Engineering Commons^™

Adversaguard: A Distributed Data-Poisoning Benchmark For Parallel Ai, Yulia Kumar, Solomon Thomas, Dejaun Gayle, J. Jenny Li, Dov Kruger

Center for Cybersecurity

As organizations scale model training across large clusters and clouds, data poisoning has emerged as a significant practical threat. Most existing research focuses on data poisoning in single-node environments. Far fewer studies have compared the effectiveness of attacks across parallel training strategies, where factors like gradient aggregation and distributed memory ceilings fundamentally alter attack detectability and its impacts. To address this gap, we introduce AdversaGuard, a reproducible benchmark and accompanying application designed specifically for distributed settings to protect AI training pipelines. This research makes the following key contributions: (1) A comprehensive Distributed Data Poisoning (DDP) benchmark spanning seven distributed systems …

Enhancing Non-Player Character Dialogue In Video Gages: An Evaluation Of Large Language Model-Generated Responses, Lam P. Quach

Master's Theses

As video games increasingly emphasize narrative depth and player immersion, the quality of Non-Player Character (NPC) dialogue has become crucial for creating engaging gaming experiences. This thesis investigates the potential of Large Language Models (LLMs) to generate high-quality NPC dialogue by comprehensively evaluating four state-of-the-art models: Gemma 3 27B, Mistral 7B, QWEN 2.5, and LLAMA 3.1. The study employs a mixed-methods approach, combining human evaluation (N=50 participants) with AI-based assessment across five key benchmarks: coherence, personality expression, engagement, style/tone appropriateness, and overall quality. Participants evaluated 32 dialogue samples (8 per model) generated for a fantasy game context featuring two distinct …

Improved Triboelectric Nanogenerator By As-Prepared Lithium Niobate For Energy Harvesting And Sensing Applications, Jahid Inam Chowdhury, Md. Wasikur Rahman, Md Arafat Hossain, Nicholas Dimakis, Mohammed Jasim Uddin

School of Integrative Biological & Chemical Sciences Faculty Publications

Triboelectric nanogenerators (TENGs) have garnered significant research interest due to their ability to harvest mechanical energy efficiently. In this study, we report a TENG composed of polydimethylsiloxane (PDMS) and polyvinyl alcohol (PVA) as triboelectric layers. To enhance charge generation in the PDMS composite polymer, we incorporated lithium niobate (LiNbO3) nanoparticles, leveraging their piezoelectric and ferroelectric properties. The LiNbO3 nanoparticles were synthesized using a solid-state reaction method, resulting in two distinct phases: triclinic LiNbO3 and monoclinic LiNb3O8. Various weight percentages of LiNbO3 and LiNb3O8 nanoparticles were added to the PDMS matrix to optimize power generation. The maximum open-circuit voltage (VOC) and …

Sainik Vol 1 Issue 1, Sastra Deemed To Be University

SAINIK

No abstract provided.

Enhancing Network Security: Dynamical Intrusion Detection Systems Leveraging Zero Trust Architecture, Ekramul Haque

Tennessee State University Alumni Theses and Dissertations

This thesis discussed two original methodologies for proposing security frameworks incorporating machine learning (ML) and Zero Trust Architecture (ZTA) principles to manage advanced persistent threats faced by Unmanned Aerial Vehicles (UAVs) and Network intrusion detection systems (IDS). The first methodology examined the use of RF signals and deep learning to identify and classify UAVs. The RF signal characteristics used in the method for detecting UAVs improved the ability to determine RF drone protocols. Although the models led to promising findings, their lack of ability to generalize to new drone types identified the need to improve both the data set and …

Hybrid Forecasting Of University Electricity Demand Using Time Series And Deep Learning, Minsoo Baek, Youngguk Seo

