Digital Commons Network^™

Federated And Explainable Spiking Neural Networks For Fair And Privacy-Preserving Nail Disease Diagnostics, Ch Pavani Reddy, Krishnanaik Vankdoth

Mansoura Engineering Journal

Automated nail disease diagnostics provide a non-invasive pathway for identifying underlying systemic health conditions; however, conventional centralized deep learning approaches often raise concerns related to privacy, fairness, and interpretability. Although the original NeuroNail-SNN framework demonstrated an energy-efficient and edge-ready diagnostic solution, its broader clinical adoption remained limited by unresolved trust, transparency, and ethical considerations. In this study, we propose the Federated and Explainable NeuroNail-SNN, which extends the original spiking neural architecture by integrating federated learning (FL), explainable artificial intelligence (XAI), fairness evaluation, and uncertainty quantification within a unified framework. Federated learning enables decentralized model training across hospitals and mobile clinics …

Towards Supporting Real-Time Estimation Of Vehicle Fuel Consumption And Co2 Emissions In Smart City Applications, Abrar Alali, Stephan Olariu

Computer Science Faculty Publications

This paper evaluates a simplified physics-based energy demand model designed to estimate vehicle fuel consumption and CO₂ emissions—a critical tool for sustainable transportation planning and smart city applications. Unlike data-driven regression models that lack generalizability for user-defined conditions or complex physics-based approaches that rely on extensive, often proprietary data, the simplified model is distinguished by its minimal parameter requirements, depending primarily on a single, overarching powertrain efficiency value. A key contribution is the comprehensive empirical evaluation of the simplified model against official Environmental Protection Agency (EPA) test data across multiple driving cycles and vehicle types, providing a rigorous validation previously …

Robust Broadband Characterization Of Flexible Absorbers, Sheet Material, And Liquids Using A Coaxial Structure, Joseph Christopher Stecher

Masters Theses

Modern high-frequency measurement systems require reliable calibration and sample positioning to ensure measurement fidelity. This thesis presents three studies addressing practical limitations in broadband material parameter extraction and instrumentation.

The first study introduces a modified Nicolson–Ross–Weir (NRW) technique for flexible, compression-sensitive materials from 100 MHz to 18 GHz. Rigid 3D-printed spacers ensure precise sample positioning, and a T-matrix–based de-embedding procedure removes spacer effects. Validation using microstrip measurements and full-wave simulation confirms accurate permittivity extraction across compression levels.

The second study extends NRW to sheet materials enabling accurate material characterization. Independent validation using toroidal inductors with leakage correction and parallel-plate capacitors …

Multi-Scale Sensor-Aware Variational Autoencoders For Adaptive Iot Security: Integrating Drift Calibration And Cyberattack Detection, Md Kamal Hossain, Iftekhar Ahmad, Daryoush Habibi

Research outputs 2022 to 2026

Existing approaches to Industrial Internet of Things (IoT) anomaly detection treat sensor drift calibration and cyberattack detection as separate problems, overlooking their strong interdependence and the compound failure modes that arise when drift dynamics are exploited by stealthy adversaries. Industrial Internet of IoT sensor networks therefore require a unified framework that jointly models both phenomena. This paper proposes the first unified multi-scale sensor-aware variational autoencoder (MS-VAE) framework that jointly models sensor drift and malicious activity within shared latent representations, enabling integrated calibration and security monitoring. The framework introduces three key components: (i) a multi-scale latent architecture that captures short-term anomalies …

Sensing-Then-Beamforming: Robust Transmission Design For Ris-Empowered Integrated Sensing And Covert Communication, Xingyu Zhao, Min Li, Ming Min Zhao, Shihao Yan, Min Jian Zhao

Research outputs 2022 to 2026

Traditional covert communication often relies on the knowledge of the warden's channel state information, which is inherently challenging to obtain due to the non-cooperative nature and potential mobility of the warden. The integration of sensing and communication technology provides a promising solution by enabling the legitimate transmitter to sense and track the warden, thereby enhancing transmission covertness. In this paper, we develop a framework for sensing-then-beamforming in reconfigurable intelligent surface (RIS)-empowered integrated sensing and covert communication (ISACC) systems, where the transmitter (Alice) estimates and tracks the mobile aerial warden's channel using sensing echo signals while simultaneously sending covert information to …

