopher John Morris

Mechanical & Aerospace Engineering Theses & Dissertations

Post-cure through thickness reinforcement is a method used to increase the mechanical properties of composite laminates in the transverse direction. This study conducted a test on skin-stringer structures bonded together in three configurations using an epoxy or thermoplastic adhesive at the interface with reinforcing pins inserted through the laminate thickness located at the edge of the stringer at differing angles between -30º and 30º. The fabrication of these samples in configurations B and C consisted of the use of carbon fiber prepeg laminate at a ply orientation of [0₂90₂]_2s for the skin and [0 90] …

Switching Methods For Three-Dimensional Rotational Dynamics Using Modified Rodrigues Parameters, Matthew Jarrett Banks

Mechanical & Aerospace Engineering Theses & Dissertations

A rigid body in space has three degrees of rotational freedom. As a result, a minimum of three independent parameters is required to define the three-dimensional orientation of a rigid body. As is well known, every set of three independent parameters has at least one orientation where mathematical or geometrical singularities are encountered; therefore, when the use of a three-parameter representation is desired, a method for singularity avoidance must also be considered. A common practice for singularity avoidance is to switch between parameter sets whose singularities occur at different orientations. With this in mind, modified Rodrigues parameters (MRP) are considered …

Towards Intelligent Runtime Framework For Distributed Heterogeneous Systems, Polykarpos Thomadakis

Computer Science Theses & Dissertations

Scientific applications strive for increased memory and computing performance, requiring massive amounts of data and time to produce results. Applications utilize large-scale, parallel computing platforms with advanced architectures to accommodate their needs. However, developing performance-portable applications for modern, heterogeneous platforms requires lots of effort and expertise in both the application and systems domains. This is more relevant for unstructured applications whose workflow is not statically predictable due to their heavily data-dependent nature. One possible solution for this problem is the introduction of an intelligent Domain-Specific Language (iDSL) that transparently helps to maintain correctness, hides the idiosyncrasies of lowlevel hardware, and …

Reinforcing Digital Trust For Cloud Manufacturing Through Data Provenance Using Ethereum Smart Contracts, Trupti Narayan Rane

Engineering Management & Systems Engineering Theses & Dissertations

Cloud Manufacturing(CMfg) is an advanced manufacturing model that caters to fast-paced agile requirements (Putnik, 2012). For manufacturing complex products that require extensive resources, manufacturers explore advanced manufacturing techniques like CMfg as it becomes infeasible to achieve high standards through complete ownership of manufacturing artifacts (Kuan et al., 2011). CMfg, with other names such as Manufacturing as a Service (MaaS) and Cyber Manufacturing (NSF, 2020), addresses the shortcoming of traditional manufacturing by building a virtual cyber enterprise of geographically distributed entities that manufacture custom products through collaboration.

With manufacturing venturing into cyberspace, Digital Trust issues concerning product quality, data, and intellectual …

Development, Experimental Validation, And Progressive Failure Modeling Of An Ultra-Thin High Stiffness Deployable Composite Boom For In-Space Applications, Jimesh D. Bhagatji

Mechanical & Aerospace Engineering Theses & Dissertations

To maximize the capabilities of nano- and micro-class satellites, which are limited by their size, weight, and power, advancements in deployable mechanisms with a high deployable surface area to packaging volume ratio are necessary. Without progress in understanding the mechanics of high-strain materials and structures, the development of compact deployable mechanisms for this class of satellites would be difficult. This research focuses on fabrication, experimental testing, and progressive failure modelling to study the deformation of an ultra-thin composite beam. The research study examines deformation modes of a boom under repetitive pure bending loads using 4-point bending setup. The material and …

Design And Implementation Of A Launching Method For Free To Oscillate Dynamic Stability Testing, Kristen M. Carey

Mechanical & Aerospace Engineering Theses & Dissertations

Magnetic Suspension and Balance Systems (MSBS) allow for static, forced oscillation and free to oscillate dynamic stability testing in a wind tunnel without the need for a physical support. The objectives of study are to assist in the application of the free to oscillate testing method in an MSBS to determine dynamic stability characteristics for various re-entry capsule designs.

