Publication detail

k-Nearest Neighbors Algorithm in Profiling Power Analysis Attacks

MARTINÁSEK, Z. ZEMAN, V. MALINA, L. MARTINÁSEK, J.

Original Title

k-Nearest Neighbors Algorithm in Profiling Power Analysis Attacks

English Title

k-Nearest Neighbors Algorithm in Profiling Power Analysis Attacks

Type

journal article in Web of Science

Language

en

Original Abstract

Power analysis presents the typical example of successful attacks against trusted cryptographic devices such as RFID (Radio-Frequency IDentifications) and contact smart cards. In recent years, the cryptographic community has explored new approaches in power analysis based on machine learning models such as Support Vector Machine (SVM), RF (Random Forest) and Multi-Layer Perceptron (MLP). In this paper, we made an extensive comparison of machine learning algorithms in the power analysis. For this purpose, we implemented a verification program that always chooses the optimal settings of individual machine learning models in order to obtain the best classification accuracy. In our research, we used three datasets, the first contains the power traces of an unprotected AES (Advanced Encryption Standard) implementation. The second and third datasets are created independently from public available power traces corresponding to a masked AES implementation (DPA Contest v4). The obtained results revealed some interesting facts, namely, an elementary \textit{k}-NN (\textit{k}-Nearest Neighbors) algorithm, which has not been commonly used in power analysis yet, shows great application potential in practice.

English abstract

Power analysis presents the typical example of successful attacks against trusted cryptographic devices such as RFID (Radio-Frequency IDentifications) and contact smart cards. In recent years, the cryptographic community has explored new approaches in power analysis based on machine learning models such as Support Vector Machine (SVM), RF (Random Forest) and Multi-Layer Perceptron (MLP). In this paper, we made an extensive comparison of machine learning algorithms in the power analysis. For this purpose, we implemented a verification program that always chooses the optimal settings of individual machine learning models in order to obtain the best classification accuracy. In our research, we used three datasets, the first contains the power traces of an unprotected AES (Advanced Encryption Standard) implementation. The second and third datasets are created independently from public available power traces corresponding to a masked AES implementation (DPA Contest v4). The obtained results revealed some interesting facts, namely, an elementary \textit{k}-NN (\textit{k}-Nearest Neighbors) algorithm, which has not been commonly used in power analysis yet, shows great application potential in practice.

Keywords

Power Analysis; Machine Learning; Template Attack; Comparison; Smart Cards

Released

01.06.2016

ISBN

1210-2512

Periodical

Radioengineering

Year of study

1

Number

1

State

CZ

Pages from

11

Pages to

28

Pages count

19

Documents

BibTex


@article{BUT118680,
  author="Zdeněk {Martinásek} and Václav {Zeman} and Lukáš {Malina} and Josef {Martinásek}",
  title="k-Nearest Neighbors Algorithm in Profiling Power Analysis Attacks",
  annote="Power analysis presents the typical example of successful attacks against trusted cryptographic devices such as RFID (Radio-Frequency IDentifications) and contact smart cards.
In recent years, the cryptographic community has explored new approaches in power analysis based on machine learning models such as Support Vector Machine (SVM), RF (Random Forest) and Multi-Layer Perceptron (MLP).
In this paper, we made an extensive comparison of machine learning algorithms in the power analysis.
For this purpose, we implemented a verification program that always chooses the optimal settings of individual machine learning models in order to obtain the best classification accuracy.
In our research, we used three datasets, the first contains the power traces of an unprotected AES (Advanced Encryption Standard) implementation.
The second and third datasets are created independently from public available power traces corresponding to a masked AES implementation (DPA Contest v4).
The obtained results revealed some interesting facts, namely, an elementary \textit{k}-NN (\textit{k}-Nearest Neighbors) algorithm, which has not been commonly used in power analysis yet, shows great application potential in practice.
",
  chapter="118680",
  doi="10.13164/re.2016.0365",
  howpublished="print",
  number="1",
  volume="1",
  year="2016",
  month="june",
  pages="11--28",
  type="journal article in Web of Science"
}