Faculty Articles

University buildings are energy-intensive and operate on complex schedules, making electricity demand forecasting particularly challenging. This study develops and evaluates monthly forecasting models for a public university campus in Georgia using six years of data (January 2019–December 2024) that integrate weather variables and academic calendar indicators. Three modeling approaches are compared: Seasonal Autoregressive Integrated Moving Average (SARIMA), SARIMA with exogenous variables (SARIMAX), and a hybrid SARIMAX–Long Short-Term Memory (LSTM) model. Feature selection methods, correlation analysis, Granger causality, Random Forest importance, Recursive Feature Elimination (RFE), and Least Absolute Shrinkage and Selection Operator (LASSO) regression, were applied to optimize input relevance. The …

The Floodnet Community Engagement Guide, Véronëque Ignace, Sofia Mariyamis, Kendra Krueger, Polly Pierone, Hayley Elszasz, Hannah Eisler Burnett

The Science and Resilience Institute at Jamaica Bay, SRIJB

"At the Intersection of Science, Policy, and Community: The FloodNet NYC Community Engagement Strategy” is a public-facing community engagement guide that documents the strategies, tools, and lessons developed through FloodNet NYC, a cross-sector partnership among researchers at NYU and CUNY and New York City agencies. Designed as a practical resource, this guide shares our approach to community engagement and dissemination so that researchers, practitioners, community organizations, and public agencies can adapt these methods to their own urban climate science projects and other community-centered efforts that address climate challenges.

Grounded in community-based participatory research (CBPR) principles, the guide presents community engagement …

Multi-Modal Depth Estimation Using Camera And Mmwave Sensors, Alston D. Devero-Belfon

Student Theses

For accurately estimating the depth of environments with varying lighting conditions, reliable methods are limited. By utilizing wireless sensor technology in conjunction with cameras, a wide range of environments can be visualized, and objects within these environments can be tracked and monitored. Such methods offer cost-effective alternatives and provide a more secure, data-at-rest option for individuals with low vision, while also enhancing machine perception. In this work, we develop such a prototype that utilizes wireless sensors and cameras, which act in sync, enabling us to estimate the depth of objects within varying lighting environments to a level that is recognizable …

Analysis Of Traffic Crash Data In Kentucky 2020-2024, Paul Ross, Eric Green, Christopher Blackden, Christopher Van Dyke

Kentucky Transportation Center Research Report

No abstract provided.

Ibi-Dt: A Novel Approach Combining Individualized Bayesian Inference And Decision Tree For Identifying Cancer Drivers And Their Interactions, Md Asad Rahman, Gregory F. Cooper, Jinying Zhao, Xinghua Lu, Jinling Liu

Engineering Management and Systems Engineering Faculty Research & Creative Works

Cancer is mainly caused by a relatively small portion of somatic genome alterations (SGAs), called cancer drivers. Despite success in identifying a good number of cancer drivers, many more remain to be discovered to explain various cancers. Moreover, limited tools are available to identify potential interactions among cancer drivers for a better understanding of oncogenesis. To tackle these challenges, we have developed a novel approach called individualized Bayesian inference using a decision tree (IBI-DT). IBI-DT recognizes the genetic heterogeneity among cancer patients, where different individuals or patient subgroups of distinct genomic makeup may have different drivers. IBI-DT works by constructing …

Lung Injury Risk Curves From Behind Armor Blunt Trauma Using A Live Swine Model, Narayan Yoganandan, Lewis Somberg, Danielle Wilson, Alok Shah, Jared Michael Koser, Brian D. Stemper, Valeta Carol Chancey, Joseph Mcentire

Biomedical Engineering Faculty Research and Publications

Introduction

From structural, anatomical, and functional perspectives, components of the thoracoabdominal region are heterogeneous and physiologically and functionally different. Although their tolerances to injury are expected to be different, the current Roma Plastilina No. 1 clay penetration criterion for behind armor blunt trauma (BABT) is not specific to the body region. It is important to develop regional injury criteria to ensure its specificity. The objective of the study is to conduct impact tests on the lung region using a live animal model and develop injury risk curves using velocity and deflection metrics via parametric survival analysis.

