Cryptology ePrint Archive: Report 2012/082

Semi-Supervised Template Attack

Liran Lerman and Stephane Fernandes Medeiros and Nikita Veshchikov and Cedric Meuter and Gianluca Bontempi and Olivier Markowitch

Abstract: Side channel attacks take advantage of the information leakage in a cryptographic device. A template attack is a family of side channel attacks which is reputed to be extremely effective. This kind of attacks supposes that the attacker can fully control a cryptographic device before attacking a similar one. In this paper, we propose a method based on a semi-supervised learning strategy to relax this assumption. The effectiveness of our proposal is confirmed by software simulations as well as by experiments on a 8-bit microcontroller.

Category / Keywords: Side channel attack, Template attack, Power analysis, Machine learning, Semi-supervised learning, Clustering, Hamming weight

Date: received 22 Feb 2012

Contact author: llerman at ulb ac be

Available formats: PDF | BibTeX Citation

Version: 20120223:214550 (All versions of this report)

Discussion forum: Show discussion | Start new discussion