Engineering Commons^™

Ai-Based Dynamic Spectrum Allocation Model For Wireless Network Management, Sai Sashank Peddibhotla

Master's Projects

The growth of wireless communication has introduced challenges in the dynamic and resource contrived space which is the efficient utilization of bandwidth and spectrum. This research presents a model for dynamic spectrum allocation with the help of Convolutional Neural Network (CNN) for feature extraction and the Deep Q-Network (DQN) model’s reinforcement learning architecture. The CNN captures both spatial and temporal features of the network states and gives them to the DQN for optimal allocation decision making. This CNN-DQN architecture effectively implements spectrum resource allocation in wireless networks and adapts to resource allocation changes within performance bounds. The system’s performance is …

Improving Contextual Retrieval For Long Documents In Q & A Systems, Sakshi Garg

Master's Projects

With the vast amount of information available on the internet distributed across several lengthy documents, finding relevant information has become more important and challenging. The goal of this project is to develop advanced techniques to retrieve information from long texts in order to deliver accurate and relevant results while ensuring speed and efficiency. As part of this work, we employ techniques to address unique difficulties posed by large and complex documents. This paper presents a custom Retrieval-Augmented Generation (RAG) framework designed to improve contextual retrieval in long and multi-document settings. In this paper, we employ several techniques like summarization, semantic …

Evorgcn: Harnessing Esm-2 Evolutionary Embeddings With Relational Gcns For High-Fidelity Protein-Protein Interaction Prediction, Mohit Kunder

Master's Projects

Accurately predicting protein-protein interactions (PPIs) is essential for understanding cellular function and advancing biomedical discovery. We model PPIs as graphs, where nodes represent proteins and edges denote interactions. Using interaction data from the STRING database, we use two samples of it, namely the benchmark datasets—SH27K and SH148K—filtered by confidence score and annotated by interaction mode (multiple relations). In this project, we present EvoRGCN, a graph-based machine learning framework for PPI prediction that integrates both sequence-level (ESM-2 embeddings) and network-level information. We incorporate various Graph Neural Network architectures, including Graph Convolutional Networks, Graph Attention Networks, and Relational Graph Convolutional Networks. Our …

Synthetic Malware Generation Using Generative Ai, Phanidhar Sai Sravan Chandana

Master's Projects

Malware grows in numbers and complexity, evading conventional signature-and anomaly-based defenses and worsening extreme data sparsity and class imbalance problems for machine learning based detection. Generative models, specifically GANs conditioned on contextual embeddings like BERT have proved effective augmenting training corpora to improve classifier accuracy, but these approaches have largely produced family-specific samples In this paper, we propose a generalized augmentation scheme for generating robust malware embeddings for various families. We begin by extracting opcode sequences from 13 malware families and encoding them into three embedding methods: CountVectorizer, TF-IDF, and BERT’s ‘[CLS]‘ vectors. We therefore train standard GANs and Wasserstein …

Comparative Analysis Of Embedding Techniques With Clustering Algorithms For Malware Opcodes, Ayush Koul

Master's Projects

Malware detection and classification remain critical challenges in cybersecurity, especially as malicious software becomes increasingly sophisticated and prevalent. While much of the work involving embeddings has traditionally relied on supervised learning approaches, there is significant potential in leveraging unsupervised learning techniques to discern hidden structures in malware data. By employing embedding techniques to convert malware samples into high-dimensional vector representations, we can capture the subtle and complex patterns inherent in malicious code without relying on pre-labeled data. This unsupervised approach helps categorize malware into predefined malware families, greatly aiding in developing cybersecurity solutions. In contrast to traditional supervised models that …

Gen Ai For Malicious Network Data, Aneesh Maturu

Master's Projects

Though botnet attacks are on the rise, they also have become sophisticated and difficult to detect. Such a rising threat demands more and more sophisticated cybersecurity that leverages machine learning technology. Nevertheless, one of the biggest bottlenecks remains the unavailability of large and well-balanced datasets, particularly for malicious traffic, which hampers the efficacy of detection models. In an attempt to address this issue, our research utilizes Generative Adversarial Networks (GANs) to produce synthetic samples of botnet traffic from the CTU-13 dataset. While the majority of generative models have been targeting image data, we use GANs for a new application: generating …

Digital Modeling Of Temperature-Dependent Processes In Industrial Wastewater: Advancing Ultimate Oxygen Demand Prediction For Treatment Optimization, Fatima Iqbal

