PhasePrint: Exposing Cloud FPGA Fingerprints by Inducing Timing Faults at Runtime
| dc.contributor.author | Mahmod, Jubayer | en |
| dc.contributor.author | Hicks, Matthew | en |
| dc.date.accessioned | 2025-04-04T12:14:04Z | en |
| dc.date.available | 2025-04-04T12:14:04Z | en |
| dc.date.issued | 2025-03-30 | en |
| dc.date.updated | 2025-04-01T07:47:46Z | en |
| dc.description.abstract | Cloud FPGAs, with their scalable and flexible nature, are rapidly gaining traction as go-to hardware acceleration platforms for compute-intensive workloads. However, their increasing adoption introduces unique security challenges. The hardware-level access that FPGAs provide leads to many vulnerabilities, including the leakage of sensitive information through data remanence and the creation of analog-domain covert channels among users. A foundational requirement in these scenarios is the ability to target an individual FPGA; knowing this, cloud vendors prevent FPGA localization by restricting access to low-level information of the underlying hardware. Beyond aiding adversaries, FPGA localization enables defenders to strategically rotate FPGA usage, preventing prolonged exposure that can lead to confidential data leakage due to long-term data remanence. This paper introduces PhasePrint, a cloud FPGA localization approach using dynamic timing faults in functionally valid circuits. PhasePrint induces timing faults in a specially crafted circuit at runtime and infers delay characteristics from the resulting error pattern—without incorporating information sources blocked by cloud vendors. PhasePrint utilizes an FPGA’s internal clock synthesizer to derive a clock pair with a strict phase relationship. By adjusting the phase relationship of these clocks, PhasePrint intentionally causes timing faults at runtime that reveal manufacturing variations among FPGA chips. We transform these fault locations into feature vectors to create device signatures and train a multiclass classifier on a dataset from 300 unique FPGAs across four AWS geographic regions. This entirely on-chip signature extraction method achieves >99% accuracy, operates 13× faster, and costs 92% less than the state-of-the-art. | en |
| dc.description.version | Published version | en |
| dc.format.mimetype | application/pdf | en |
| dc.identifier.doi | https://doi.org/10.1145/3676641.3716012 | en |
| dc.identifier.uri | https://hdl.handle.net/10919/125146 | en |
| dc.language.iso | en | en |
| dc.publisher | ACM | en |
| dc.rights | Creative Commons Attribution 4.0 International | en |
| dc.rights.holder | The author(s) | en |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | en |
| dc.title | <i>PhasePrint</i>: Exposing Cloud FPGA Fingerprints by Inducing Timing Faults at Runtime | en |
| dc.type | Article - Refereed | en |
| dc.type.dcmitype | Text | en |