Digital Commons Network^™

Removing Eog Artifacts From Eeg Recordings Using Deep Learning, Christian O'Reilly, Scott Huberty

Faculty Publications

The electroencephalogram (EEG) directly measures the electrical activity generated by the brain. Unfortunately, it is often contaminated by various artifacts, notably those caused by eye movements and blinks (EOG artifacts). Such artifacts are usually removed using an independent component analysis (ICA) or other blind source separation techniques. However, it is difficult to assess whether subtracting EOG components estimated through ICA removes some neurogenic activity. It is crucial to address this question to avoid biasing EEG analyses. Toward that objective, we developed a deep learning model for EOG artifact removal that exploits information about eye movements available through eye-tracking (ET). Using …

From Adversarial Attacks To Robust Classifiers - A Study In Social Media Spam Detection - Black Box & White Box, Jonathan Jose Penaloza Rumie

Undergraduate Theses

Adversarial attacks pose a significant threat to the reliability of machine learning-based spam detection systems in social media. This undergraduate thesis, "From Adversarial Attacks to Robust Classifiers: A Study in Social Media Spam Detection – Black Box & White Box," systematically examines the impact of both black-box and white-box adversarial attacks on a range of spam classifiers, including Logistic Regression, Decision Trees, Random Forests, K-Nearest Neighbors, Bagging, Gradient Boosting, and Support Vector Machines. Leveraging a novel dataset derived from Twitter spam messages and enhanced with adversarial perturbations such as synonym replacement and character-level modifications, this study evaluates classifier performance under …

31 - Shaped Adversarial Patches, Huong Quach

Undergraduate Research Symposium

In recent years, the development and deployment of computer vision models have become widespread, with applications ranging from autonomous vehicles to security systems. Among these, object detection algorithms like YOLO are particularly significant due to their real-time performance and accuracy in identifying and localizing objects within an image. However, the robustness of these models is increasingly challenged by adversarial attacks, which are deliberate manipulations designed to deceive the model's predictions.

In this paper, I present an approach to advancing the deception capabilities of adversarial patches, specifically targeting YOLO-based person detectors. The objective is to design and implement shaped adversarial patches …

On The Robustness Of Adaptive Resonance Theory Neural Networks, Shane Cairns, Leonardo Enzo Brito Da Silva, Sasha Petrenko, Donald C. Wunsch

Miners Solving for Tomorrow Research Conference

No abstract provided.

Antenna Multi-Band Enhancement By Employing Negative Impedance Converters, Kevin Pepin

Doctoral Dissertations and Master's Theses

This thesis presents the implementation of a Negative Impedance Converter (NIC) integrated with a planar monopole antenna, originally centered at 2.9 GHz, to create a multi-band network. The network exhibits resonances at frequencies of 0.9 GHz, 1.5 GHz and 1.9 GHz. The work highlights the design process of pairing a NIC with a capacitive, single-band antenna, addressing key challenges such as overcoming the limitations associated with low self-resonant frequency (SRF) inductors. The design process included the use of measurement-based component models and radio-frequency simulations using Keysight Advanced Design Systems (ADS). The developed NIC emulates a -0.99 pF capacitor using two …

On The Provenance Of Software Systems: Automating Software Traceability With Knowledge Graph And Large Language Model Synergy, Tyler Procko

Doctoral Dissertations and Master's Theses

The present dissertation delineates a system that enables those engaged in software development to automatically generate and maintain project life cycle provenance. All projects are implemented and made manifest with the development of artifacts, e.g., papers, code files, etc. Tools exist to accelerate artifact creation, but little focus is paid to the processes that produce them. In terms of Ontology, or, from Ancient Greek, the study of being, the two most basic entities in reality are Continuant and Occurrent, or, roughly, “Artifact” and “Process”. This dissertation posits that for any created artifact, its process of creation, i.e., its life …

Developing Fixed-Bias Langmuir Probes For Multi-Point Constellation Deployments, Henry Carter Valentine

Doctoral Dissertations and Master's Theses

Since their initial development in the early 20th century, electrostatic Langmuir probes have proved invaluable in terrestrial and interplanetary ionospheric sounding applications. When deployed aboard rocket and satellite platforms, these probes yield high-cadence, in-situ measurements of key plasma parameters such as electron density, ion density, and electron temperature. Furthermore, the efficacy of Langmuir probes in characterizing the three-dimensional structure and dynamics of ionospheric plasmas can be augmented by the technique of multi-payload deployments. In this work, we discuss the design, development, and analysis of fixed-bias Langmuir probes constructed for two multi-point science campaigns: the Mars-bound, dual-satellite Escape and Plasma Acceleration …