College of Graduate Studies: Theses & Dissertations

The pulp and paper industry is the third-largest consumer of freshwater globally and faces mounting pressure to optimize water usage and minimize environmental impact. Aerated stabilization basins are extensively utilized in the pulp and paper industry and play a crucial role in treating wastewater from these operations. The existing management practices fail to effectively optimize treatment processes due to the prolonged time required for testing key quality parameters. The utilization of models developed for treatment facilities presents a viable solution; however, existing open-source models often fall short in accurately predicting treatment efficiency across varying conditions and lack comprehensive accounting of …

Pulsed Arc Additive Manufacturing Of A Functionally Graded Er2209 Duplex Stainless Steel And Hsla-100 Structure: Morphology, Characterization, And Mechanical Performance, Stevens G. Hill Jr

College of Graduate Studies: Theses & Dissertations

Wire arc additive manufacturing is a process well suited to the efficient production of large structures. Duplex stainless steel exhibits high corrosion resistance and good strength which can be highly beneficial for industrial use. However, its use is limited due to its cost and complexity in controlling microstructure to achieve desired properties. In many cases, it can be highly beneficial to manufacture a component which uses specialty steel grades such as duplex stainless steel only where necessary, and utilizes more affordable, commonly available steels for reinforcement or bulk structural support. A functionally graded material satisfies these requirements by providing a …

An Investigative Study Of Thermal Fluid Properties When Introducing Metallic Nanoparticles Through A Heat Exchanging System, Levi P. Mckinney

College of Graduate Studies: Theses & Dissertations

Heat-exchanging systems are essential in applications ranging from automobiles to air-conditioning units, and ongoing improvements aim to enhance efficiency while reducing system size. Conventional coolant upgrades often rely on ethylene glycol, which increases thermal stability and lowers freezing point but reduces water’s inherent heat-transfer capability. Advances in nanotechnology provide an alternative approach: suspending nanoparticles in base fluids can significantly modify thermal properties, with some formulations exhibiting conductivity increases of up to 60%. This thesis examines the thermal and tribological performance of three working fluids—distilled water, a 50:50 water–ethylene glycol mixture, and the same mixture enhanced with Al₂O₃ nanoparticles. Distilled water …

Evaluating The Effectiveness Of Lidar And Photogrammetry For Inspecting Mse Walls In Bridges, Shakil Ahmed

College of Graduate Studies: Theses & Dissertations

Mechanically Stabilized Earth (MSE) retaining walls are vital components of transportation infrastructure, providing structural support for embankments and bridge approaches. As many walls approach or exceed their design life, concerns regarding long-term stability, settlement, and displacement have increased. Traditional visual inspection remains the most common assessment method; however, it is subjective and often unable to detect subtle geometric changes that may precede structural distress or failure. This study evaluates Terrestrial LiDAR (Light Detection and Ranging) Scanning (TLS) and Close-Range Photogrammetry (CRP) as nondestructive, noncontact alternatives for detecting displacement in MSE retaining walls. Three representative sites were investigated: a Control Wall …

Metal-Organic Thin Film Coated On Optical Fiber, Nahideh Salehifar

Doctoral Dissertations

This dissertation explores the development of metal-organic framework (MOF)-based optical fiber sensors for detecting volatile organic compounds (VOCs) at low concentrations (parts-per-billion to parts-per-million). In the first part of the study, theoretical calculations were performed using effective medium approximation (EMA) models, including Lorentz–Lorentz, Maxwell–Garnett, and Bruggeman equations, to predict the refractive index changes of MOFs upon gas adsorption. These models were applied to MOFs such as ZIF-7, ZIF-8, ZIF-90, MIL-101(Cr), and HKUST-1 to evaluate their potential for gas sensing.

In the second part of the dissertation, experimental work was conducted to validate the theoretical predictions. MOF-coated optical fibers were fabricated …

Evaluating And Relaxing The Limits On Flexural Reinforcement Ratio Of Masonry Shear Walls, Tousif Mahmood

Doctoral Dissertations

Reinforced masonry shear walls (RMSWs), essential for lateral and out-of-plane load-resisting systems in modern construction are constrained by TMS 402/602 code limits on reinforcement ratios ("ρ" _"max" ) and axial compressive stresses (≤10% of masonry compressive strength, f_m^'), undermining masonry’s inherent compression capacity under high axial loads. This dissertation investigates the seismic performance of reinforced masonry shear walls (RMSWs) subjected to high axial compressive stresses (10–20% of f_m^'), with a focus on walls violating the maximum reinforcement ratio ("ρ" _"max") and axial load limits of TMS 402-22. Through experimental testing of 30 large-scale fully grouted (FG) and partially grouted (PG) …

Experimental Study Of Defects In Coaxial Wire-Based Laser Metal Deposition, Remy Mathenia

