TechRxiv
hw-em.pdf (826.93 kB)

Detecting the Exploitation of Hardware Vulnerabilities using Electromagnetic Emanations

Download (826.93 kB)
preprint
posted on 28.07.2021, 17:34 by Giancarlo Canales BarretoGiancarlo Canales Barreto, Nicholas Lamb
We present a cache attack monitoring methodology that leverages statistical machine learning models to detect n-day hardware attacks by analyzing the electromagnetic emanations of a device. Experimental results from a Raspberry Pi 4 hosting Linux and a Jetson TX2 development board running a Linux guest hosted by seL4 demonstrate that our approach can sense Spectre attacks with a concordance statistic of 97% and 95%.

History

Email Address of Submitting Author

canalesbarreto@battelle.org

ORCID of Submitting Author

0000-0002-0965-5420

Submitting Author's Institution

Battelle Memorial Institute

Submitting Author's Country

United States of America