Security Analysis of Digital-Based Physically Unclonable Functions: Dataset Generation, Machine Learning Modeling, and Correlation Analysis
Bibliographic record
Abstract
Physically unclonable functions (PUFs) are circuit primitives that offer a promising and cost-effective solution for various security applications, such as integrated circuits (IC) counterfeiting, secret key generation, and lightweight authentication. PUFs leverage semiconducting variations of ICs to extract intrinsic responses based on applied challenges, establishing unique challenge-response pairs (CRPs) for each device. The security analysis of PUFs is crucial to identify the device weaknesses and ensure response integrity. Accordingly, CRP-based examination plays a major role in defining the resistivity of the block against general and modeling-based attacks. Such analysis requires an updated and representative dataset for training and evaluation. However, there is a lack of benchmark datasets for assessing the effectiveness and resistance of PUF devices. Motivated by this, in this work, we generate a dataset of 300K CRPs for a digital-based PUF implemented on a field programmable gate array (FPGA). The dataset provides a significant number of CRPs for a multi-bit response, where the spatial and temporal adjacency are implicitly defined in the extracted CRPs. Moreover, we investigate different approaches utilizing the generated dataset such as machine learning-based modeling, correlation analysis, and entropy analysis. The CRPs are employed to train linear and nonlinear Support Vector Machine (SVM) models, and the prediction accuracy of SVM models is used as an indicator of the PUF's vulnerability to modeling attacks. As the prediction accuracy does not exceed 65% over 10K CRPs, the extracted dataset sufficiently verifies the resiliency of the device against ML-based modeling attacks. Additionally, Pearson's coefficient is computed on a 10 K-bit vector to determine the correlation between the bits of the response. The calculations expose some correlations between ±0.25, which warns from potential threads. Finally, the paper discusses some potential future research directions and challenges that are envisioned to enhance the security performance of PUFs.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.006 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".