Hardware-Efficient Configurable Ring-Oscillator-Based Physical Unclonable Function/True Random Number Generator...

Area of Science:

• * Electronic Engineering
• * Cybersecurity
• * Cryptography

Background:

• * Physical unclonable functions (PUFs) are crucial for verifying electronic device authenticity and securing data.
• * Existing PUF solutions often lack hardware efficiency and sufficient output bit generation.
• * Modern programmable devices offer opportunities for novel PUF/TRNG architectures.

Purpose of the Study:

• * To develop a configurable PUF/TRNG module leveraging Xilinx 7 Series FPGA structures.
• * To enhance hardware efficiency and increase output bit count compared to prior art.
• * To facilitate integration into embedded systems and accelerate on-device validation.

Main Methods:

• * Design of a configurable PUF/TRNG module based on ring oscillators (ROs).
• * Exploitation of RO configurability for statistical characterization.
• * Development of a resource-optimized module version with increased output bits.
• * Integration as a parameterized intellectual property (IP) module with standard interfaces.

Main Results:

• * The module demonstrates adequate reliability, uniqueness, and unpredictability as a PUF.
• * Acceptable randomness and entropy levels were achieved when functioning as a TRNG.
• * A compact 4x4 CLB implementation can obfuscate/recover up to 4096-bit cryptographic keys.

Conclusions:

• * The developed PUF/TRNG module offers an efficient and versatile solution for device authentication and secure data handling.
• * Its design facilitates seamless integration into embedded systems, enhancing security.
• * The module's performance metrics support its application in securing critical data exchange over the internet.