On the Relevance of Feature Selection for Profiled Side-channel Attacks

In the process of profiled side-channel analysis there is a number of steps one needs to make. One important step that is often conducted without a proper attention is selection of the points of interest (features) within the side-channel measurement trace. Most of the related work start with an assumption that the features are selected and various attacks are then considered and compared to find the best approach. In this paper, we concentrate on the feature selection step and show that if a proper selection is done, most of the attack techniques offer satisfactory results. We investigate how more advanced feature selection techniques stemming from the machine learning domain can be used to improve the side-channel attack efficiency. Our results show that the so-called Hybrid feature selection methods result in the best classification accuracy over a wide range of test scenarios and number of features selected.

[1]  Annelie Heuser,et al.  Intelligent Machine Homicide - Breaking Cryptographic Devices Using Support Vector Machines , 2012, COSADE.

[2]  Huan Liu,et al.  Feature Selection for High-Dimensional Data: A Fast Correlation-Based Filter Solution , 2003, ICML.

[3]  Ron Kohavi,et al.  Irrelevant Features and the Subset Selection Problem , 1994, ICML.

[4]  Markus G. Kuhn,et al.  Efficient Template Attacks , 2013, CARDIS.

[5]  Yongbin Zhou,et al.  How to Compare Selections of Points of Interest for Side-Channel Distinguishers in Practice? , 2014, ICICS.

[6]  N. Meinshausen,et al.  Stability selection , 2008, 0809.2932.

[7]  Christof Paar,et al.  Templates vs. Stochastic Methods , 2006, CHES.

[8]  Olivier Markowitch,et al.  Power analysis attack: an approach based on machine learning , 2014, Int. J. Appl. Cryptogr..

[9]  P. Rohatgi,et al.  Test Vector Leakage Assessment ( TVLA ) methodology in practice , 2013 .

[10]  José Carlos Cortizo,et al.  Multi Criteria Wrapper Improvements to Naive Bayes Learning , 2006, IDEAL.

[11]  Stefan Mangard,et al.  Power analysis attacks - revealing the secrets of smart cards , 2007 .

[12]  Ron Kohavi,et al.  Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[13]  Emmanuel Prouff,et al.  Breaking Cryptographic Implementations Using Deep Learning Techniques , 2016, SPACE.

[14]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[15]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[16]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[17]  Eibe Frank,et al.  Large-scale attribute selection using wrappers , 2009, 2009 IEEE Symposium on Computational Intelligence and Data Mining.

[18]  Ian H. Witten,et al.  Data mining - practical machine learning tools and techniques, Second Edition , 2005, The Morgan Kaufmann series in data management systems.

[19]  A. Ng.  Feature selection, L1 vs. L2 regularization, and rotational invariance , 2004, Twenty-first international conference on Machine learning - ICML '04.

[20]  Nir Friedman,et al.  Bayesian Network Classifiers , 1997, Machine Learning.

[21]  Trevor Hastie,et al.  An Introduction to Statistical Learning , 2013, Springer Texts in Statistics.