Heat-Pipe-Based Thermal Management System Design For A 250 Kw Gan-Based Integrated Modular Motor Drive, Seyed Iman Hosseini Sabzevari, Salar Koushan, Armin Ebrahimian, Towhid Islam Chowdhury, Nathan Weise, Ayman El-Refaie

Electrical and Computer Engineering Faculty Research and Publications

Integrated modular motor drive (IMMD) is an effective approach for realizing high-efficiency, high-power-density, and fault-tolerant electric machines. However, designing an efficient thermal management system (TMS) for the motor drive becomes a challenge, particularly due to space constraints. This article presents the design of a TMS based on 3-mm heat pipes for a 250-kW IMMD intended for aviation applications. The power electronics module is simulated using PLECS software where an electrothermal analysis is conducted. A simplified thermal resistance model of the system is developed to estimate the die junction temperature of gallium nitride (GaN) semiconductors. The performance of the proposed TMS …

Enhancing Iot Security Using Lightweight Machine Learning Algorithms: A Comprehensive Approach Using Ensemble Learning, Feature Selection, And Federated Transfer Learning, Khawlah Harahsheh

Electrical & Computer Engineering Theses & Dissertations

The rapid expansion of the Internet of Things (IoT) has introduced significant security vulnerabilities due to the resource-constrained nature of IoT devices and their exposure to cyber threats. Traditional security solutions are often infeasible due to the high computational and storage demands they impose. This dissertation presents a lightweight, AI-driven security framework that enhances IoT network resilience by integrating feature selection, ensemble learning, and federated transfer learning while maintaining data privacy and minimizing computational overhead.

The proposed framework consists of three primary components: Feature Selection for Intrusion Detection, which optimizes performance by reducing redundant data and improving detection accuracy with …

Streamer Discharge Simulation For Plasma-Assisted Combustion, Stuart Jairo Reyes

Electrical & Computer Engineering Theses & Dissertations

A common and successful method to achieve atmospheric pressure fuel-air plasma-assisted combustion is through repetitive ns pulsed discharges and dielectric-barrier discharge. The transient phase in these discharges is dominated by transport influenced by strong space charges produced by ionization fronts, this can be best represented by the streamer model. The function of non-thermal plasma in these discharges is to excite the species in the fuel-air mixture to produce radicals which accelerate the chemical conversion reactions which directly lead to temperature rise, ultimately culminating in ignition. Therefore, the characterization of the streamer and its energy partitioning is essential to developing a …

Enhancing Ai Reliability: Probabilistic Inference And Neural Optimization, Behnam Zeinali Rizi

USF Tampa Graduate Theses and Dissertations

Deep neural networks have recently shown better performance than traditional machine learning algorithms in various applications such as computer vision, natural language processing, indoor navigation, and biomedical signal/bio-informatics data processing tasks. However, some main challenges still exist when harnessing these models for different applications. The first challenge is the performance of these methodologies for some specific types of data, such as Biomedical images, IMU measurements, and even bioinformatics data, such as genome expression. Moreover, deploying these models on mobile and IoT devices is computationally challenging. Most devices use cloud computing, where powerful deep-learning models analyze the data on a server. …

Design And Development Of A Low-Cost Educational Robot: A Scalable And Affordable Software Learning Platform, Diana Refaat Henry Jacob, Mohamed Tarek Elawa, Ashraf Mohamed Hafez

Journal of Engineering Research

This study presents the design and development of a low-cost educational robot to enhance learning experiences in robotics and programming. The system integrates robust software development, ensuring seamless interaction between sensors, actuators, and a custom-designed graphical user interface (GUI). The robot is equipped with various sensors and actuators to support a broad range of experiments and educational applications. The software architecture is detailed, focusing on its modularity and ease of use, enabling students to intuitively program and interact with the robot. Additionally, a series of experiments were conducted to evaluate the performance of the sensors and actuators, providing insights into …

Non-Orthogonal Multiple Access Empowered Physical Layer Security Systems: Review, Issues And Challenges, Aml Fawzy, Mahmoud Selim, Maha Elsabrouty, Sameh Napoleon, Mustafa M. Abd Elnaby