This thesis discusses the development and testing of a launching method called the grabber for use in the MSBS Subsonic Wind Tunnel at NASA Langley Research Center. Aerodynamic tests were run to support the use of this method and compare the …

Attention And Task Engagement During Automated Driving, James Richard Unverricht

Psychology Theses & Dissertations

Many young drivers suffer fatal crashes each year in the United States at a rate approximately three times greater than more experienced drivers. Automated driving systems may serve to mitigate young drivers high crash rates but remain underexplored in research. This dissertation project examined the effects of levels of automation and interestingness of auditory clips on latent hazard anticipation in young drivers during simulated driving. Participants drove a vehicle at varying levels of vehicle automation (SAE Level 0, 2, or 3) in simulated scenarios, each containing a latent hazard event during which a boring, neutral, or interesting auditory clip was …

A Comparative Study Of Vinti-Based Orbit Propagation And Estimation For Cubesats In Very Low Earth Orbits, Ethan Michael Senecal

Mechanical & Aerospace Engineering Theses & Dissertations

In recent years, there has been a growing interest in CubeSats and very low Earth orbit (VLEO) space missions. Mission SeaLion, a collaborative CubeSat mission between Old Dominion University, the U.S. Coast Guard Academy, and U.S. Air Force Institute of Technology, planned to launch a 3U CubeSat into VLEO. The VLEO mission is a particularly challenging environment for navigation and orbit propagation because drag introduces a significant perturbation for orbit models such as SGP4. Additionally, mission requirements left no capacity for attitude determination or control, further reducing knowledge of drag behavior of the satellite in flight. This deficiency is a …

Towards A Robust Defense: A Multifaceted Approach To The Detection And Mitigation Of Neural Backdoor Attacks Through Feature Space Exploration And Analysis, Liuwan Zhu

Electrical & Computer Engineering Theses & Dissertations

From voice assistants to self-driving vehicles, machine learning(ML), especially deep learning, revolutionizes the way we work and live, through the wide adoption in a broad range of applications. Unfortunately, this widespread use makes deep learning-based systems a desirable target for cyberattacks, such as generating adversarial examples to fool a deep learning system to make wrong decisions. In particular, many recent studies have revealed that attackers can corrupt the training of a deep learning model, e.g., through data poisoning, or distribute a deep learning model they created with “backdoors” planted, e.g., distributed as part of a software library, so that the …

Chemical And Physical Interaction Mechanisms And Multifunctional Properties Of Plant Based Graphene In Carbon Fiber Epoxy Composites, Daniel W. Mulqueen

Mechanical & Aerospace Engineering Theses & Dissertations

Graphene has generated substantial interest as a filler due to its exceptional strength, flexibility, and conductivity but faces obstacles in supply and implementation. A renewable, plant-based graphene nanoparticle (pGNP) presents a more accessible and sustainable filler with the same properties as mineral graphenes. In this study, the mechanisms of graphene reinforcement in carbon fiber reinforced plastic (CFRP) were examined, along with the resulting improvements to mechanical strength, resistance to crack propagation, electrical and thermal conductivity at elevated temperatures. pGNP, produced from renewable biomass, was shown to have a graphitic structure with flakes 3-10 layers thick and a median lateral size …

Assessing The Prevalence And Archival Rate Of Uris To Git Hosting Platforms In Scholarly Publications, Emily Escamilla

Computer Science Theses & Dissertations

The definition of scholarly content has expanded to include the data and source code that contribute to a publication. While major archiving efforts to preserve conventional scholarly content, typically in PDFs (e.g., LOCKSS, CLOCKSS, Portico), are underway, no analogous effort has yet emerged to preserve the data and code referenced in those PDFs, particularly the scholarly code hosted online on Git Hosting Platforms (GHPs). Similarly, Software Heritage is working to archive public source code, but there is value in archiving the surrounding ephemera that provide important context to the code while maintaining their original URIs. In current implementations, source code …