Tamper Resistance Evaluation of Midori128 against Differential Power Analysis

The use of lightweight cryptography enhances the security of the Internet of Things (IoT) devices. Midori is one of the typical lightweight ciphers. Although the cryptographic algorithm is computationally secure, the risk of side-channel analysis (SCA) has been pointed out. For tamper resistance evaluation, it is important to evaluate various power analysis methods. Midori128 has been evaluated using correlation power analysis and deep learning-based power analysis. However, no evaluation has been reported for differential power analysis (DPA) though it is a typical power analysis method. Therefore, we proposed a DPA method of Midori128 for tamper resistance evaluation. In the experiment, Midori128 was implemented in a field programmable gate array. Then, we measured the power consumption waveforms and evaluated the tamper resistance using actual measured waveform data. As a result, the proposed method was used to successfully estimate more than 93% of secret keys by using 5,000 waveforms and showed that the tamper resistance of Midori128 can be quantitatively evaluated.