Doctoral Dissertations

This work seeks to improve the usability and capability of coaxial wire-based laser metal deposition (LMD) through the experimental study of process parameters on output geometry, directional effects, and defect formation. Wire-based LMD is a directed energy deposition (DED) strategy that uses a focused laser heat source to melt and fuse metal wire as it is deposited. This process is used to build parts layer-by-layer until a desired geometry is accomplished. LMD enables the creation of complex components at a high build rate with low material and energy waste. This work focuses on the deposition of titanium wire in the …

Measurement Method And Applications Of Transfer Function In Rf Desensitization Problem, Xiangrui Su

Doctoral Dissertations

Radio frequency (RF) desensitization issues comprise two components: noise radiation sources and the transfer function from noise sources to the victim antenna. RFI is a critical challenge in modern electronic systems, particularly in densely packed environments. This work presents a comprehensive study of RFI, addressing key aspects through three novel contributions. First, a transfer function measurement method is developed for compact metallic enclosures. This method provides a precise characterization of the electromagnetic (EM) environment within confined spaces, enabling accurate identification of interference pathways. Second, an EM emission management analysis framework is proposed, leveraging transfer functions to quantify and mitigate interference …

The Structure, Properties And Dissolution Behaviors Of Phosphate Glasses, Han Zhang

Doctoral Dissertations

The poor chemical durability remains a critical challenge for the application of phosphate glasses. This study investigates the compositional influences on the structure, properties and chemical durability of Li2O-ZnO-P₂O₅ glasses. Their structural characteristics were analyzed utilizing high-performance liquid chromatography, Raman spectroscopy, and X-ray photoelectron spectroscopy. The incorporation of (Li2O+ZnO) in LZeq glasses depolymerizes the phosphate network. In LZ40P and LZ45P glasses, Li⁺ initially replaces Zn²⁺ associated with non-bridging oxygens (NBOs) in Q² tetrahedra. Once the substitution in Q² is complete, further Li⁺ incorporation leads to the replacement of Zn²⁺ in Q^{1 …}

Design Of Real-Time And Energy-Efficient Driver Assist Systems Using Electroencephalogram And Neuromorphic Computing, Nathan Alan Lutes

Doctoral Dissertations

Despite the technological breakthroughs in advanced driver assist systems, distracted driving persists as a major challenge to roadway safety. This investigation advances the body of knowledge towards a solution by developing an individualized driver-state detection method using electroencephalogram (EEG) and neuromorphic computing to provide a less invasive and more energy efficient ADAS solution. It furthermore explores the changes in brain functional connectivity under distracted conditions to better understand brain state information that could be used for neuro-feedback intervention systems. The first contribution introduces the concept of using Convolutional Spiking Neural Networks (CSNNs) for recognition of patterns with movement-intention predictive power …

Novel Nano-Formulations Based On Flash Nanoprecipitation For Ocular Drug Delivery, Lin Qi

Doctoral Dissertations

Effective ocular drug delivery has always been a challenge for clinical and research. Although many nanotechnology-based formulations have been developed and showed great potential in ocular drug delivery, this is far from enough to meet clinical requirements. These studies aimed at the design and synthesis of novel nanoparticles using multi-inlet vortex mixer (MIVM) to improve ocular drug delivery efficiency. The first project involves packaging two different anti-glaucoma drugs, brimonidine (BM) and betaxolol (BX), into solid drug nanoparticles (SDNs) by MIVM, which can achieve both functions, reducing aqueous humor production and promoting aqueous humor efflux. The studies demonstrate that the SDNs …

Higher-Order Statistics And Normalized Decay Analysis For Detector Deadtime Characterization, Abdallah Wazzan

Doctoral Dissertations

Detector deadtime limits radiation measurement accuracy at high count rates, yet current methods rely on idealized paralyzable or non-paralyzable models. Real detectors exhibit hybrid behavior requiring advanced characterization approaches. This study explores deadtime characterization using two Monte Carlo simulation approaches: higher-order statistical analysis of inter-arrival times and simplified deadtime correction with hybrid models.