A Novel Hybrid Robust Transfer Learning-Based Adaptive Fractional-Order Super-Twisting Sliding Mode Controller Enhanced For Speed Regulation Of Brushless Dc Motor, Seyyed Morteza Ghamari, Asma Aziz, Daryoush Habibi

Research outputs 2022 to 2026

Brushless DC (BLDC) motors are widely used in applications that are highly-efficient, reliable, and compact, such as electric vehicles, robotics, and medical devices. However, the inherent nonlinearities and load sensitivity of BLDC motors require a robust and adaptive control strategy to ensure satisfactory performance under various operating conditions. Sliding mode control (SMC) has been widely used for the BLDC drives. However, because of its simplicity and robustness, the control effectiveness of the control is limited by the sensitivity to the disturbances and the chattering phenomenon. To remedy this, super-twisting (ST) technique has been proposed to achieve smoother response and better …

Unified Bayesian And Machine Learning-Based State Of Charge Estimation In A Pv Battery System, Showrov Rahman

Open Access Dissertations

The state of charge (SOC) of a battery indicates the remaining charge in the battery relative to its nominal maximum charge capacity. The accurate estimation of SOC is of utmost importance for efficient energy management as it indicates the level of energy stored in the battery.

Bayesian and machine learning (ML) approaches represent two distinct estimation strategies for battery state estimation. Bayesian filters, such as the Kalman filter (KF) and particle filter (PF), operate sequentially by incorporating physical models of battery dynamics. In contrast, ML-based methods are data-driven, and their performance largely depends on the quantity and quality of training …

Dataless Neural Networks For Boolean Satisfiability And Network Optimization, Andrew Evan Gautier

Graduate Theses, Dissertations, and Problem Reports (ETD)

Combinatorial optimization problems (COPs) require searching over a finite solution space subject to constraints, with the goal of satisfying an objective function. They arise in operations research, scheduling, resource allocation, circuit design, and many other fields. Many problems in combinatorial optimization (including satisfiability and network design) are NP-hard. Traditionally, researchers have built approximate solvers that return near- optimal solutions efficiently by developing increasingly sophisticated heuristics and meta- heuristics. Deep learning has provided new opportunities for improving combinatorial solvers by leveraging neural guidance to prune the search space. Traditional neural networks have distinct drawbacks in this context: separate training and …

Multi-Timescale Monitoring And Modeling For Resilient Smart Grids: Pmu Anomaly Detection And Battery Digital Twins, Muhammad Imran Hossain

Graduate Theses, Dissertations, and Problem Reports (ETD)

Modern power systems face failures at very different timescales. Cyber-physical disturbances may emerge within seconds, while battery degradation develops over hundreds of operating cycles. Both monitoring problems share the same underlying difficulty: power systems produce measurement data in abundance, but reliably labeled examples of abnormal or degraded operation are scarce. Rare grid events are difficult to label, and battery degradation data are heterogeneous across cells, cycling protocols, and chemistries. This thesis addresses these challenges through two complementary domain-informed learning frameworks that constrain representation learning using information specific to each physical problem.

First, at the system level, T-BiGAN, a Transformer-augmented bidirectional …

Data Augmentation And The Reliability Of Conformal Prediction For Uncertainty Quantification In Medical Imaging, Rizwan Ahamed

Graduate Theses, Dissertations, and Problem Reports (ETD)

The safe clinical deployment of deep learning models for high-stakes medical imaging tasks requires more than high average accuracy; it requires demonstrable, per-case reliability. Uncertainty quantification (UQ) provides the missing signal that tells a clinician when a model prediction can be trusted and when a case should be escalated for expert review. Among UQ approaches, conformal prediction (CP) is especially attractive because it produces prediction sets that are guaranteed, under the assumption of exchangeability, to contain the true label with a user chosen probability, and it does so without assumptions about the model or the data distribution. This report first …