Materials and Methods

Live …

Sparse-Data Orbit Estimation In Low Earth Orbit Using The Markov Chain Monte Carlo Ensemble Gaussian Mixture Filter, Nicholas J. Oden

Master's Theses

The number of space objects (SOs) in low Earth orbit (LEO) continues to increase rapidly, creating challenges for the current ground-based tracking network, which cannot accommodate the projected growth in SOs. Catalog maintenance relies on frequent observations for reliable reacquisition, with Two-Line Element (TLE) sets typically generated daily to mitigate rapid error growth from poor TLE accuracy. This constraint limits the ability to track more objects with existing infrastructure. This work evaluates the Markov Chain Monte Carlo Ensemble Gaussian Mixture Filter (MCMC EnGMF), a nonlinear, non-Gaussian filter well-suited for sparse tracking scenarios where higher post-update accuracy is needed to reduce …

Low-Resource Ecoacoustic Audio Classification, Enis Berk Coban

Dissertations, Theses, and Capstone Projects

Ecoacoustic monitoring via machine learning enables scalable analysis but is often constrained by labeled data scarcity, particularly in remote regions like the Arctic. This thesis confronts low-resource ecoacoustic audio classification by developing and evaluating complementary machine learning methodologies. We introduce EDANSA, the first publicly available, expert- labeled Arctic dataset of its kind, curated via novel active learning, alongside a baseline CNN. We systematically evaluate transfer learning, showing general audio embeddings effectively bootstrap classifiers for challenging Arctic sounds, significantly outperforming direct label mapping. Optimizing label utility, we investigate standard data augmentation and introduce novel audio data valuation via Shapley values, revealing …

Toward A Generalizable Perceptual Hashing Framework For Image Manipulation Detection, Priyanka Samanta

Dissertations, Theses, and Capstone Projects

This thesis contributes to research in adversarial image manipulation detection. The primary motivation is the increasing need to verify digital images, especially for legal evidence, journalistic proof, or social media content—where manipulated or fabricated images can mislead, defame, or distort reality. A key application and contribution of this work is the development of eWitness, a blockchain application that generates and registers image provenance at capture time to enable independent verification of authenticity. The secret sauce behind the system is SmartHash, a novel and efficient perceptual hashing algorithm designed for real-world deployment in systems like eWitness. Unlike existing algorithms, SmartHash targets …

Storage Location Optimization In Automated Storage And Retrieval Systems: A Deep Reinforcement Learning Approach, Lingjun Wang, Aldy Gunawan, Pieter Vansteenwegen

Research Collection School Of Computing and Information Systems

This study investigates the optimization of storage location in automated storage and retrieval systems (AS/RS). We introduce an optimization approach based on the Deep Q-Network (DQN) algorithm to enhance warehouse task efficiency and minimize stacker travel during storage and retrieval. To accelerate the algorithm training process, we integrate a prioritized experience replay mechanism. Furthermore, we decouple action selection from value estimation within the DQN framework to address the issue of value overestimation. The proposed model is evaluated against three heuristic methods. The experimental results demonstrate that our approach significantly outperforms these baselines.

Mitigating Melanin-Induced Bias In Pulse Oximetry: Optical, Algorithmic, Engineering, Hardware And Modeling Tools, Mckenzie Bradley, Sydnee Barrett, Ty Mckelvey, Jeremiah Carpenter, Delphine Dean

Publications

Melanin, the primary determinant of skin pigmentation, absorbs light at wavelengths that can have significant impact on the accuracy of pulse oximetry and other optical biosensing methods. This narrative review examines key factors influencing melanin-dependent pulse oximetry inaccuracies, including optical interference in transmission and reflectance modes. These inaccuracies further highlight the need for use of standardized skin tone metrics in device testing and design such as the Monk Skin Tone scale and Individual Typology Angle for performance stratification. There are several approaches in development that hope to address the errors in pulse oximetry measurements on melanin-rich skin. These include algorithmic …