Using MATLAB (PULSE-WIZ), we analyzed full decay curves spanning ~4.5 half-lives of Cobalt-60 and Vanadium-52, examining coefficient of variation (CV), skewness, and kurtosis of inter-arrival times, plus normalized decay curves with Full Width at Half Maximum (FWHM) analysis. Hybrid models incorporated paralyzable and non-paralyzable components with dead …

Development, Characterization And Testing Of Traditonal And Advanced Nuclear Fuel Cladding Materials, Joshua Eddy Rittenhouse

Doctoral Dissertations

Kanthal D and FeCrAl alloys in general, are prospective candidates as accident tolerant nuclear fuel cladding materials. The work presented herein focuses on applying two techniques of severe plastic deformation, equal channel angular pressing (ECAP) and high-pressure torsion (HPT), as means of grain refinement to improve irradiation resistance. Samples of as-received, ECAP, and HPT processed Kanthal D were exposed to neutron irradiation to a dose of 2 DPA at two different temperatures, 300 °C and 500 °C. Detailed characterization was performed including mechanical and microstructural, and several positive improvements with regards to irradiation resistance were identified in the ECAP and …

Enhanced Optimization Of Mass Transfer For Carbon Capture And Wastewater Remediation In Algal Systems Through Algorithmic And Bioprocessing Techniques, Peter Ofuje Obidi

Doctoral Dissertations

The scalability and industrial deployment of algal cultivation systems are limited by suboptimal mass transfer, constraining their effectiveness in carbon capture and wastewater remediation. This research investigated these challenges through integrated optimization methodologies that combine algorithmic frameworks with enhanced bioprocessing techniques to enhance efficiency, economy, and scalability. A System-of-Systems (SoS) meta-architecture was developed using genetic algorithms and fuzzy assessor functions to demonstrate a pathway toward cost reduction. Rigorous mechanical and chemical characterizations were quantitatively analyzed to reveal existing optimization strategies and further evaluated the best strategies to use in enhancing mass transfer for improved biomass yield. The work also integrates …

Dynamic Response, Assessment, And Retrofitting Of Prestressed Concrete Bridge Girders Subjected To Lateral Impact Loads, Haitham A. Abdelmalek

Doctoral Dissertations

Over height vehicle collisions pose a growing threat to the resilience of bridge infrastructure across the United States. According to the American Road & Transportation Builders Association (ARTBA, 2024), approximately 36% of all U.S. bridges require major repair or replacement, with an estimated cost of $400 billion. This dissertation investigates the structural dynamic response, damage assessment, and retrofitting strategies for prestressed concrete (PC) bridge girders subjected to lateral impact loading.

The research comprised two main phases: (1) numerical modeling and (2) experimental testing. A validated 3D nonlinear finite element (FE) model was developed to conduct parametric studies on impact behavior, …

Synthesis Of Graphene Using Carbonaceous Materials In An Ultrasonic Reactor, Paul Chukwuma Ani

Doctoral Dissertations

The synthesis of high-quality graphene from sustainable carbonaceous feedstocks offers an avenue to reduce the environmental and economic costs associated with conventional graphite-based production. This research investigates the co-utilization of biochar and graphite as precursors for graphene fabrication in an ultrasonic reactor. Biochar was produced from biomass via downdraft gasification at 850 °C, yielding a high fixed-carbon, partially graphitized material with favorable surface area and porosity. Graphite, selected for its crystalline structure, was investigated with biochar to assess the influence of precursor composition on exfoliation efficiency, layer thickness distribution, and defect density. Ultrasonic-assisted liquid-phase exfoliation was employed, with process parameters …

Advancements In Signal Processing And Image Reconstruction For Active Microwave Thermographic Measurements, Logan Martin Wilcox

Doctoral Dissertations

Active microwave thermography, or AMT, is a coupled electromagnetic-thermographic nondestructive testing and evaluation technique. AMT has found success in a variety of inspection needs in the aerospace, space, and infrastructure fields due to its unique type of thermal excitation. During an AMT inspection, a specimen is exposed to microwave energy from a radiating source (i.e., an antenna). This exposure to microwave energy results in dielectric/magnetic heating, which causes an increase in temperature and potential defect indications to manifest on an inspection surface (which is measured via an infrared camera). Due to the use of an antenna, there is a spatially …

Deep Learning Architecture Design For Nano-Satellite Image Super-Resolution, William Everette Symolon

Doctoral Dissertations

Increasing threats to U.S. national security satellite constellations have resulted in an increased interest in constellation resilience and satellite redundancy. NanoSats have contributed to commercial, scientific and government applications in remote sensing, communications, navigation, and research. They also have the potential to enhance satellite constellation resilience. However, the inherent size, weight, and power limitations of NanoSats enforce constraints on imaging hardware; the small lenses and short focal lengths result in imagery with low spatial resolution, which limits the utility of CubeSat images for military planning purposes and national intelligence applications. This research proposed a deep learning architecture capable of enhancing …

Investigating The Effects Of Fly Ash On The Properties Of Silica-Deficient High Alumina Materials, Sai Akshay Ponduru