Adaptive Multi-Grade Deep Learning For Highly Oscillatory Fredholm Integral Equations Of The Second Kind, Jie Jiang, Yuesheng Xu

Mathematics & Statistics Faculty Publications

This paper studies the use of Multi-Grade Deep Learning (MGDL) for solving highly oscillatory Fredholm integral equations of the second kind. We provide rigorous error analyses of continuous and discrete MGDL models, showing that the discrete model retains the convergence and stability of its continuous counterpart under sufficiently small quadrature error. We identify the DNN training error as the primary source of approximation error, motivating a novel adaptive MGDL algorithm that selects the network grade based on training performance. Numerical experiments with highly oscillatory (including wavenumber 500) and singular solutions confirm the accuracy, effectiveness and robustness of the proposed approach.

Architecting A Complex Adaptive System Model For Selecting Policies To Reduce Kidney Discard, Lirim Ashiku

Doctoral Dissertations

"The kidney allocation system is a complex, evolving system involving multiple heterogeneous agents. Each agent exhibits emergent behavior that may not fully align with the complex system goals. Therefore, there is a need for a transdisciplinary systems approach to visualize the interdependency among agents and understand the dominant patterns that shape the kidney allocation system.

First, this research presented an incremental hierarchical system engineering approach in identifying the agents’ needs and behaviors toward the complex systems’ goal of maximizing deceased donor kidney utilization and reducing kidney discard. The hierarchical systems approach linked with model-based system engineering aided in eliciting agents’ …

H2 Matrix Compression For Fast Magnetic Field Post-Processing, Calvin M. Demps

University of Kentucky Master's Theses

Large-scale computational electromagnetic simulations frequently require the assembly of dense system matrices, whose computation time and memory cost grow rapidly with problem size and can render high-resolution simulations impractical. This thesis presents a benchmark of the H² hierarchical matrix method applied to the construction of a coupling matrix that maps magnetization current in a structure to the magnetic field at observation points external to the structure. The underlying field interaction is derived from Maxwell's equations and discretized using a Nyström approach, producing a dense matrix.

The method is benchmarked across various geometries of increasing complexity and for increasing mesh …

Element-Based Predictive Modeling Of Hydrothermal Liquefaction Bioproducts Derived From Corn Stover, Isamu Umeda, Meicen Liu, Yi Zheng, Jiefu Wang, Zhiwu Wang, Sandeep Kumar

Civil & Environmental Engineering Faculty Publications

The hydrothermal liquefaction (HTL) process offers an energetic advantage over pyrolysis because it does not require prior drying of the biomass feedstock. However, there are significant challenges in simultaneously estimating both the yields and characteristics of products from the HTL of biomass with theoretical support. This study developed a unique element-based kinetic model to predict the yields, higher heating values, and fuel characteristics of solid residue and heavy bio-oil, based on the temperature, residence time, solid loading, and elemental composition (C, H, N, and O) of corn stover. Furthermore, the model predicted the weights of dissolved carbon and nitrogen in …

Formalizing Asymmetric Control-Telemetry Separation In Distributed Industrial Control Systems, Andrew Manison

College of Graduate Studies: Theses & Dissertations

Distributed industrial control systems often place control and telemetry traffic on the same communication substrate even though the two workloads impose different requirements. Control paths need bounded request-response latency and predictable acknowledgement semantics, whereas telemetry paths benefit from scalable publish-subscribe fanout and tolerance for consumer-side delay. This thesis argues that, for the tested class of mixed workloads on shared commodity infrastructure, these communication roles should be separated architecturally rather than forced through a single protocol. To evaluate that claim, the thesis formalizes an asymmetric control- telemetry pattern and instantiates it in the Asymtra framework using gRPC for synchronous control and …

State-Dependent Queueing For Adaptive Signal Control: A Simulation-Based Performance Evaluation, Shaimaa Alseddiek, Usama Elrawy Shahdah, Hala B. Nafea, Hossam El-Din Moustafa, El-Said Ahmed Marzouk, Mohamed M. Ashour

