Predicting Hard and Soft-Responses and Identifying Stable Challenges of MUX PUFs using ANNs

摘要

In this paper, we present artificial neural network (ANN) models to predict hard and soft-responses of three configurations of arbiter based physical unclonable functions (PUFs): standard, feed-forward (FF) and modified feed-forward (MFF). The models are trained using data extracted from 32-stage arbiter PUF circuits fabricated using IBM 32 nm HKMG process. The contributions of this paper are two-fold. First, we evaluate the unpredictability of the PUFs by predicting hard responses using ANNs and comparing these with ground truth. Second, ANNs are trained to predict soft-responses and a probability based thresholding scheme is used to define stability. The obtained soft-response models are used to identify unstable responses.