Automation Of Lcmc Logistics Processes: A Delphi Approach, Kline M. Alt

Theses and Dissertations

As the U.S. Air Force confronts growing complexity in system acquisition, the implementation of digital models in system design and logistics process management allows the incorporation of digital tools and the possibility for automation of portions of logistics processes. This thesis investigates where these technologies can be most effectively integrated within the Air Force Life Cycle Management Center logistics enterprise (AFLCMC). Using a three round Delphi study, AFLCMC logistics subject matter expert (SME) opinions were solicited from program-level senior logisticians, program managers to identify high-need areas, key success factors, and potential barriers to adoption. Quantitative consensus from Likert-scale and ordinal …

A Review Of The United States' Long Term War Support Capabilities In The Indo-Pacific Command Region, Brian J. Mullin Jr.

Theses and Dissertations

This study examines U.S. maritime transportation readiness in the Indo-Pacific, highlighting fleet age, mariner shortages, shipyard decline, and port vulnerabilities. It also considers contested logistics and technological threats. Recommendations include fleet recapitalization, mariner pipeline growth, port diversification, and defensive upgrades. The study concludes that secure sea line assumptions are outdated and calls for greater resilience, with follow-on efficiency analysis proposed for ports and ships.

Implementing Lean Principles To Enhance Warehouse Operations At King Abdulaziz Air Base(Kaab): A Case Study Of Royal Saudi Air Force (Rsaf), Saleh A. Alghamdi

Theses and Dissertations

The Royal Saudi Air Force (RSAF) relies on efficient logistics to sustain readiness. At King Abdulaziz Air Base, warehouse receiving inefficiencies caused delays and waste. This study used Lean principles and a six-month time–motion analysis, with Pareto and Fishbone tools, to identify 55% waste in dead pile and 75% in palletized shipments. Standard times of 5.98 and 6.55 minutes were set. Key recommendations include SOPs, cross-training, forklift certification, layout redesign, and RFID. Lean adoption could save 100+ labor hours and $4,000 annually, improving safety, accuracy, and mission readiness.

Distributed Coherent Beamforming At 60 Ghz Enabled By Optically-Established Coherence, Drake Silbernagel, Yu Rong, Isabella Lenz, Prithvi Hemanth, Carl Morgenstern, Owen Ma, Nolan Matthews, Nadar Zaki, Kyle W. Martin, John D. Elgin, Jacob Holtom, Daniel W. Bliss, Kimberly Frey

Space Dynamics Laboratory Publications

We implement and experimentally demonstrate a 60 GHz distributed system leveraging an optical time synchronization system that provides precise time and frequency alignment between independent elements of the distributed mesh. Utilizing such accurate coherence, we perform receive beamforming with interference rejection and transmit nulling. In these configurations, the system achieves a coherent gain over an incoherent network of N nodes, significantly improving the relevant signal power ratios. Our system demonstrates extended array phase coherence times, enabling advanced techniques. Results from over-the-air experiments demonstrate a 14.3 dB signal-to-interference-plus-noise improvement in interference-laden scenarios with a contributing 13.5 dB null towards interference in …

Mesospheric Gravity Waves Observed By Nasa Atmospheric Waves Experiment (Awe), Yucheng Zhao, Jiarong Zhang, Pierre-Dominique Pautet, Jun Ma, Joe Mcinerney, Hanli Liu, Steve Eckermann, Ludger Scherliess, Michael John Taylor, Burt Lamborn, Russ Kirkham

