A 0.16pJ/bit Recurrent Neural Network Based PUF for Enhanced Machine Learning Atack Resistance
This addresses the problem of machine learning attack resistance for IoT device security, but it appears incremental as it builds on existing PUF designs with specific modifications.
The paper tackles the susceptibility of existing strong Physically Unclonable Functions (PUFs) to machine learning attacks by proposing a Recurrent-Neural-Network PUF (RNN-PUF), which achieves a machine learning attack accuracy of 62% and reliability above 93%, representing a 33.5% improvement in their Figure-of-Merit.
Physically Unclonable Function (PUF) circuits are finding widespread use due to increasing adoption of IoT devices. However, the existing strong PUFs such as Arbiter PUFs (APUF) and its compositions are susceptible to machine learning (ML) attacks because the challenge-response pairs have a linear relationship. In this paper, we present a Recurrent-Neural-Network PUF (RNN-PUF) which uses a combination of feedback and XOR function to significantly improve resistance to ML attack, without significant reduction in the reliability. ML attack is also partly reduced by using a shared comparator with offset-cancellation to remove bias and save power. From simulation results, we obtain ML attack accuracy of 62% for different ML algorithms, while reliability stays above 93%. This represents a 33.5% improvement in our Figure-of-Merit. Power consumption is estimated to be 12.3uW with energy/bit of ~ 0.16pJ.