Computer Sciences Commons^™

Detecting Bitstream-Level Fpga Trojans With An Snn, Kylie Arnett

Shelby Hall Graduate Research Forum Presentations

Limited research has been conducted on SNNs for FPGA Trojan detection. FPGA design is often handled by manufacturers outside the U.S. FPGA manufactures outsource production to third-party foundries. This multi-step process introduces security vulnerabilities and increases risk of Hardware Trojan insertion.

Key Questions: To what extent can an FPGA be manipulated at the bitstream level to enable or disable encryption algorithms?

Can SNNs accurately detect the presence of Trojans within an FPGA?

Designing For Trust In Chat-Based Question Answering Systems: An Exchange-Based Retrieval Approach, Nathan Mccutchen

Master's Theses

Community chat platforms such as Discord and Slack support spontaneous, collaborative communication but make it difficult to retrieve previously discussed information. As conversations accumulate, valuable exchanges become buried, leading to repeated questions and sustained burden on experienced community members.

This work contributes a set of design requirements for question-answering systems operating over unstructured chat data, a Discord bot prototype implementing those requirements named Echo, and an empirical evaluation of how such a system affects user trust. Rather than encoding discrete question-answer pairs or generating synthetic responses with a language model, Echo indexes conversation topics for semantic retrieval and presents results …

A Scalable Iteration Of The Horizon Simulation Framework Using Multithreading Techniques, Jason E. Beals

Master's Theses

The Horizon Simulation Framework (HSF) occupies a unique space in the modern aerospace modeling landscape, enabling flexible, modular modeling of mission-level agent behavior through an object-oriented, hierarchical design. HSF's hallmark breadth-first search scheduling algorithm explores a "multiverse" of possible mission execution pathways, enabling exhaustive evaluation of schedule combinations against user-defined heuristics.

As aerospace systems become increasingly complex, HSF faces critical challenges in establishing verifiable, deterministic behavior. The framework's core scheduling algorithm had not undergone systematic validation, leaving questions about temporal consistency, state management correctness, and reproducibility across different program executions. Furthermore, the exponential growth of schedule combinations creates computational bottlenecks …

Graph Convolution Neural Network And Deep Q-Network Optimization-Based Intrusion Detection With Explainability Analysis, Kelvin Mwiga, Mussa Dida, Leandros Maglaras, Ahmad Mohsin, Helge Janicke, Iqbal H. Sarker

Research outputs 2022 to 2026

As networks expand in size and complexity, coupled with an exponential increase in intrusions on network and IoT systems, this leads to traditional models failing to capture increasingly intricate correlations among network components accurately. Graph Convolution Networks (GCNs) have recently acquired prominence for their capacity to represent nodes, edges, or entire graphs by aggregating information from adjacent nodes. However, the correlations between nodes and their neighbours, as well as related edges, differ. Assigning higher weights to nodes and edges with high similarity improves model accuracy and expressiveness. In this paper, we propose the GCN-DQN model, which integrates GCN with a …

Stock Market Price Prediction Using Big Data Models Comparison Analysis, Vibhor Pal

Shelby Hall Graduate Research Forum Posters

The stock market consists of complex financial datasets, and achieving stock price real time prediction needs an efficient big data framework for processing. This paper compares big data distributed data processing frameworks for forecasting stock prices using Graph Neural Networks (GNNs) - Apache Flink and Apache Spark. We analyze 70 publicly traded companies’ monthly data for the last 5 years from Yahoo Finance, ranked by Price-to-Earnings (P/E). In the companies’ datasets, there may be a connection or similarity between companies, and this can lead to similar stocks’ price behavior. These interfirm relationships are maintained by GNNs models, and their output …

Evaluating Software-Based Hardware Abstraction As A Fault Injection Countermeasure, Tristan Clark, J. Todd Mcdonald

Shelby Hall Graduate Research Forum Posters

Despite how theoretically secure a system may be, it can be compromised if an adversary has physical access to the device. This can be done by injecting hardware corruptions directly into the physical system, which is called a fault injection (FI) attack. To combat this, there is a need for robust fault-tolerant countermeasures. One such countermeasure is obfuscating the program to introduce redundancy and complexity, particularly through software-based hardware abstraction (SBHA).