Space Dynamics Laboratory Publications

Although smaller scale gravity waves (GWs) (horizontal wavelengths ~30–300 km) are thought to account for the dominant energy and momentum inputs at the Ionosphere-Thermosphere-Mesosphere (ITM) altitudes, the sources, variability, and influences of these smaller-scale GWs are still major unknowns. The NASA Atmospheric Waves Experiment (AWE) is designed to measure these GWs in the mesopause region. In November 2023, AWE was successfully launched and deployed on the International Space Station (ISS) and science data collection was started. The AWE instrument maps the nighttime hydroxyl (OH) layer (~87 km), providing 2D GW fields in mesospheric temperature and OH band intensity over a …

Orographic Wave Activity At Mesospheric Altitude Over The Southern Ocean Islands Observed By The Atmospheric Waves Experiment, P.-D. Pautet, S. D. Eckermann, L. Scherliess, J. Ma, Y. Zhao

Space Dynamics Laboratory Publications

A major source for the gravity waves (GW) measured in the stratosphere and higher is the orographic forcing produced by the tropospheric wind blowing over mountainous regions. Recent studies have shown that even small islands can generate waves which, under appropriate conditions, are able to penetrate above 80 km altitude, transporting large amount of momentum into the upper atmosphere. Investigating the exact effects of those islands is challenging because ground-based and even airborne measurements are limited above those generally isolated places.

The Utah State University (USU) Atmospheric Waves Experiment (AWE) was designed and built to study mesospheric GW globally, even …

Comparative Analysis Of Different Magnetic Anomaly Datasets Using Navigation Performance With Flight Test Data, Aaron P. Nielsen, Brandon M. Blakely, Patrick Duff

Faculty Publications

Magnetic Anomaly Navigation (MagNav) is a map-based method of navigation which relies on accurately obtaining the anomaly field to a high level of precision to achieve good navigation results. This requires utilizing high quality sensors, accurately modeling disturbance fields from the aircraft & other sources, and creating high-fidelity maps. Aeromagnetic survey data or marine track survey data must be processed into a product that can be used as a reference for a magnetic navigator and a variety of techniques can be utilized for this processing. The data collection for different survey types reflects choices typically made to study the underlying …

Optimizing Object Detection For Remote Monitoring With Owl Ducklink Radios, Colin Ren-Jin Babian

Electrical Engineering

This project develops an AI powered object detection system integrated with OWL DuckLink radios to enable real time remote monitoring. The system benchmarks multiple AI accelerators using a Raspberry Pi 5 to determine their efficiency in low power, long range communication environments. By evaluating the Raspberry Pi 5 CPU (7.62W, 14.5 camera FPS, 3.8 inference FPS), Raspberry Pi AI Camera with Sony IMX500 (6.45W, 30 camera FPS, 9.2 inference FPS), and Raspberry Pi AI Hat with Hailo-8L chipset (7.80W, 30 camera FPS, >30 inference FPS), this project identifies the most effective combination of hardware and software for edge AI applications. …

Restoration Of 6-Axis Industrial-Grade Robotic Arms, Bryce Andrew Paulson, Warren Douglas Howard

Electrical Engineering

The Staubli TX60 and TX60CR are industrial-grade, six-axis robotic arms originally designed to perform complex automation tasks in manufacturing environments. Following their retirement from active industrial use, both units were restored to full operational status. As part of this project, they were reprogrammed to execute a pick-and-place routine using kids shape-sorting toys that mirrors typical workflows found in factory settings. The six degrees of freedom afforded by each robot enable a wide range of motion, making them well-suited for tasks such as material handling, product sorting, and repetitive operations typically carried out by human workers. Each model is equipped with …

A Biocompatible Nitinol Based Triboelectric Stent Sensor For Prospective Cardiovascular Health Monitoring, Ulises Vidaurri Romero, Sk Shamim Hasan Abir, Najlah Karam, Mariana Torres, Shahria Ahmed, Md. Wasikur Rahman, Bahareh Azimi, Serena Danti, Jianzhi Li, M. Jasim Uddin

