The use of a strong physical unclonable function (PUF) for device authentication in resource-
limited settings is promising but susceptible to machine learning (ML) attacks. Several
defenses have been suggested, but they often have drawbacks such as increased hardware
requirements, compromised reliability, and limited effectiveness against sophisticated ML
attacks. To overcome these challenges, a one-time key-based (OTK) obfuscation approach
has been proposed aiming to bolster the security of robust PUFs when facing machine�…