Correlation Power Analysis with Companding Methods

AbstractCompanding methods have been profoundly applied in signal processing for quantization. And various companding schemes have been proposed to improve the PAPR (Peak to Average Power Ratio) of OFDM systems. In this paper, based on the exploration of the features of μ-law functions, we propose Correlation Power Analysis (CPA) with μ-law companding methods. μ-law expanding function is used to preprocess the power traces collected during AES encryption on ASIC and FPGA respectively. Experiments show that it reduces the number of power traces to recover all the key bytes as much as 25.8% than the conventional CPA