Nano-intrinsic security primitives for internet of everything

Kim, J 2019, Nano-intrinsic security primitives for internet of everything, Doctor of Philosophy (PhD), Engineering, RMIT University.

Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Kim.pdf Thesis application/pdf 72.71MB
Title Nano-intrinsic security primitives for internet of everything
Author(s) Kim, J
Year 2019
Abstract With the advent of Internet-enabled electronic devices and mobile computer systems, maintaining data security is one of the most important challenges in modern civilization. The innovation of physically unclonable functions (PUFs) shows great potential for enabling low-cost low-power authentication, anti-counterfeiting and beyond on the semiconductor chips. This is because secrets in a PUF are hidden in the randomness of the physical properties of desirably identical devices, making it extremely difficult, if not impossible, to extract them. Hence, the basic idea of PUF is to take advantage of inevitable non-idealities in the physical domain to create a system that can provide an innovative way to secure device identities, sensitive information, and their communications. While the physical variation exists everywhere, various materials, systems, and technologies have been considered as the source of unpredictable physical device variation in large scales for generating security primitives. The purpose of this project is to develop emerging solid-state memory-based security primitives and examine their robustness as well as feasibility.

Firstly, the author gives an extensive overview of PUFs. The rationality, classification, and application of PUF are discussed. To objectively compare the quality of PUFs, the author formulates important PUF properties and evaluation metrics. By reviewing previously proposed constructions ranging from conventional standard complementary metal-oxide-semiconductor (CMOS) components to emerging non-volatile memories, the quality of different PUFs classes are discussed and summarized. Through a comparative analysis, emerging non-volatile redox-based resistor memories (ReRAMs) have shown the potential as promising candidates for the next generation of low-cost, low-power, compact in size, and secure PUF.

Next, the author presents novel approaches to build a PUF by utilizing concatenated two layers of ReRAM crossbar arrays. Upon concatenate two layers, the nonlinear structure is introduced, and this results in the improved uniformity and the avalanche characteristic of the proposed PUF. A group of cell readout method is employed, and it supports a massive pool of challenge-response pairs of the nonlinear ReRAM-based PUF. The non-linear PUF construction is experimentally assessed using the evaluation metrics, and the quality of randomness is verified using predictive analysis.

Last but not least, random telegraph noise (RTN) is studied as a source of entropy for a true random number generation (TRNG). RTN is usually considered a disadvantageous feature in the conventional CMOS designs. However, in combination with appropriate readout scheme, RTN in ReRAM can be used as a novel technique to generate quality random numbers. The proposed differential readout-based design can maintain the quality of output by reducing the effect of the undesired noise from the whole system, while the controlling difficulty of the conventional readout method can be significantly reduced. This is advantageous as the differential readout circuit can embrace the resistance variation features of ReRAMs without extensive pre-calibration.

The study in this thesis has the potential to enable the development of cost-efficient and lightweight security primitives that can be integrated into modern computer mobile systems and devices for providing a high level of security.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Engineering
Subjects Microelectronics and Integrated Circuits
Circuits and Systems
Keyword(s) Resistive random access memory
Security primitives
Physical unclonable function
Internet of Things
Hardware security
Version Filter Type
Access Statistics: 223 Abstract Views, 88 File Downloads  -  Detailed Statistics
Created: Tue, 16 Apr 2019, 13:41:05 EST by Keely Chapman
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us