Hybrid Denoising Based Correlation Power Analysis for AES

The correlation power analysis is greatly affected by the noise in the power data. To address this issue, this paper proposes a hybrid denoising-based correlation power analysis method. Power consumption data of AES-128 algorithm in DPAContest dataset are used. The data is first preprocessed by Butterworth low-pass filter and then fed to Empirical Mode Decomposition for decomposition. Wavelet threshold denoising method is used to further remove noise in the high frequency part. By attacking the filtered data, the hybrid denoising-based correlative power analysis method is compared to the correlative power analysis with only Butterworth filter and the common correlative power analysis method without filtering. Experimental results show that the hybrid denoising method has better performance.