Variational mode decomposition based denoising in side channel attacks
Juan Ai, Zhu Wang, Xinping Zhou, Changhai Ou · 2016
In this paper, we present two kinds of noise reduction methods for improving the attack performance in side channel attacks (SCA). One is wavelet transform combined with variational mode decomposition (VMD-WT), the other is an iterative variation mode decomposition combined with wavelet transform (IVMD-WT). We test them both on simulated piece-regular signal and practical leakages including software implementation and hardware implementation of AES. Results of all experiments show that the proposed methods outperform widely used wavelet transform and can improve the success rate of correlation power analysis (CPA) significantly. The proposed methods not only are more suitable to general condition but also are superior to some existing methods from the view point of usability and efficiency for denoising.