Design and implementation of memory physically unclonable functions on low-power devices

Aguilar Rios, Manuel Antonio (2023) Design and implementation of memory physically unclonable functions on low-power devices. Doctoral thesis, Northern Arizona University.

Text Aguilar_Rios_2023_design_implementation_memory_physically_unclonable_f.pdf - Published Version Download (15MB)

Abstract

Cryptography plays a vital role in safeguarding our sensitive information from falling into the wrong hands. It relies on symmetrical encryption methods like Advanced Encryption Standard (AES) and asymmetrical encryption methods such as Rivest-Shamir-Adleman (RSA) and Elliptic Curve Cryptography (ECC) for authentication between parties. However, the management of cryptographic keys introduces vulnerabilities that hackers can exploit. Moreover, the security of asymmetric encryption methods like RSA and ECC can be compromised by powerful quantum computers.Moreover, a key primitive in cryptography is the random number generator (RNG). Randomness is required in various cryptographic protocols, such as randomizing encryption operations or generating random challenges in authentication. Standard libraries often provide Pseudo-Random Number Generators (PRNGs), which are deterministic and vulnerable to attacks. Cryptographically-Secure Pseudo-Random Number Generators (CSPRNGs) produce highly unpredictable sequences, but they are complex and resource-intensive, making them impractical for low-power devices. Physically Unclonable Functions (PUFs) present a potential solution for addressing challenges in authentication, key management, and random number generation. PUFs offer a unique digital fingerprint that cannot be influenced or replicated, making them well-suited for generating one-time keys. Moreover, PUFs can be True Random Number Generators (TRNGs), producing true random sequences with minimal power consumption and computational resources. By utilizing PUFs for these purposes, cryptographic systems can enhance their security by generating secure keys and the availability of highly unpredictable random numbers. There are various different PUFs, and one widely used PUF is the static random-access memory (SRAM) PUF. However, there are also promising emerging memory technologies like resistive random-access memory (ReRAM) and magnetoresistive random-access memory (MRAM) PUFs. These innovative PUFs show great potential in enhancing the overall security of cryptographic systems. By leveraging these advanced PUF technologies, cryptographic systems can further bolster their security measures. The primary aim of this research is to implement and examine MRAM, SRAM, and ReRAM PUF devices on low-power client devices and incorporate them into cryptographic applications. The study demonstrates that SRAM and MRAM devices suit a unique ternary-addressable physically unclonable function (TAPUF) design. This design not only generates dependable cryptographic keys but also produces true random sequences that successfully pass the National Institute of Technology (NIST) Statistical Testing Suite for random numbers. Moreover, it also shows that the analog responses of pre-formed ReRAM cells can encapsulate a message, even when applied on a low-power microcontroller.

Item Type:	Thesis (Doctoral)
Publisher’s Statement:	© Copyright is held by the author. Digital access to this material is made possible by the Cline Library, Northern Arizona University. Further transmission, reproduction or presentation of protected items is prohibited except with permission of the author.
Keywords:	Authentication; Random number generators; Cryptography; Cybersecurity; Physically Unclonable Functions; Random Number Generators; Random-Access Memory
Subjects:	Q Science > QA Mathematics > QA76 Computer software
NAU Depositing Author Academic Status:	Student
Department/Unit:	Graduate College > Theses and Dissertations College of Engineering, Informatics, and Applied Sciences > School of Informatics, Computing, and Cyber Systems
Date Deposited:	23 Oct 2025 17:46
Last Modified:	23 Oct 2025 17:46
URI:	https://openknowledge.nau.edu/id/eprint/6266

Actions (login required)

IR Staff Record View

Downloads