The nature of statistical learning theory
The nature of statistical learning theory
Fast training of support vector machines using sequential minimal optimization
Advances in kernel methods
An introduction to support Vector Machines: and other kernel-based learning methods
An introduction to support Vector Machines: and other kernel-based learning methods
CHES '02 Revised Papers from the 4th International Workshop on Cryptographic Hardware and Embedded Systems
Pattern Recognition and Machine Learning (Information Science and Statistics)
Pattern Recognition and Machine Learning (Information Science and Statistics)
Power Analysis Attacks: Revealing the Secrets of Smart Cards (Advances in Information Security)
Power Analysis Attacks: Revealing the Secrets of Smart Cards (Advances in Information Security)
A note on Platt's probabilistic outputs for support vector machines
Machine Learning
A Unified Framework for the Analysis of Side-Channel Key Recovery Attacks
EUROCRYPT '09 Proceedings of the 28th Annual International Conference on Advances in Cryptology: the Theory and Applications of Cryptographic Techniques
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
Template attacks in principal subspaces
CHES'06 Proceedings of the 8th international conference on Cryptographic Hardware and Embedded Systems
Templates vs. stochastic methods
CHES'06 Proceedings of the 8th international conference on Cryptographic Hardware and Embedded Systems
Which is the best multiclass SVM method? an empirical study
MCS'05 Proceedings of the 6th international conference on Multiple Classifier Systems
WISA'04 Proceedings of the 5th international conference on Information Security Applications
COSADE'12 Proceedings of the Third international conference on Constructive Side-Channel Analysis and Secure Design
| Hi-index | 0.00 |
Common template attacks are probabilistic relying on the multivariate Gaussian distribution regarding the noise of the device under attack. Though this is a realistic assumption, numerical problems are likely to occur in practice due to evaluation in higher dimensions. To avoid this, a feature selection is applied to identify points in time that contribute most information to an attack. An alternative to common template attacks is to apply machine learning in form of support vector machines (SVMs). Recent works brought out approaches that produce comparable results, respectively better in the presence of noise, but still not optimal in terms of efficiency and performance. In this work we show how to adapt the SVM template approach in order to considerably reduce the effort while carrying out the attack and how to better exploit the side-channel information under the assumption of an attack model with a strict order, e.g. Hamming weight model.