MRAM PUF

Author:

Das Jayita1,Scott Kevin1,Bhanja Sanjukta1

Affiliation:

1. University of South Florida

Abstract

In this work, we have studied two novel techniques to enhance the performance of existing geometry-based magnetoresistive RAM physically unclonable function (MRAM PUF). Geometry-based MRAM PUFs rely only on geometric variations in MRAM cells that generate preferred ground state in cells and form the basis of digital signature generation. Here we study two novel ways to improve the performance of the geometry-based PUF signature. First, we study how the choice between specific geometries can enhance the reliability of the digital signature. Using fabrications and simulations, we study how the rectangular shape in the PUF cells is more susceptible to lithography-based geometric variations than the elliptical shape of the same aspect ratio. The choice of rectangular over elliptical masks in the lithography process can therefore improve the reliability of the digital signature from PUF. Second, we present a MRAM PUF architecture and study how resistances in MRAM cells can be used to generate analog voltage output that are easier to detect if probed by an adversary. In the new PUF architecture, we have the choice between selection of rows and columns to generate unique and hard-to-predict analog voltage outputs. For a 64-bit response, the analog voltage output can range between 20 and 500 mV, making it tough for an adversary to guess over this wide range of voltages. This work ends with a discussion on the threat resilience ability of the new improved MRAM PUF to attacks from probing-, tampering-, reuse-, and simulation-based models.

Publisher

Association for Computing Machinery (ACM)

Subject

Electrical and Electronic Engineering,Hardware and Architecture,Software

Cited by 7 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3