Hirschman optimal transform based correlation frequency electromagnetic analysis
Zhu Wang, Xinping Zhou, Victor DeBrunner, Weiqing Huang, Degang Sun, Yan Wang, Changhai Ou, Juan Ai · 2016
Correlation Electromagnetic analysis (CEMA) has been effective in revealing the cryptographic key on cryptosystems. Random delay insertion (RDI) causes misalignments to prevent the action of these attacks in the time domain to avoid information leakage. In this paper, we first use the newly proposed time-frequency transformation Hirschman Optimal Transform (HOT) to transform the signal from the time domain to the frequency domain for analysis. Experimental results show that this method conquers the weakness in the time domain analysis in which the samples have to be aligned accurately. So, our proposed method can be used to break cipher chips with random delay. We find that our method is more efficient than a similar approach based on the DFT.