Journal of Engineering Research

As 5G networks advance, the demand for higher data rates, enhanced spectral efficiency, and increased connectivity intensify. Non-orthogonal multiple Access (NOMA) addresses these needs by allowing multiple users to share the same time and frequency resources, thus optimizing network resource utilization and significantly boosting system capacity and throughput. NOMA's influence extends beyond traditional communication scenarios, impacting various vertical industries that require extensive connectivity, such as the Internet of Things (IoT). This transformative approach is crucial for industrial and critical mission applications. Given the importance of safeguarding these communications from potential eavesdroppers, Physical Layer Security (PLS) emerges as a vital tool. …

Deep Learning Applications For Predictive Modeling In Cancer Therapy And Enzyme Encoding, Mengmeng Liu

LSU Doctoral Dissertations

Predictive modeling has revolutionized computational biology and molecular bioinformatics, enabling significant advancements in cancer therapy and functional enzyme characterization. Despite considerable progress, significant challenges remain in accurately predicting combinational cancer therapies and systematically representing enzyme functions for computational applications. Traditional methods struggle with capturing the complex interactions between drugs and biological networks, as well as representing hierarchical relationships within enzyme classifications. This dissertation addresses these limitations by developing advanced deep learning models tailored to enhance predictive performance in both domains.

First, a data augmentation strategy is introduced to improve anticancer drug synergy prediction by generating pharmacologically relevant drug pairs based …

Role Of Eye-Tracking Technology And Software Algorithms In Enhancing Adhd Detection And Diagnosis: A Systematic Literature Review, Lauren E. Perkins

Honors College Theses

This systematic literature review explores the role of eye-tracking technology and software algorithms in enhancing the detection and diagnosis of ADHD. ADHD, a neurodevelopmental disorder affecting both children and adults, is traditionally diagnosed through behavioral assessments, which may lack objectivity. Recent studies suggest that eye-tracking, specifically focusing on saccades, fixations, and blink rates, offers the potential for more accurate and objective measures of ADHD. The review examines clinical trials, observational studies, and machine learning research to assess the correlation between ADHD and eye movement patterns. Results indicate that individuals with ADHD exhibit distinct eye movement patterns, which can be quantified …

The Impact Of Solar Angle And Cloud Shadows On 3d Reconstruction Of Rolling Stock Cargo, Carlina M. Ostrand, Adam D. Reiman, Frank W. Ciarallo, Scott L. Nykl, Clark N. Taylor, Joshua F. Krutz

Faculty Publications

Meeting the relentless demand for more efficient air cargo transportation is of paramount importance for commercial needs and military missions. This study describes an experiment to test an innovative approach that harnesses cutting-edge stereoscopic vision technology to create 3D point clouds of rolling stock cargo across varying solar angles and cloud shadow conditions. Virtual cargo point clouds are generated by calibrating and systematically organizing the depth and location points from an RGB-D camera and then reprojecting them in a virtual environment. Measurement accuracy was rigorously tested across six camera positions in various combinations of weather conditions against physical ground truth …

Optimizing Parameters For Efficient Computation With Fully Homomorphic Encryption Schemes, Cavi̇dan Yakupoğlu Karaağaç, Kurt Rohloff

Turkish Journal of Electrical Engineering and Computer Sciences

In this study, we aim to provide a parameter selection approach for the BFVrns scheme, one of the prominent fully homomorphic encryption (FHE) schemes. Selecting parameters for lattice-based FHE schemes poses a practical challenge for both experts and nonexperts. To solve this problem, we introduce a hybrid approach that combines theoretical approach with experimental analysis. First, we employ regression analysis to examine the impact of parameters on both performance and security. The varying behavior of FHE parameters in terms of performance, security, and ciphertext expansion factor (CEF) makes parameter selection more challenging. To address this issue, we employ a multi-objective …

Decomposition Lstm With Dual Multi-Head Self-Attention For Wind Turbine Drivetrain State Forecasting, Haikun Jia, Huini Sun, Shuang Bai

Turkish Journal of Electrical Engineering and Computer Sciences

Due to the clean and renewable nature of wind energy, accurate prediction of rotor loads and operating states for wind turbine units has become of paramount importance. Currently, traditional methods relying on expert analysis combined with instrument testing for qualitative reasoning are both time-consuming and labor-intensive, and their accuracy guarantees are limited. In response to wind farm data entailing the interweaving of data from multiple sources and the diverse interrelations across various features and time steps, this study introduces a method for predicting rotor loads and operating states. Initially, we employ an iterative multi-scale seasonal-trend decomposition block to capture latent …