Mechanical Engineering Faculty Publications

Triboelectric nanogenerators (TENGs) have been considered as an effective approach for self-powered systems. Currently, coronary heart disease remains the leading cause of death in the United States. This can be easily resolved by balloon angioplasty or a specialized mesh tube called a stent. This study demonstrates a stent sensor made of nitinol, a nickel—titanium alloy used in the medical field for its pseudo-elasticity and strong corrosion resistance, poly(vinylidene fluoride) (PVDF) and polydimethylsiloxane (PDMS), which can measure several physiological parameters while placing it in the arteries. This nitinol health monitor sensor (NHMS) device thus integrates the TENG with a specific medical …

Design Of Robust Adaptive Nonlinear Backstepping Controller Enhanced By Deep Deterministic Policy Gradient Algorithm For Efficient Power Converter Regulation, Seyyed Morteza Ghamari, Asma Aziz, Mehrdad Ghahramani

Research outputs 2022 to 2026

Power converters play an important role in incorporating renewable energy sources into power systems. Among different converter designs, Buck and Boost converters are popular, as they use fewer components and deliver cost savings and high efficiency. However, Boost converters are known as non–minimum phase systems, imposing harder constraints for designing a robust converter. Developing an efficient controller for these topologies can be difficult since they exhibit nonlinearity and distortion in high frequency modes. The Lyapunov-based Adaptive Backstepping Control (ABSC) technology is used to regulate suitable outputs for these structures. This approach is an updated version of the technique that uses …

Renewable-Based Isolated Power Systems: A Review Of Scalability, Reliability, And Uncertainty Modeling, Mehrdad Ghahramani, Daryoush Habibi, Seyyedmorteza Ghamari, Hamid Soleimani, Asma Aziz

Research outputs 2022 to 2026

Electric power systems are increasingly becoming more decentralized. Many communities depend on isolated power systems that operate independently of the main grid. Remote, islanded, and isolated systems face challenges due to the intermittency and unpredictability of renewable energy sources. This paper reviews the current status of renewable integration and control in stand-alone power systems. It examines techniques to enhance system reliability through energy storage, hybrid systems, and advanced predictive models. Additionally, the issues related to connecting stand-alone systems, focusing on reliability and renewable penetration, are discussed. The scalability of stand-alone power systems is analyzed based on classifications of small-, medium-, …

Effect Of Short-Chain Polymer Binders On The Mechanical And Electrochemical Performance Of Silicon Anodes, Fei Sun, L. Zurita-Garcia, Dean R. Wheeler

Faculty Publications

Polymer binders are crucial components in providing both mechanical support and chemical stability to the structure of porous Li-ion electrodes. Particularly in silicon anodes, the active material undergoes substantial volume expansion of up to 275%. Due to the mechanical constraint of the current collector, these silicon materials tend to expand in the normal direction while exhibiting substantial particle rearrangement and plastic deformation. Conventional rigid binders such as polyacrylic acid (PAA) and polyimide (PI), while providing satisfactory initial capacity, do not eliminate diminished long-term performance. Our research attempts to develop binder formulations that can accommodate sufficient flexibility for the substantial volume …

Wind Turbine Rotor Design Using High-Fidelity Aerostructural Optimization, Marco Mangano, Sicheng He, Yingqian Liao, Denis-Gabriel Caprace, Andrew Ning, Joaquim R. R. A. Martins

Faculty Publications

Large wind turbines yield more energy but demand careful aeroelastic blade design. Coupled multiphysics design strategies can reduce wind energy costs exploiting fluid-structure interactions. This work presents the first high-fidelity aerostructural optimization study of a large wind turbine rotor.We use blade-resolved fluid dynamics and structural solvers in a monolithic gradient-based optimization framework to explore steady-state torque and blade mass trade-offs. The coupled-adjoint approach computes gradients efficiently, enabling the optimization of over 100 structural and geometric parameters simultaneously. Our optimization study modifies a DTU 10 MW benchmark with a simplified structure and isotropic material properties. The tightly coupled optimizations increase torque …