A Research of Power Analysis Based on Multiple Classification Models

Aiming at the problem that the single model classification algorithm has a low success rate when the number of training samples is low, We present a power analysis method that combines multiple classification models. We use DPA_Contest_V4 dataset to complete our experiment. First we use the traditional method to break the mask, and then we use SVM, RF and kNN classification algorithm to train and predict as base learners. Finally, we combine these models with ensemble learning or semi-supervised learning. The experimental results show that these two methods are both superior to the single model. Especially when the number of traces in the training set is small, the accuracy can be increased by more than 10%.

[1]  Avrim Blum,et al.  The Bottleneck , 2021, Monopsony Capitalism.

[2]  Zhi-Hua Zhou,et al.  Tri-training: exploiting unlabeled data using three classifiers , 2005, IEEE Transactions on Knowledge and Data Engineering.

[3]  Olivier Markowitch,et al.  Side channel attack: an approach based on machine learning , 2011 .

[4]  Xin Yao,et al.  Ensemble learning via negative correlation , 1999, Neural Networks.

[5]  Joos Vandewalle,et al.  Machine learning in side-channel analysis: a first study , 2011, Journal of Cryptographic Engineering.

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

[7]  N. Altman.  An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression , 1992 .

[8]  Máire O'Neill,et al.  Neural network based attack on a masked implementation of AES , 2015, 2015 IEEE International Symposium on Hardware Oriented Security and Trust (HOST).

[9]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[10]  Elisabeth Oswald,et al.  Practical Template Attacks , 2004, WISA.

[11]  Vladimir Cherkassky,et al.  The Nature Of Statistical Learning Theory , 1997, IEEE Trans. Neural Networks.

[12]  Romain Poussier,et al.  Template Attacks vs. Machine Learning Revisited (and the Curse of Dimensionality in Side-Channel Analysis) , 2015, COSADE.

[13]  Olivier Markowitch,et al.  A Machine Learning Approach Against a Masked AES , 2013, CARDIS.

[14]  Kerstin Lemke-Rust,et al.  Efficient Template Attacks Based on Probabilistic Multi-class Support Vector Machines , 2012, CARDIS.

[15]  Olivier Markowitch,et al.  A machine learning approach against a masked AES , 2014, Journal of Cryptographic Engineering.

[16]  Dawu Gu,et al.  An Improved Side-Channel Attack Based on Support Vector Machine , 2014, 2014 Tenth International Conference on Computational Intelligence and Security.