This research proposes using SBHA as a countermeasure to fault injection attacks. By transforming point-function password programs, the proposed countermeasure aims to increase the difficulty of conducting successful FI …

Explainable Deep Reinforcement Learning For Real-Time Network Intrusion Detection, Sebastian Bustamante

Shelby Hall Graduate Research Forum Posters

This research aims to enhance current Deep Reinforcement Learning (DRL)-based Intrusion Detection System (IDS) models by adding transparency using Explainable Artificial Intelligence (XAI). This study proposes a DRL-based IDS architecture that incorporates explainability to provide interpretable reasons for IDS decisions. A Deep Q-Network (DQN) agent will be trained in a simulated network using well-known datasets to learn traffic behavior. XAI methods will be applied to extract feature importance and allow users to understand why alerts were generated. The outcomes of this research will contribute to improving network security by providing more insight on how XAI could be adopted into modern …

Text Corpus Combined Method And Tools For Music Textual Analysis, Yuwei Lu

Shelby Hall Graduate Research Forum Posters

While there has been a great deal of research conducted on how to search images and video using text, there has been less focus on how to retrieve music using multiple facets such as mood and lyrical content. or its features This is due in part to the historical lack of available musical corpuses. A musical corpus is a specialized text corpus specifically designed to capture information about music. Recently, several musical corpora for music information retrieval have been built and made available. However, these corpora and the tools designed to exploit them are typically designed to facilitate only one …

Integrating Nonlinear Phase Space Analysis And Image-Based Representation For Network Intrusion Detection, Chakriya Suon

Shelby Hall Graduate Research Forum Posters

With the rise of cyber threats, cybersecurity continues to play a critical role in the ever-changing landscape of technology by protecting and defending against threat agents. Our research applies novel machine learning (ML)techniques to detect network intrusions effectively. Our primary focus is to extend prior research, which has used network flows that are processed by a nonlinear phase space algorithm (NLPSA). The NLSPA approach has proven extremely effective in detecting anomalous or malicious traffic patterns on representative data but requires extensive training time.

Our contribution integrates deep learning into the anomaly detection approach by creating image-based representations of the adjacency …

Evaluating The Effects Of Anti-Forensic Activities In Additive Manufacturing Devices, Daniel B. Miller

Shelby Hall Graduate Research Forum Posters

Additive Manufacturing (AM) is a newer famlily of production tecchologies that constructs objects by fusing layers of material into the desired shape. Methods for achieving this as described in Gibson et al. [1] are varied and include Fused Filament Deposition, Selective Laser Sintering, Stereolithography (SLA), and Powder Bed Fusion. Computers are integral to the processes being responsible for creating and decoding design instructions, collecting and processing sensor data, and, ultimately, directing the activity of the machines which implement the process. Additionally, the AM industry is rapidly expanding, worth an estimated $23 billion in 2023 and projected to reach $88 billion …

Cellebrite Reliability In Digital Forensics, Christina Huynh

Shelby Hall Graduate Research Forum Posters

Forensic tools like Cellebrite are commonly used in court to gather and interpret raw data for evidence. Cellebrite does not only collect data but creates and interprets the artifacts of data to create a scene of the process it has been through. With this, evidence can be influenced by software designs and not just the data on the mobile device. Courts and police use Cellebrite to gather evidence and reconstruct it to create an easily readable dataset. These tools lack reproducibility, transparency, integrity, and chain of evidence command. Cellebrite is often used in court and by police without further vetting …

Deconstructing Digital Disinformation: Social Media Data Preparation And Analysis For Healthcare Research, Russell W. Cantrell, Matt Campbell

Shelby Hall Graduate Research Forum Posters

The spread of medical misinformation poses significant threats to public health, healthcare system stability, and the quality of patient care. Our research examines misinformation targeting the U.S. healthcare system. It uses a mixed-methods approach that includes social media data analysis, surveys of practicing nurses, and agent-based simulation. This poster focuses on the initial phase, which attempts to detect potential misinformation and disinformation by analyzing patterns in social media posts and user account behaviors. A detailed account of the data preparation process lays the groundwork for examining how disinformation operates online. This phase draws on the Pushshift repository, which offers historical …

Detecting Sensor Data Manipulation, Ricky Green, Michael Black

Shelby Hall Graduate Research Forum Posters

The integration of Information Technology (IT) and Operational Technology (OT) have made OT devices vulnerable to threats that have been successfully exploited with devastating results. Many modern techniques for hardening and securing enterprise IT systems are either incompatible with OT components in an Industrial Control System (ICS), reduce the efficiency of processes, or are prohibitively expensive to implement. Research in the area of ICS security focuses on a top-down approach, such as intrusion prevention by securing the perimeter of the network at layers 3 – 5 of the Purdue model by hardening IT systems. This approach is useful in Enterprise …