Mansoura Engineering Journal

Urban traffic congestion persists as a critical challenge to transportation system efficiency, sustainability, and safety. Traditional queuing models utilizing fixed service rates inadequately represent the dynamic feedback between congestion and capacity in real vehicular flow. State-Dependent Queuing Models (SDQMs) address this limitation by modelling service rate as a function of queue length or density. This research advances SDQM application for adaptive traffic signal control through development of a calibrated state-dependent departure rate implemented within a microscopic simulation environment using SUMO and TraCI. Six control strategies including fixed-time, actuated, and two SDQM variants were evaluated across traffic demands ranging from undersaturated …

Consensus And Controversies Of International Guidelines For The Diagnosis, Surveillance, And Management Of Fetal Growth Restriction: An Updated Comparison, Daniele Diane Mascio, Suneet P. Chauhan, Tullio Ghi, Asma Khalil, Juliana G. Martins, Sara Sorrenti, Tamara Stampalija, Fabrizio Zullo, Francesc Figueras

Department of Obstetrics & Gynecology Faculty Publications

OBJECTIVE: To compare areas of consensus and disagreements across contemporary international and national guidelines on the diagnosis, surveillance, and management of fetal growth restriction (FGR).

DATA SOURCES: Electronic searches of MEDLINE from database inception up to March 2026 using MeSH terms and keywords related to FGR and guidelines. STUDY ELIGIBILITY CRITERIA: Critical, structured comparison of national or international guidelines on FGR published since 2010. Final inclusion required unanimous agreement from all authors.

STUDY APPRAISAL AND SYNTHESIS METHODS: Pre-specified extraction across domains: definition; prediction/prevention; surveillance tools and frequency; delivery timing and mode; and labor induction methods. Dual data …

Incremental Cluster Validity Indices And Their Role In Interpreting Lifelong Learning Systems, Niklas Max Melton

Doctoral Dissertations

Clustering and supervised learning are often treated as distinct paradigms, yet both rely on structure in feature space. This dissertation investigates the relationship between cluster validity indices (CVIs) and supervised learning in real-time and lifelong learning settings where data arrive incrementally and cannot be revisited. Across four studies, it develops methods for online cluster validation, uses supervised learning to improve their interpretability, and applies these ideas to evaluating performance degradation in continual learning.

The first study extends incremental cluster validity indices (iCVIs), enabling widely used validation metrics to operate in streaming environments. Experiments on synthetic and real-world datasets show systematic …

Bond Graph And Extended Generalized Average Method With Applications In Cyber-Physical Systems And Power Electronics, Arnold Anthony Fernandes

Doctoral Dissertations

"This research examines two applications of control theory. The first application considers the bond graph (BG) modeling technique, which is used to develop the MATLAB structural analysis toolbox (MATSAT), an open-source toolbox for sensor placement and qualitative system analysis that considers the observability and fault-detection capabilities of multi-domain cyberphysical systems. The toolbox provides information on redundant sensors, guiding the system designer in cost and security trade-offs. The toolbox uses traditional BG causality assignment procedures. Additionally, MATSAT provides optimal causality assignment methods that perform significantly better at assigning causality to BGs with increased junctions, sources, and simple meshes, without encountering causality …

Validating A Low-Cost Radio Frequency Characterization Framework For Development-Grade Software-Defined Radios In Short-Duration Cubesat Missions, Thomas Wayne Francois

Masters Theses

Software-defined radios (SDRs) and CubeSat platforms have reduced the cost and complexity of space-based communication systems, enabling broader participation in satellite missions. While low-cost radio hardware is increasingly accessible, the ability to characterize and validate its performance remains constrained by the high cost and limited access to traditional RF test equipment. This disparity creates a challenge for small satellite development teams, which must characterize communication-system technical performance with limited access to laboratory-grade instrumentation.

