Deep Learning Based Side-Channel Analysis for Lightweight Cipher PRESENT

The lightweight cipher is attracting attention as cryptographic technique for IoT devices. PRESENT, the target of this study, which is standardized in ISO/IEC29192-2 and it is one of the most popular lightweight ciphers. On the other hand, the risk of side-channel analysis (SCA) has been reported as a security issue for cryptographic circuits. Moreover, the deep learning based SCA (DL-SCA) has recently been reported as a more powerful analysis method. However, DL-SCA for PRESENT with hardware implementation has not been reported so far. Therefore, this study proposes DL-SCA for the lightweight cipher PRESENT. The experimental results using FPGA show that all round keys can be successfully estimated by using 300 waveforms, and that the proposed method can be analyzed with a smaller number of waveforms than the conventional SCA method CPA.