FPGA Hardware Trojan Detection: Golden-Free Machine Learning Approach

Ensuring trust in the semiconductor IC supply chain necessitates the critical detection of Hardware Trojans, yet current methods relying on side-channel analysis often require the use of golden chips for verification. This research paper presents a novel approach to detect Hardware Trojans in the semiconductor IC supply chain, addressing the need for trust and eliminating the use of golden chips. By combining unsuper-vised machine learning and side-channel analysis, the proposed technique leverages unique features from on-chip ring-oscillator networks to identify anomalies through unsupervised clustering. Evaluation on FPGA chips with Trojan insertion demonstrated exceptional accuracy, surpassing alternative methods with a 99 % accuracy rate. The centroid-based clustering model exhibited superior performance with a slight edge in false positive rate and an fl score. This research contributes to enhancing trust in semiconductor IC supply chains by offering a fresh perspective on Hardware Trojan detection.