Node-wise Hardware Trojan Detection Based on Graph Learning

In the fourth industrial revolution, securing protection of supply chains has become an ever-growing concern. One such cyber threat is a hardware Trojan (HT), malicious modification to IC. HTs are often identified during manufacturing process but should be removed earlier in design process. Machine learning-based HT detection gate-level netlists efficient approach identifying at early stage. However, feature-based modeling limitations terms discovering appropriate set features. We thus propose NHTD-GL this paper, novel node-wise method based on graph learning (GL). Given formal analysis features obtained from domain knowledge, bridges gap between representation and detection. The experimental results demonstrate that achieves 0.998 accuracy 0.921 F1-score outperforms state-of-the-art methods. extracts without heuristic feature engineering.