Brain Computer Interfaces: Enhancing Low-Cost Eeg Performance Through Deep, Anwar Rassoul

Shelby Hall Graduate Research Forum Posters

The field of Brain Computer Interfacing (BCI) has traditionally been confined to clinical and research environments due to the high cost and complexity of medical-grade EEG systems. However, the emergence of low-cost hardware exemplified has catalyzed a shift toward accessible, portable BCI applications. While these devices lower the barrier to entry for developers and researchers, they often suffer from a lower signal-to-noise ratio (SNR). This increased noise makes it difficult to extract the clean neural signatures required for high-accuracy control, particularly when operating in non-shielded, real-world environments.

This research focuses on Steady-State Visually Evoked Potentials (SSVEP), a robust BCI paradigm …

Detecting Sophisticated Cyberattacks On Public Water Infrastructure, Ayrton Purdy

Shelby Hall Graduate Research Forum Posters

From around the globe, malicious actors continually probe critical infrastructure assets for weaknesses. Their backgrounds, goals, and motives may vary, but the purpose of their attacks is the same: to damage, undermine, or exploit the functionality of these assets [2]. At a fundamental level, critical infrastructure is any essential system and asset vital to national security. Critical infrastructure includes assets such as power, transportation, telecommunications, water and wastewater systems (WWS), and many more [3].

Algorithm For Detecting Luks2-Encrypted Containers In Forensic Images, Nicholas Flynn, Michael Black

Shelby Hall Graduate Research Forum Posters

As technology becomes increasingly integrated into daily life, the way in which society interacts with digital content continues to rapidly change. Though there is growth in advancements that help the average user, there is a similar upward trend in crimes committed involving a computer. Figure 1 illustrates the growth in research across many disciplines of digital forensics reflecting the demand for tools which can combat a wide variety of cyber crimes.In the past two decades, with a massive spike since 2017, there has been much literature produced in response to this demand. It can be inferred from the Federal Bureau …

Improving Consensus In Blockchain, Nelson Navas

Shelby Hall Graduate Research Forum Posters

Called the 4th industrial revolution, Industry 4.0 is the latest paradigm for implementing industrial applications. This new approach relies heavily on increased automation, smart machines, human-machine interaction, AI, and telecommunications. Industry 4.0 applications introduce the idea of the smart factory. The integration of information technology (IT) and operational technology (OT) is a key factor that promotes efficiency in the supply chain. All this is predicated in the generation, sharing, and storage of large quantities of data and transactions to facilitate management, traceability, and control of industrial processes. Increased reliance on interconnectedness causes cybersecurity challenges. Access to machinery, infrastructure, IT systems, …

Temporal Eclectic Rule Extraction: Exploring Trustworthy Explainable Artificial Intelligence For Recurrent Neural Networks, Micah Israel

Shelby Hall Graduate Research Forum Posters

Enhancing temporal neural network interpretability can greatly increase the effectiveness of Intrusion Detection Systems (IDS). While explainable Deep Neural Networks (DNN) have been researched heavily in the literature for intrusion detection, explainable temporal neural networks lack the same attention. Current state-of-the-art XAI techniques rely on black-box surrogate explainers, which attempt to generate post-hoc explanations without valuable information inside the model's hidden neurons. To address this, this proposal introduces a novel white-box XAI method, Temporal Eclectic Rule Extraction (TERE), which is designed to provide explainable rules directly from temporal models. TERE aims to enhance decision transparency in IDS by offering interpretable …

Machine Learning On The Edge: Performance And Security Evaluation Of Cnn Implementations In Embedded Systems, Krista Stacey

Shelby Hall Graduate Research Forum Posters

Embedded systems increasingly integrate Machine Learning (ML) for real-time decision-making across loT, infrastructure, and critical systems. However, ecosystems differ significantly in: Latency, Throughput, Energy use, Accuracy of Models Security exposure. Most research evaluates performance or security, not both together. There is a need for a unified cross-platform performance-security evaluation framework

Generative Ai For Text-To-Video Generation: Recent Advances And Future Directions, Kadhim Hayawi, Sakib Shahriar