Fuzzy-Virtual Inertia Control To Improve The Frequency Response Of Multi-Area Power Systems, Nourelhouda Djaraf, Yacine Daili, Abderrahim Zemmit, Abdelghani Harrag

Turkish Journal of Electrical Engineering and Computer Sciences

Virtual inertia control (VIC) is essential for power systems dominated by electronic devices to compensate for the lack of inertia and ensure frequency regulation. However, most existing VICs often focus solely on optimizing the virtual inertia parameter to adapt to the high penetration of renewable energy sources (RESs) without considering the damping factor. This oversight can lead to significant fluctuations and power mismatches, especially in interconnected systems where the coordination between MGs is sensitive and essential, and there is a risk of propagation of deviations between MGs, which makes the control more complex. To address these issues, this paper presents …

Conversational Open-Domain Question Answering For Resource-Constrained Languages, Emrah Budur, Tunga Güngör

Turkish Journal of Electrical Engineering and Computer Sciences

The growing interest in Conversational AI has led to the development of Conversational OpenQA systems as a crucial step for meeting users' information needs in real world scenarios. Conversational OpenQA systems enhance standard OpenQA performance by leveraging conversation history of the users. However, building effective Conversational OpenQA systems requires large-scale Conversational OpenQA datasets, often limited to the English language, hindering progress in low-resource languages. We present a robust Conversational OpenQA system enhanced by conversational context, designed for languages with limited resources and exemplified in our case study for Turkish. To address data limitations in a cost-effective way, we repurpose existing …

A Case Study Of Gray-Box Fuzzing With Byte- And Tree-Level Mutation Strategies In Xml-Based Applications For Exposing Security Vulnerabilities, Şerafetti̇n Şentürk, Vahi̇d Garousi, Nejat Yumuşak

Turkish Journal of Electrical Engineering and Computer Sciences

Fuzzing is an automated process for detecting crashes and vulnerabilities in software system and it is classified as grammar- or mutation-based in terms of input generation. While the grammar-based fuzzing generates inputs from a specification and takes highly-structured inputs, mutation-based fuzzing generates inputs by modifying input files and abstract syntax trees randomly. There are not many case studies comparing the crash detection capabilities in the scope of mutation-based fuzzing. To add to the body of empirical evidence in this area, this case study compares fuzzing with different mutation strategies to evaluate their effectiveness in three aspects: fault detection effectiveness, fault …

Multi-Channel Features Combined With Gray Features For Remote Sensing Image Classification, Khaled Hasan, Hussien Harb, Lamiaa A. Elrefaei, Hala M. Abdel-Kader

Mansoura Engineering Journal

Image classification in remote sensing is crucial for various applications like deforestation monitoring and Land Use/Land Cover (LULC) mapping. While grayscale images offer simplicity, recent advancements have enhanced classification by integrating spatial and relative data. However, relying solely on grayscale images overlooks valuable color information from different channels. To overcome this, researchers have explored two strategies: combining diverse features and utilizing multi-channel features. These methods merge spatial information and exploit shared features across channels. Despite their improved performance compared to grayscale, they face increased dimensionality, necessitating dimensionality reduction techniques. In our study, we propose a novel approach that integrates spatial …

Distributed Repeaters For Extending The Range Of Harmonic Rf Tags, Trevor Saunders

USF Tampa Graduate Theses and Dissertations

Passive harmonic Radio Frequency (RF) tags are highly advantageous compared to theiractive counterparts due to their ability to operate effectively in environments that are cluttered or where access to the tag for battery maintenance is challenging, dangerous and/or time consuming. They are also resistant to the local oscillator (LO) phase noise in radar applications where the doppler frequency from slow moving targets is buried beneath the noise. One of the major constraints of passive harmonic RF tags, however, lies in their operating range and the significant amount of power required to overcome the conversion loss and activate the nonlinear device …

Modeling And Simulations Of An Input Coupler For A High-Power Traveling Wave Tube Amplifier As A Preliminary Investigation Into A Multi-Stream Traveling Wave Tube Amplifier, Ethan Wade

Electrical and Computer Engineering ETDs

The University of New Mexico (UNM) is working on developing a multi-stream Traveling Wave Tube (TWT) amplifier. One of the first steps to be taken during the design of a TWT amplifier is to design an input coupler to couple power from a lower power High Power Microwave (HPM) seed source to the slow wave structure (SWS) of the amplifier. In order to aid in the development of a TWT which operates in X-band, multiple options for input couplers have been designed for use in a greater TWT project at UNM. The primary tasks to be addressed with these input …

