2011
DOI: 10.1007/978-3-642-27257-8_15
|View full text |Cite
|
Sign up to set email alerts
|

An Exploration of the Kolmogorov-Smirnov Test as a Competitor to Mutual Information Analysis

Abstract: Abstract. A theme of recent side-channel research has been the quest for distinguishers which remain eective even when few assumptions can be made about the underlying distribution of the measured leakage traces. The Kolmogorov-Smirnov (KS) test is a well known non-parametric method for distinguishing between distributions, and, as such, a perfect candidate and an interesting competitor to the (already much discussed) mutual information (MI) based attacks. However, the side-channel distinguisher based on the K… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

1
39
0

Year Published

2011
2011
2021
2021

Publication Types

Select...
4
3
1

Relationship

2
6

Authors

Journals

citations
Cited by 50 publications
(40 citation statements)
references
References 17 publications
(27 reference statements)
1
39
0
Order By: Relevance
“…The authors verified the effectiveness of the distinguisher in one typical attack scenario; the work of [36] presents a more in-depth analysis which we further extend here-finding it to be a reasonable substitute for MIA in first-order scenarios, with some evidence that it is more noise robust. Its adaptation to second-order scenarios is less successful and it does not share MIA's potential for higher-order adaptations.…”
Section: Our Contributionsupporting
confidence: 53%
See 2 more Smart Citations
“…The authors verified the effectiveness of the distinguisher in one typical attack scenario; the work of [36] presents a more in-depth analysis which we further extend here-finding it to be a reasonable substitute for MIA in first-order scenarios, with some evidence that it is more noise robust. Its adaptation to second-order scenarios is less successful and it does not share MIA's potential for higher-order adaptations.…”
Section: Our Contributionsupporting
confidence: 53%
“…This paper explains in greater depth, and applies in greater breadth, the alternative approach presented in [35] and further used in [36]. The objective is to abstract away altogether from the confounding problem of estimation by focussing on the theoretic properties of distinguishers as employed in various given leakage scenarios.…”
Section: Our Contributionmentioning
confidence: 99%
See 1 more Smart Citation
“…b) Non-parametric tests: One can use more general nonparametric tests, such as Kolmogorov-Smirnov [WOM11]. The advantage is that these tests typically rely on fewer assumptions about the underlying distributions; the disadvantage is that they may converge slower and require more samples.…”
Section: Step 3: Apply Statistical Testmentioning
confidence: 99%
“…computed for a wrong sub-key value or a wrong time) and the leakage values tends toward its minimum value (often 0) when the sample size N increases (see e.g. [20] [40]), both assumptions are often not verified in practice, where the adversary must for instance deal with the ghost peaks phenomenon. The situation is even worst for the LRA attacks since the vector of coefficients − → β (and thus the set of predictions) depends not only onk but also on the attack time u.…”
Section: Which Does Not Depend On the Leakage Values Is Denoted By Pkmentioning
confidence: 99%