All Works

Text-to-video (T2V) generation has recently emerged as a transformative technology within the field of generative AI, enabling the creation of realistic, temporally coherent videos based on natural language descriptions. This paradigm provides significant added value in many domains such as creative media, human-computer interaction, immersive learning, and simulation. Despite its growing importance, systematic discussion of T2V is still limited compared with adjacent modalities such as text-to-image and image-to-video. To alleviate the scarcity of discussions in the T2V field, this paper provides a systematic review of works published from 2024 onward, consolidating fragmented contributions across the field. We survey and categorize …

Transfer Learning Neural Networks For Nuclear Forensic Image Morphology Using Image Splitting Techniques, Niko A. Petrocelli, Lee C. Lambert, Brett J. Borghetti, Abigail A. Bickley

Faculty Publications

Manual morphological analysis of actinide particles from scanning electron microscope (SEM) imagery is a critical component of nuclear forensics but is prone to significant inter-analyst variability. To address this challenge, this work develops and evaluates an automated classification method using deep learning. We introduce a methodology based on partitioning 1906 SEM images, representing 13 classes of uranium compounds, into smaller patches for analysis. Three convolutional neural network (CNN) architectures of increasing complexity were compared: a custom baseline CNN, a simple transfer learning model using ResNet50v1, and a complex model featuring hierarchical feature extraction and a spatial attention mechanism built upon …

An Explainable Transformer-Based Model For Phishing Email Detection: A Large Language Model Approach, Mohammad Amaz Uddin, Md Mahiuddin, Iqbal H. Sarker

Research outputs 2022 to 2026

Phishing email is a serious cyber threat that tries to deceive users by sending false emails with the intention of stealing confidential information or causing financial harm. Attackers, often posing as trustworthy entities, exploit technological advancements and sophistication to make the detection and prevention of phishing more challenging. Despite extensive academic research, phishing detection remains an ongoing and formidable challenge in the cybersecurity landscape. In this research paper, we present a fine-tuned transformer-based masked language model, RoBERTa (Robustly Optimized BERT Pretraining Approach), for phishing email detection. In the detection process, we employ a phishing email dataset and apply the preprocessing …

Explainable Artificial Intelligence Models For Detecting Suspicious Bank Transactions, Narasimha Kumar Narasapuram, Syed Afaq Ali Shah, Mohd Fairuz Shiratuddin, Ferdous Sohel

Research outputs 2022 to 2026

Detecting financial crime is a complex challenge due to evolving criminal strategies and fragmented detection systems, particularly in the areas of money laundering and fraud. While it is easy to implement, traditional rule-based approaches lack adaptability to new threats, and machine learning models, though more effective, often function as opaque "black boxes," limiting their practical use in regulated domains like banking, where interpretability and accountability are essential. This research presents a novel framework that combines intrinsic and post-hoc XAI techniques to detect suspicious bank transactions. Intrinsic methods provide model-inherent transparency, while post-hoc methods offer behavior-level explanations, enabling robust cross-verification of …

Driver Behavior Analyzer 2.0: A Modular Framework For Interpretable Driver Safety Analysis From Obd-Ii And Gps Telemetry, Sangwhan Cha, Venkata Sundar Kamesh Durvasula

Harrisburg University Other Works

Driver behavior analysis plays a central role in advancing road safety and enabling data-driven driver feedback. Although commercial telematics platforms offer sophisticated analytics, they are frequently expensive, proprietary, and optimized for enterprise-scale use. At the same time, low-cost On-Board Diagnostics II (OBD-II) adapters make telemetry collection widely accessible, but they typically do not provide higher-level behavioral interpretation.

In this paper, we present Driver Behavior Analyzer 2.0 (DBA 2.0), an offline-first, modular analytics framework that converts OBD-II and GPS telemetry into interpretable safety insights. DBA 2.0 supports ingestion of telemetry logs in CSV and JSON formats, data normalization, rule-based detection of …

Codeultrafeedback: An Llm-As-A-Judge Dataset For Aligning Large Language Models To Coding Preferences, Martin Weyssow, Aton Kamanda, Xin Zhou, Houari Sahraoui

Research Collection School Of Computing and Information Systems

Evaluating the alignment of large language models (LLMs) with user-defined coding preferences is a challenging endeavor that requires a deep assessment of LLMs' outputs. Existing methods and benchmarks rely primarily on automated metrics and static analysis tools, which often fail to capture the nuances of user instructions and LLM outputs. To address this gap, we introduce the LLM-as-a-Judge evaluation framework and present CodeUltraFeedback, a comprehensive dataset for assessing and improving LLM alignment with coding preferences. CodeUltraFeedback consists of 10,000 coding instructions, each annotated with four responses generated from a diverse pool of 14 LLMs. These responses are annotated using GPT-3.5 …