Doctoral Dissertations

This research is made up of five studies mainly focused on attaining sustainability in high alumina cements like calcium aluminate cement (CAC) and calcium sulfoaluminate belite cements (CSAB) by partially replacing them with additives and supplementary cementitious materials (SCMs) like fly ash (FA). The first study focused on understanding the reaction kinetics of CAC and FA binders at different replacement levels (i.e. between 10%-to-50% by mass) and their effect on mechanical properties. It was also shown that the reactive nature of FAs can be determined by calculating a single parameter called number of constraints (nc) derived from topological constraint theory. …

Data-Driven Mulitiscale Modeling Of Electrochemical Transport, Non-Linear Electrical Contact, And Additive Manufacturing Processes, Emmanuel Olugbade

Doctoral Dissertations

The growing demand for high-efficiency energy systems and advanced manufacturing technologies requires predictive frameworks that link atomic-scale physics with engineering-scale performance. This dissertation develops a unified multiscale modeling approach that integrates molecular dynamics, density functional theory, finite-element analysis, and machine learning to connect structure, transport, and performance across materials and processes. Depending on the interactions involved, the framework employs loose coupling for parameter transfer, tight coupling for two-way feedback, and hybrid coupling where machine-learning surrogates accelerate high-fidelity simulations while retaining physical interpretability. In electrochemical systems, an XGBoost-enhanced single-particle model reproduces P2D-level electrolyte potential dynamics at roughly one-hundredth the computational cost, …

Characterization And Deportment Of Anode Impurities In Copper Electrorefining, Charles Michael Campbell

Doctoral Dissertations

The objective of this research was to study the deportment of the group 15 elements, arsenic, antimony and bismuth during copper electrorefining. Samples were collected from six industrial copper anodes with different compositions. Specimens were physically characterized and electro refined to understand the differences in the behavior of selected impurities. Inclusions in the cast metal structures were characterized using automated scanning electron microscopy and energy dispersive spectroscopy to measure and correlate their size, shape and composition. Arsenic and lead were found to have a positive correlation between concentration and size of inclusions. Using wavelength dispersive spectroscopy, multiphase inclusions were examined, …

Optical Sensor Instrumentation For Enhanced Continuous Caster Development, Hanok Wondimagegnehu Tekle

Doctoral Dissertations

Fiber-optic sensors are an emerging technology that can enhance process monitoring and control in steelmaking. They are especially valuable in the continuous caster’s harsh environment. This dissertation presents the industrial application and demonstration of three sensor types: single-mode silica fiber with Rayleigh-based Optical Frequency Domain Reflectometry (OFDR), sapphire fiber (single-crystal alumina) with Fiber Bragg Gratings (sFBG), and an in-line Raman spectroscopy probe. Each sensor served a distinct role in the continuous caster. Rayleigh OFDR sensors were embedded in tundishes for distributed thermal mapping at 7-mm spatial resolution across ≈4 m. Measurements were taken during preheating, casting, and ladle exchanges, and …

Identification Of Corrosion Damage, Vibration, And Loose Conections In Aircraft Data Transmission Lines Using Reflected And Transmitted Signals; Formulation Of Vegetable Oil-Based Nanofluids As Cutting Fluids For Mql Machining, Saidanvardzhon Valiev

Doctoral Dissertations

This research presents a novel Frequency Domain Transmitometry (FDT) method that uses transmitted signals (S21) to detect and characterize aircraft data transmission lines (ADTL) corrosion damage without additional reflectometry circuitry. Corrosion experiments were conducted according to ASTM G85-A5 over 14 weeks. Combined FDT and reflected signals (S11) analyses revealed distinct three-peak signatures associated with damage and corrosion. The square root of the Area Under the Curve (AUC) of the FDT damage peak and the Full Width at Three-Quarter Maximum (FW3QM) of the S11 damage peak is correlated with corrosion propagation depth and width. S11 signals were further used for vibration …

Deep Learning-Driven Biometric Security: Advancing Liveness Detection And Anti-Spoofing Techniques, Banafsheh Adami

Graduate Theses, Dissertations, and Problem Reports (ETD)

Biometric authentication has become a key part of our everyday lives—from unlocking smartphones with a fingerprint or face to verifying identities in banks and airports. These systems rely on our unique physical or behavioral traits, making them both convenient and secure. Unlike passwords, biometrics cannot be forgotten or stolen in the traditional sense. However, they are not without risk. One of the biggest concerns is spoofing: attempts by attackers to fool systems using fake biometric traits, such as silicone fingerprints or AI-generated videos.

As generative AI tools become more powerful and accessible, the ability to create convincing fake biometric data …