Date of Award
2018
Document Type
Thesis
Degree Name
Master of Science in Engineering (MSE)
Department
Electrical and Computer Engineering
Committee Chair
Aleksandar Milenkovic
Committee Member
Jeffrey Kulick
Committee Member
Biswajit Ray
Subject(s)
Flash memories (Computers), Random number generators
Abstract
This thesis introduces a new technique for generating true random numbers that exploits read noise of perturbed cells in NOR Flash memories. The existing techniques for generating true random numbers rely on additional hardware resources that serve as a source of entropy or exploit memory components such as SRAM or NAND Flash memories. The proposed technique utilizes NOR Flash memories that are readily available in modern microcontrollers. We characterize behavior of the NOR Flash memories, introduce an algorithm for inducing a perturbed state of Flash memory cells, and introduce an algorithm for extracting randomness from these cells and generating true random numbers. The proposed technique is demonstrated and experimentally evaluated on a TI MSP430 family of microcontrollers. The experimental evaluation shows that the proposed technique enables highthroughput and low-cost extraction of random sequences that pass tests from the NIST statistical suite. The proposed technique requires no hardware modifications, is entirely implemented in software, and can be tailored to work in low-end and lowcost embedded systems.
Recommended Citation
Poudel, Prawar, "Using NOR flash memory in microcontrollers for generating true random numbers" (2018). Theses. 249.
https://louis.uah.edu/uah-theses/249