Neurophisology Biosignals Of Cognitive Training Classification In Virtual Reality Environment Using Deep Learning Model, Nourhan Zayed, Mohamed Reda

Mechanical Engineering

This research investigates the potential of neurophysiological biosignals fusion, such as electroencephalogram (EEG) and Eye Tracking signals (ET), to classify cognitive states during virtual reality (VR) training, specifically for the rehabilitation of neurodegenerative diseases. By analyzing EEG data collected from participants engaged in VR-based cognitive exercises, we aim to identify patterns associated with different cognitive states and develop a robust classification system. A Convolutional Neural Network (CNN) model was developed to predict task performance utilizing neurophysiological inputs in an immersive world. This system could be used to monitor cognitive function, assess treatment efficacy, and provide real-time feedback to adapt the …

Conversational Voice User Interfaces Supporting Individuals With Down Syndrome: A Literature Review, Franceli L. Cibrian, Concepción Valdez, Lauren Min, Vivian Genaro Motti

Engineering Faculty Articles and Research

Conversational Voice User Interfaces (CVUIs) are widely used in commercial applications such as personal assistants. CVUIs are beneficial for most users as they enable interaction through speech and natural language. However, recent studies indicate that underrepresented user groups, such as individuals with speech impairments and specifically those with Down syndrome, face challenges in using voice commands to control CVUIs. The anatomical and physiological differences affecting the voice, speech, fluency, and prosody of users with Down syndrome hinder their experience with CVUIs. This article presents the results of 43 papers related to the use of voice user interfaces supporting individuals with …

Free Space Optical Communications: Challenges, Mitigation Techniques, Classification Framework, And Standardization, Moataz Mohamed Roshdy, Ahmed Hosny Elghandour, Kamel Mohamed Hassan

Future Engineering Journal

Over the past several years, there has been a renewed interest in Free Space Optical (FSO) communication technology, commonly referred to as Optical Wireless Communications (OWC). FSO is sometimes viewed as a substitute for current technology, such radio frequency. In other situations, FSO is viewed as a potent contender to work in tandem and complement next-generation technology, such 5G wireless networks. FSO technology is therefore widely used in a variety of indoor (such as data centers), terrestrial (such as mobile networks), space (such as intersatellite and deep space communication), and underwater systems (such as underwater sensing). Nevertheless, despite the enormous …

Mg And Fe Co-Doped Perovskites For The Electrochemical Oxidative Coupling Of Methane In Solid Oxide Electrolyzers, Luke H. Denoyer

Chemical and Biological Engineering ETDs

More efficient usage of shale gas reserves and natural gas resources will allow for higher olefin yields and lower greenhouse gas emissions. The oxidative coupling of methane (OCM) is a direct pathway for converting methane to ethylene at temperatures >700 °C. The OCM process is typically regulated via the choice of catalyst, reactor conditions, temperature range and gas mixing. The electrochemical oxidative coupling of methane (E-OCM) is a modification to the OCM reaction by controlling the oxygen content via externally applied potentials to an SOEC. Perovskite oxide catalysts are useful for E-OCM due to their flexible bond networks and desired …

Multiscale Materials Characterization And In-Situ Process Monitoring In Additive Manufacturing Via Non-Contact Optical Thermometry Techniques, Rifat-E-Nur Hossain

Doctoral Dissertations

While additive manufacturing (AM) is experiencing rapid growth, its development is uneven across different branches. Some areas are still emerging, while even the more established branches are still facing ongoing challenges that require further development. Regardless of their development stage, both emerging and mature AM require process monitoring and part characterization. Process monitoring helps to achieve more control over the process and build a self-adaptive system, while characterization of printed parts speeds up process optimization and ensures required quality. Together, process monitoring and build characterization will transform AM into a more dependable and commercially viable technique. Build surface temperature is …

Novel Approach To Traveling-Wave-Based Fault Location In Nonhomogeneous Transmission Lines, John Parker Burrell

Master's Theses

Accurate fault location is a critical aspect of power system protection, ensuring grid reliability and minimizing downtime. Traditional traveling-wave-based fault location methods face limitations when applied to nonhomogeneous transmission lines due to the reliance on precise segment velocities and propagation time data. This thesis addresses these challenges by proposing a novel algorithm that leverages historical fault data to estimate segment velocities and refine these estimates as more faults occur. The algorithm was rigorously tested using the digital model of an 11-segment, 65.694 km real overhead transmission line and validated using both simulations and hardware tests using commercially available time-domain protective …