Private Set Intersection: A Systematic Review, Yunbo Yang, Defan Zhu, Jianting Ning, Qi Feng, Xiaoguo Li, Yuejia Cheng, Guomin Yang, Kui Ren

Research Collection School Of Computing and Information Systems

Various services, such as search engines, are increasingly deployed in cloud-based and distributed systems. However, data are typically managed by trusted servers, making user privacy and data security critical concerns. Private set intersection (PSI) is a powerful cryptographic primitive that enables multiple parties to compute the intersection of their datasets without revealing private inputs. It has been extensively studied over the past two decades, leading to significant gains in computational and communication efficiency. Yet, in many real-world scenarios, revealing the raw intersection may still leak sensitive information. To address this, numerous PSI variants have been developed to meet different application …

Comprehensively Evaluating The Perception Systems Of Autonomous Vehicles Against Hazards, Xiaodong Zhang, Jie Bao, Jianlei Chi, Jun Sun, Zijiang Yang

Research Collection School Of Computing and Information Systems

Perception systems are vital for the safety of autonomous driving. In complex autonomous driving scenarios, autonomous vehicles must overcome various natural hazards, such as heavy rain or raindrops on the camera lens. Therefore, it is essential to conduct comprehensive testing of the perception systems in autonomous vehicles against these hazards, as demanded by the regulatory agencies of many countries for human drivers. Since there are many hazard scenarios, each of which has multiple configurable parameters, the challenges are (1) how do we systematically and adequately test an autonomous vehicle against these hazard scenarios, with measurable outcome; and (2) how do …

Improving Credit Card Transaction Fraud Detection Using Cvqboosting, Bethel Hui Ting Loke, Nirvik Sahoo, Bingyan Guan, Minrui Xu, Dev Verma, Paul R. Griffin

Research Collection School Of Computing and Information Systems

This paper introduces a novel hybrid quantum-classical approach to credit card fraud detection using CVQBoost, a hybrid quantum-classical boosting algorithm executed on the photonic Dirac-3 processor from Quantum Computing Inc. (QCi). By integrating a diverse set of weak classifiers, which includes K-nearest neighbours (KNN), linear discriminant analysis, logistic regression, and XGBoost, within a hybrid quantum-classical ensemble, the proposed method demonstrates significant improvements over the latest published classical benchmarks. Experiments on a Kaggle credit card fraud dataset show that the quantum-enhanced model achieves a mean AUC-PR score of over 0.8, corresponding to an approximately 9% relative improvement over the best published …

Deep Learning Approaches For Anti-Money Laundering On Mobile Transactions: Review, Framework, And Directions, Jiani Fan, Lwin Khin Shar, Ruichen Zhang, Ziyao Liu, Wenzhuo Yang, Dusit Niyato, Kwok-Yan Lam

Research Collection School Of Computing and Information Systems

Money laundering is a financial crime that obscures the origin of illicit funds, necessitating the development and enforcement of anti-money laundering (AML) policies by governments and organizations. The proliferation of mobile payment platforms and smart IoT devices has significantly complicated AML investigations. As payment networks become more interconnected, there is an increasing need for efficient real-time detection to process large volumes of transaction data on heterogeneous payment systems by different operators such as digital currencies, cryptocurrencies, and account-based payments. Most of these mobile payment networks are supported by connected devices, many of which are considered loT devices in the FinTech …

Learning To Search For Vehicle Routing With Multiple Time Windows, Kuan Xu, Zhiguang Cao, Chenlong Zheng, Lindong Liu

Research Collection School Of Computing and Information Systems

In this study, we propose a reinforcement learning-based adaptive variable neighborhood search (RL-AVNS) method designed for effectively solving the Vehicle Routing Problem with Multiple Time Windows (VRPMTW). Unlike traditional adaptive approaches that rely solely on historical operator performance, our method integrates a reinforcement learning framework to dynamically select neighborhood operators based on real-time solution states and learned experience. We introduce a fitness metric that quantifies customers’ temporal flexibility to improve the shaking phase, and employ a transformer-based neural policy network to intelligently guide operator selection during the local search. Extensive computational experiments are conducted on realistic scenarios derived from the …