Side channel resistance quantification and verification

Aaron E. Cohen, Keshab K. Parhi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

In this paper, a formal method for quantifying the side channel resistance based on a statistical approach is developed and a formal CAD method for verifying the side channel resistance of a circuit design is presented. Our analysis indicates that designs which have small means for the difference in power consumption between key bits and high standard deviations for the difference can achieve higher security. Our analytical results for quantifying side channel resistance indicate that it is not feasible to achieve 2 80 security against side channel attacks with low standard deviation or with large means in the difference between key bit runs. Additionally, our analytical results for a preliminary CAD approach to detecting side channel leakage show that significant computing resources are required to gain enough statistics on the power consumption of a circuit to make reasonable approximations about the system's side channel resistance performance prior to implementation when the side channel attacks are unknown; however, our results indicate that it is feasible to detect individual side channel vulnerabilities when the side channel attacks are known.

Original languageEnglish (US)
Title of host publication2007 IEEE International Conference on Electro/Information Technology, EIT 2007
Pages130-134
Number of pages5
DOIs
StatePublished - Dec 1 2007
Event2007 IEEE International Conference on Electro/Information Technology, EIT 2007 - Chicago, IL, United States
Duration: May 17 2007May 20 2007

Publication series

Name2007 IEEE International Conference on Electro/Information Technology, EIT 2007

Other

Other2007 IEEE International Conference on Electro/Information Technology, EIT 2007
Country/TerritoryUnited States
CityChicago, IL
Period5/17/075/20/07

Fingerprint

Dive into the research topics of 'Side channel resistance quantification and verification'. Together they form a unique fingerprint.

Cite this