PCB Authentication by Exploiting Chip-level Intrinsic Current Variations

Faculty:Swarup Bhunia
Project Description:Printed circuit boards (PCBs) are inseparable building-blocks of modern electronic systems. However, the long supply chain of PCB manufacturing makes it prone to numerous types of tampering as well as counterfeiting attacks including overproduction, cloning, and recycling. At the PCB level, existing countermeasures either suffer from significant overhead issues or they can be easily breached by opportunistic counterfeiters. In this work, we introduce a current PUF based novel PCB authentication method to protect them from counterfeiting and tampering attacks. It exploits the dynamic current variations stemming from temporal switching activity in FPGAs to generate high quality board- specific digital signatures. The source of the entropy of this PUF is evaluated by transistor-level Monte- Carlo simulation in HSPICE. This experimental approach is applied to perform measurements on 20 different custom-made PCBs and the generated signatures show high uniqueness, robustness, and randomness features and the implementation requires modest hardware overhead (<1%). This intrinsic fluctuation in supply current across PCBs for varying workloads entail unique board-specific signatures, which are extremely difficult to clone. Additionally, remote authentication for IoT Edge devices can be achieved by deploying a robust and trusted current-measuring circuitry at the user’s end to effectively combat PCB tampering and counterfeiting attacks.