This thesis presents a low-cost RF characterization framework for assessing key radio-frequency performance metrics using readily available hardware and measurement techniques. The approach integrates frequency translation, SDR-based …

A Universal Hybrid Model-Free Deep Quantum–Transfer Learning Controller Enhanced By Grey Wolf Optimization For Dc–Dc Boost Converters With Hardware-In-Loop Validation, Seyyed Morteza Ghamari, Asma Aziz

Research outputs 2022 to 2026

This paper proposes a universal hybrid model-free quantum–transfer learning controller with enhanced online grey wolf optimization algorithm (GWO–QTL) for DC–DC boost converter. This system has the characteristics of non-minimum phase behavior, parasitic effects, and fractional-order dynamics because of high frequency operation. These characteristics make analytical modeling complicated and make it difficult to have a single traditional controller that will operate reliably over different converter types. This motivates the creation of a unified model-free control framework that is able to learn directly from the behavior of the converter without relying on the topology specific models. Reinforcement learning, where an agent interacts …

Design Of Artificial Interference Signal Waveforms For Covert Communication Aided By Multiple Friendly Nodes, Xuyang Zhao, Wei Guo, Yongchao Wang, Shihao Yan

Research outputs 2022 to 2026

In this work, we consider a covert communication scenario with multiple friendly interference nodes. The goal is to hide a legitimate communication link from a transmitter to a receiver under a warden’s surveillance. Firstly, we propose a novel strategy for generating artificial noise (AN) signals and formulate a corresponding design problem, aiming to minimize the adverse effects of AN on the legitimate receiver while enhancing communication covertness. Specifically, we optimize the basis matrix for AN signal space using statistical information of the involved channel coefficients, when precise channel state information are unavailable. Secondly, we analyze the geometric structure of the …

Optimizing Ev Battery Charging Using Fuzzy Logic In The Presence Of Uncertainties And Unknown Parameters, Minhaz Uddin Ahmed, Md Ohirul Qays, Stefan Lachowicz, Parvez Mahmud

Research outputs 2022 to 2026

The growing use of electric vehicles (EVs) creates challenges in designing charging systems that are smart, dependable, and efficient, especially when environmental conditions change. This research proposes a fuzzy-logic-based PID control strategy integrated into a photovoltaic (PV) powered EV charging system to address uncertainties such as fluctuating solar irradiance, grid instability, and dynamic load demands. A MATLAB-R2023a/Simulink-R2023a model was developed to simulate the charging process using real-time adaptive control. The fuzzy logic controller (FLC) automatically updates the PID gains by evaluating the error and how quickly the error is changing. This adaptive approach enables efficient voltage regulation and improved system …

Development Of Feature Tokenizer Deep Learning Model For Fault Diagnosis In Marine Propulsion System, Pratik Anand Deshpande, J. Preetha Roselyn, Prabha Sundaravadivel

Electrical Engineering Faculty Publications and Presentations

The fault diagnostics in Brushless Direct Current (BLDC) motor drive system is critical for operational safety and system lifespan in propulsion system applications. However, signature parameters such as currents, voltages, speed, and torque have provided nonlinear behavior, which limits the usefulness of traditional model-based approaches. This research provides a deep learning based intelligent system to monitor the failures in marine propulsion system. Each signal feature is represented as a structured token, with a specific class token used to collect global contextual information. The proposed model captures both local temporal dynamics and global inter-feature interdependence multi-layer self-attention processes, allowing for the …

Integrating Environmental Awareness In Underwater Acoustic Networks: A Comprehensive Review, Sadaf Vahabli, Iftekhar Ahmad, Quoc Viet Phung, Daryoush Habibi, Walid K. Hasan, Ruba Zaheer

Research outputs 2022 to 2026

Underwater Acoustic Networks (UANs) are critical for enabling long-range underwater communication, supporting a wide range of applications, such as environmental monitoring, resource exploration, disaster prevention and marine security. Unfortunately, UANs operate within a limited acoustic spectrum, where these spectrums often overlap with those used by marine animals for communication, navigation and foraging. Additionally, anthropogenic noise from industrial activities contributes to acoustic pollution, intensifying this problem. This frequency overlap poses significant risks to marine life. Therefore, this review highlights the urgent need to develop environmentally aware UANs that minimize harmful acoustic interference with marine mammals, fish and invertebrates. Further, it focuses …

Design, Fabrication Modeling, And Optical Metrology Of Mwir Silicon Metalenses, Weiyu Chen

Graduate Studies Theses and Dissertations 2026

Mid-wave infrared (MWIR) optical systems require compact, broadband, manufacturable components whose performance can be predicted and measured reliably. Silicon metalenses are attractive because silicon combines high refractive index, MWIR transparency, and semiconductor-compatible fabrication. This dissertation develops a scale-bridging design–fabrication-modeling–metrology framework for MWIR silicon metalenses.

