Detecting Hardware Trojans in IoT Devices Using Semi-Supervised Learning with XG-Boost Classifier

The increasing rate of hardware Trojans poses critical security risks to chips and integrated circuits (ICs) used in high precedence applications. Conventional detection methods often fall short due to the growing complexity of Trojans and the unfeasibility of extensive testing. This study introduces a comprehensive approach to extract key features from netlist data, such as connectivity, levels, fan-in, fan-out, and counts of flipflops/multiplexers, essential for identifying hardware Trojans. Principal Component Analysis (PCA) was applied to reduce feature dimensions while preserving 95% variance, and permutation importance was utilized to determine feature importance. Furthermore, Label Spreading was employed to make use of unlabeled data, improving the training dataset and model performance. To ensure optimal selection of hyperparameters and accurate net predictions, Grid-Search CV was used to fine-tune an XG-Boost classifier. By combining supervised and unsupervised learning, our method effectively detects differences that indicate the presence of hardware Trojans with minimal labeled data. The effectiveness and viability of semi-supervised hardware Trojan detection were highlighted by the suggested approach’s average true positive rate of 91.7% and true negative rate of 94.77% when evaluated on the Trust-HUB benchmarks.