Qizhi, Tian (2013)
Novel Power Trace Processing Methods for Side-Channel Analysis of Cryptosystems.
Technische Universität Darmstadt
Ph.D. Thesis, Primary publication
|
Text
Novel Power Trace Processing Methods for Side-Channel Analysis of Cryptosystems.pdf Copyright Information: CC BY-NC-ND 2.5 Generic - Creative Commons, Attribution, NonCommercial, NoDerivs . Download (4MB) | Preview |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Novel Power Trace Processing Methods for Side-Channel Analysis of Cryptosystems | ||||
Language: | English | ||||
Referees: | Sorin A., Prof. Dr. Huss ; Ray C.C., Prof. Dr. Cheung | ||||
Date: | 5 August 2013 | ||||
Place of Publication: | Darmstadt | ||||
Date of oral examination: | 23 May 2013 | ||||
Abstract: | In 1999, scientist Kocher proposed a way to reveal the secret key of cryptosystems by exploiting the leaked side channel information from a smart card. Since then, such a research field becomes more and more important. The motive for doing that arises from the interests on the one hand, and the strategic vision behind the information security in modern society accompanied with unimaginable high speed technology development on the other. In this work, our new proposed attack method, i.e., power amount analysis, is generalized and abstracted firstly, which leads to power amount analysis methodology based on the mostly utilized additive white Gaussian noise channel in the telecommunication field. This methodology conveys two important conceptions. On the one hand, it proposes a way to process the captured power traces to extract and purify the information leakage more efficiently, meanwhile, reduces the dimensionality for the analyzed data resulting to simple calculation in real attacks; on the other hand, various distinguishers may be executed for this attack rather than the calculation of the correlation coefficient. Second, in order to improve the attack methods, a least squares estimation based trace form leakage model is proposed. Based on such a model, power amount analysis mutation I and II are suggested for perusing better attack performance. Subsequently, an attack framework is given, which provides more possibilities to retrieve keys from cryptosystems. Third, a series of trace pre-processing methods are proposed to neutralize the misalignment in captured power traces produced from a random clock featured cryptosystem in terms of horizontal alignment and vertical matching. Thereafter, two trace pre-processing frameworks are given concentrating on the misaligned and originally aligned power trace pre-processing and attacks, respectively. According to the different attack requirements and implementations, one can choose appropriate trace pre-processing and attack methods selectively in real attacks to achieve a good attack performance. Last but not the least, all the proposed attack and trace pre-processing methods and frameworks are successfully verified and evaluated by exploiting different cryptographic implementations running with the different clock types and frequencies, which may be good tools to evaluate the system security for yielding safe cryptosystems and architectures in reality. |
||||
Alternative Abstract: |
|
||||
URN: | urn:nbn:de:tuda-tuprints-35663 | ||||
Classification DDC: | 000 Generalities, computers, information > 004 Computer science | ||||
Divisions: | 20 Department of Computer Science 20 Department of Computer Science > Integrated Circuits and Systems |
||||
Date Deposited: | 08 Aug 2013 12:02 | ||||
Last Modified: | 09 Jul 2020 00:30 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/3566 | ||||
PPN: | 386305560 | ||||
Export: |
View Item |