A shared full-aperture framework connects meta-atom libraries, physical layout generation, angular-spectrum propagation, point-spread-function calculation, and focal-plane energy metrics. Building on this foundation, the Dispersive Sweatt Model (DSM) incorporates meta-atom dispersion into conventional ray-tracing software, enabling broadband co-optimization of metasurfaces and refractive elements. A process-aware design framework then incorporates scanning electron microscopy (SEM)-measured height–radius relationships produced by …

Securing The Energy Transition: Cyber-Physical Security And Resilience In Next-Generation Power Systems, Airin Rahman

Graduate Studies Theses and Dissertations 2026

Modern power systems are rapidly evolving into renewable-dominated and digitally interconnected cyber-physical infrastructures due to the increasing deployment of distributed energy resources (DERs), inverter-based technologies, and advanced control platforms. Maintaining reliability under high renewable penetration requires flexible resources capable of shifting energy across extended time horizons. Long-duration energy storage (LDES), particularly hydrogen-based energy systems, has therefore emerged as an important enabler of renewable integration, grid flexibility, and resilience. However, the growing dependence on communication, sensing, and distributed control also expands the cyber-physical attack surface of modern power systems, creating security and resilience challenges that conventional operational paradigms were not designed …

Advancing Cyber-Physical Security And Resilience Of Modern Power Systems: Intelligent Monitoring, Secure Operation, And Resilient Recovery, Md Moshiur Rahman

Graduate Studies Theses and Dissertations 2026

Modern power distribution systems are rapidly evolving into cyber-physical, DER-rich, and data-driven networks that rely on extensive sensing, communication, automation, grid-edge intelligence, and operator decision support. While this transformation improves flexibility, observability and controllability, it also expands the cyber-attack surface and increases the risk that cyber intrusions can propagate into physical disturbances, compromised DER operation, degraded situational awareness, and interrupted service continuity. This dissertation advances the cyber-physical security and resilience of modern distribution systems by developing a high-fidelity real-time cyber-physical hardware-in-the-loop testbed using OPAL-RT, EXata CPS, industrial relays, SCADA/RTAC, HMI, and grid-edge devices to emulate realistic DER-integrated distribution grid operation. …

Large Language Models For Neurology: A Mini Review, Donald C. Wunsch, Daniel B. Hier

Electrical and Computer Engineering Faculty Research & Creative Works

Large language models have the potential to transform neurology by augmenting diagnostic reasoning, streamlining documentation, and improving workflow efficiency. This Mini Review surveys emerging applications of large language models in Alzheimer's disease, Parkinson's disease, multiple sclerosis, and epilepsy, with emphasis on ambient documentation, multimodal data integration, and clinical decision support. Key barriers to adoption include bias, privacy, reliability, and regulatory alignment. Looking ahead, neurology-focused language models may develop greater fluency in biomedical ontologies and FHIR standards, improving data interoperability and supporting more seamless collaboration between clinicians and AI systems. Two future developments have the potential to be particularly impactful: (1) …

Safety Aware Continual Reinforcement Learning-Based Output Tracking Control Of Nonlinear Continuous-Time Systems, Irfan Ganie, Sarangapani Jagannathan

Electrical and Computer Engineering Faculty Research & Creative Works

An output feedback (OF)-based control scheme utilizing both a scalable multilayer neural network (MNN) observer and actor–critic MNN via integral reinforcement learning (IRL)/adaptive dynamics programming (ADP) approach for a class of nonlinear systems with output constraints is introduced. The proposed observer, critic, and actor MNN weight updates are derived using a singular value decomposition (SVD) of MNN activation function gradient along with output error, Bellman and control input errors, respectively. Next, the approach incorporates continual learning (CL), utilizing a penalty function in the weight update laws for both actor–critic MNNs to consolidate knowledge from previous tasks and enhance learning in …