Open Access. Powered by Scholars. Published by Universities.®
- Discipline
- Keyword
-
- Anomaly detection (2)
- Cybersecurity (2)
- Adaptive training (1)
- Adversarial attack (1)
- Adversarial defense (1)
-
- Artificial Intelligence, Cybersecurity, Literature Review (1)
- Data Recovery (1)
- Distributed energy resources (1)
- GRU autoencoder (1)
- Industrial Internet of Things (1)
- Industrial control systems (1)
- Intrusion detection (1)
- Multimodal hate mitigation (1)
- Network flow analysis (1)
- Operational technology network (1)
- P4 (1)
- Phasor Measurement Unit (1)
- Programmable Network (1)
- Protocol gateway (1)
- Robust training (1)
- Sequence-to-sequence autoencoder model (1)
- Software-Defined Networking (1)
- Solid (1)
- Unsupervised learning (1)
- Vision-Language Models (1)
Articles 1 - 6 of 6
Full-Text Articles in Cybersecurity
F.L.A.I.R. -- A Flow-Level Autoencoder For Intrusion Recognition, Joseph P. Dumond
F.L.A.I.R. -- A Flow-Level Autoencoder For Intrusion Recognition, Joseph P. Dumond
Electrical Engineering and Computer Science Undergraduate Honors Theses
Industrial Internet of Things (IIoT) networks underpin critical infrastructure worldwide, yet securing them remains an open challenge. Traditional intrusion detection systems require labeled attack data for training, a resource that is rarely available in real industrial deployments. They also fail against novel threats, a model trained on known attacks has no basis for detecting anything outside its training set. This thesis presents a Flow-Level Autoencoder for Intrusion Recognition, or FLAIR, a fully unsupervised deep learning system for network intrusion detection in IIoT environments. FLAIR is built on a Gated Recurrent Unit (GRU) autoencoder trained exclusively on normal network traffic. Rather …
A Case Study On Using Large Language Models To Drive Convergence-Oriented Literature Discovery In Critical Infrastructure Security, Caden J. Williamson
A Case Study On Using Large Language Models To Drive Convergence-Oriented Literature Discovery In Critical Infrastructure Security, Caden J. Williamson
Electrical Engineering and Computer Science Undergraduate Honors Theses
In the world of cybersecurity, the rapid development of artificial intelligence proposes a constant challenge for researchers to defend critical infrastructure. Attacks on critical infrastructure can be catastrophic, and emerging strategies of cyber-adversaries that implement leading AI models can expose vulnerabilities in critical infrastructure much faster than previous tools. To defend against this emerging threat, the Cybersecurity Research Working Group at the University of Arkansas is aiming to develop a cross-domain and cross-discipline center of excellence. To support this effort, the group is writing a literature review on the topics of AI and critical systems security. Literature review is an …
Leveraging P4 Programmable Switches For Resilient Operation And Design Of Phasor Measurement Unit Networks, Eva Casto
Electrical Engineering and Computer Science Undergraduate Honors Theses
The power grid utilizes a device called the phasor measurement unit (PMU), allowing power system administrators to remotely monitor and manage the state of the grid in Wide Area Monitoring Systems (WAMS). The advantages of PMUs – such as fine-grained, time-synchronized measurements and efficient, decentralized monitoring – are what make them key devices in the power grid. However, PMU technology also comes with new threats of the digital age, like malfunctions and cyberattacks, which can result in missing and faulty measurements that compromise power grid observability. P4 programmable networks can be used to detect faulty PMU data in a decentralized, …
Real-Time Anomaly Detection In Ot Networks Using Gru-Based Autoencoders, Grant Austin Wilkins
Real-Time Anomaly Detection In Ot Networks Using Gru-Based Autoencoders, Grant Austin Wilkins
Graduate Theses and Dissertations
Operational Technology (OT) networks, particularly those used in critical infrastructure, face increasing cyber threats that target network-level protocols and behaviors. While most anomaly detection research for OT systems has traditionally relied on sensor data, this thesis explores the viability of detecting malicious activity directly from network telemetry. We propose a sequence-to-sequence autoencoder model based on Gated Recurrent Units (GRUs) with multilevel attention, trained to reconstruct normal patterns of packet-level communication extracted from raw PCAP data. The developed feature engineering pipeline integrates general networking attributes such as IP and MAC addresses, ports, and transport protocols with OT-specific protocol information from Modbus …
Securing Distributed Energy Resources: A Secure Gateway For Modbus To Solid Communication Using A Raspberry Pi, Donna R. Thakadipuram
Securing Distributed Energy Resources: A Secure Gateway For Modbus To Solid Communication Using A Raspberry Pi, Donna R. Thakadipuram
Electrical Engineering and Computer Science Undergraduate Honors Theses
As distributed energy resources (DERs) such as solar panels, wind turbines, and battery storage systems become more common, securing their communications has become increasingly important. Many of these systems still rely on legacy communication protocols such as Modbus, which were not designed with cybersecurity in mind. This project addresses this challenge by developing a secure communication gateway that allows Modbus RTU devices to interface with decentralized Solid pods, which are personal data storage units that give users control over their information. This system is built on a Raspberry Pi 4, and it translates telemetry data from Modbus into a Solid-compatible …
Adversarial Machine Learning: Methods For Attacks And Defenses, Minh Hao Van
Adversarial Machine Learning: Methods For Attacks And Defenses, Minh Hao Van
Graduate Theses and Dissertations
With the rapid development of machine learning in real-world applications, enhancing security plays an important role. Adversarial machine learning focuses on understanding malicious actions from attackers and developing defensive techniques against such threats when deploying machine learning systems. An attack can occur in different scenarios, such as poisoning attacks during the training stage and evasion attacks during the testing stage. Although extensive research has explored defense strategies to deal with these harmful attacks, there is a need for further research into areas such as how to counteract malicious attacks with healthy noise or how to train an adaptive defense against …