Defense Date
2026
Document Type
Thesis
Degree Name
Master of Science
Department
Electrical & Computer Engineering
First Advisor
Robert Klenke
Abstract
Security for resource-constrained devices has become critical as more of these devices are supported on Internet of Things (IoT) networks. The challenge of establishing a secure Root of Trust on these devices is simultaneously yoked by processing limitations and hardware integrity. Consequently, physically unclonable functions (PUFs) have been proposed as a security primitive to provide software entropy due to their non-deterministic operation, and hardware entropy due to their hidden (memory-free) operation. In this project, we specifically investigate the delay-difference PUF (DD-PUF), building on previous work which proposed it as a Machine-Learning resilient Root of Trust for software entropy. Our study approaches the question of DD-PUF adaptation with a focus on physical design scalability and dependencies. Uniqueness, aliasing and uniformity distributions are applied for standard evaluation of PUF devices on Artix-7 FPGA devices. We propose additional metrics for evaluating physical PUF designs with device-based and cell-based dependencies of latch delays, trigger signal delays, and layout-based parasitic effects. Results show reliability of non-deterministic outputs on Artix-7 hardware, and consistent margins for output dependencies on static attributes. From these studies, we propose parameters for detecting layout-based delay and parasitic effects on DD-PUF operation.
Rights
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
5-7-2026