Author ORCID Identifier
https://orcid.org/0000-0002-7435-4675
Defense Date
2026
Document Type
Dissertation
Degree Name
Doctor of Philosophy
Department
Computer Science
First Advisor
Irfan Ahmed
Abstract
Additive manufacturing builds objects by depositing material layer by layer. It now produces medical implants, aerospace parts, and military components. This reach into safety-critical domains makes these systems attractive targets for attackers.
Additive manufacturing is a cyber-physical system, so it leaks information into the physical world. This leakage, called side-channel data, carries useful details about the printing process. Careful analysis of this data can reveal both attacks and process faults. Most prior work secures polymer printing while 3D bioprinting remains far less studied. This dissertation addresses that gap with both offensive and defensive contributions.
On the offensive side, we study how attackers can damage and steal from bioprinting. We design and evaluate sabotage attacks that lower cell viability and print quality. We build a taxonomy of bioprinter firmware attacks. We also demonstrated a new class of intellectual property theft for 3D bioprinters using power side channel. The attack BioLeak, uses a machine learning model that recovers partial printer activities from passive power measurements. This is the first LLM-assisted IP theft attack against a bioprinter.
On the defensive side, we first systematize side-channel monitoring for additive manufacturing. This bridges the security and quality assurance communities. We then present three detection frameworks. BioSaFe, uses multiple sensors and treats G-code as ground truth to catch sabotage. WattShield, profiles power signatures to detect malicious firmware in fused-filament printers. BioShield, extends this power-based approach to extrusion-based 3D bioprinting. Together these works map the attack surface of additive manufacturing and build practical defenses.
Rights
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
7-27-2026