EMD-Based Denoising for Side-Channel Attacks and Relationships between the Noises Extracted with Different Denoising Methods
In essence, side-channel leakages produced during the execution of crypto implementations are noisy physical measurements. It turns out that various noises contained in leakages have, in general, negative effects on the key-recovery efficiency of side-channel attacks. Therefore, in practice, frequency-based denoising methods are presented and in wide use nowadays. However, most of them for reducing noises of high-frequency are not always effective, and they sometimes do little or even no help. On the other hand, the relationship between noises extracted with different denoising methods that target different frequencies, in time-domain, is not being discussed, which in turn will determine the potential power of combining these denoising methods. Motivated by this, we present two empirical mode decomposition (EMD) based denoising methods for side-channel attacks, and study their effectiveness in reducing noises of high frequency in real power traces. Compared with their counterparts, EMD-based denoising methods achieve both effectiveness and stability. Furthermore, we investigate the relationships between the noises extracted with denoising methods that target different frequencies, by performing attacks on real power traces denoised by multiple combinations of different denoising methods. For this purpose, we define the notion of overlapping coefficient, which measures how much that noises are overlapped with each other. Our results and observations are evidently verified by correlation power analysis attacks on multiple real power traces sets.
KeywordsSide-channel Cryptanalysis Correlation Power Analysis Empirical Mode Decomposition Noise Reduction Overlapping Coefficient
Unable to display preview. Download preview PDF.
- 3.Charvet, X., Pelletier, H.: Improving the DPA attack using Wavelet transform. In: Non-Invasive Attack Testing Workshop 2005 (2005)Google Scholar
- 4.Souissi, Y., Aabid, M., Debande, N., Guilley, S., Danger, J.: Novel Applications of Wavelet Transforms based Side-Channel Analysis. In: Non-Invasive Attack Testing Workshop 2011 (2011)Google Scholar
- 5.Cao, Y., Zhou, Y., Yu, Z.: On the Negative Effects of Trend Noise and Its Applications in Side-Channel Cryptanalysis, http://eprint.iacr.org/2013/102.pdf
- 10.Cohen, L.: Time-Frequency Analysis. Prentice- Hall, Englewood Cliffs (1995)Google Scholar
- 11.Flandrin, P., Rilling, G., Goncalves, P.: EMD equivalent filter banks, from interpetation to applications. In: Huang, N.E., Shen, S. (eds.) Hilbert-Huang Transform and Its Applications, 1st edn. World Scientific, Singapore (2005)Google Scholar
- 13.Kopsinis, Y., McLaughlin, S.: Development of EMD-Based Denoising Methods Inspired by Wavelet Thresholding. IEEE Transactions on Signal Processing 57(4) (April 2009)Google Scholar
- 14.Mangard, S., Oswald, E., Popp, T.: Power Analysis Attacks: Revealing the Secrets of Smart Cards. Springer (2007)Google Scholar
- 15.Kocher, P.C.: Timing Attacks on Implementations of Diffie-Hellman, RSA, DSS, and Other Systems. In: Koblitz, N. (ed.) CRYPTO 1996. LNCS, vol. 1109, pp. 104–113. Springer, Heidelberg (1996)Google Scholar