NASCTY: Neuroevolution to Attack Side-Channel Leakages Yielding Convolutional Neural Networks

Side-channel analysis (SCA) is a class of attacks on the physical implementation cipher, which enables extraction confidential key information by exploiting unintended leaks generated device. In recent years, researchers have observed that neural networks (NNs) can be utilized to perform highly effective SCA profiling, even against countermeasure-hardened targets. This study investigates new approach designing NNs for SCA, called neuroevolution attack side-channel traces yielding convolutional (NASCTY-CNNs). method based genetic algorithm (GA) evolves architectural hyperparameters automatically create CNNs analysis. The findings this research demonstrate we achieve performance results comparable state-of-the-art methods when dealing with desynchronized leakages protected masking techniques. indicates employing similar neuroevolutionary techniques could serve as promising avenue further exploration. Moreover, similarities among constructed shed light how NASCTY effectively constructs architectures and addresses implemented countermeasures.