On application of one-class SVM to reverse engineering-based hardware Trojan detection

TLDR

Hardware Trojans, introduced through outsourced IC design and fabrication, pose significant security risks, prompting research into detection methods that often rely on a golden model derived from reverse‑engineered Trojan‑free chips. This work proposes a robust reverse‑engineering method to identify Trojan‑free integrated circuits. The method employs a one‑class support vector machine to learn the normal behavior of ICs from reverse‑engineered samples. Simulations on publicly available circuits demonstrate that the approach achieves high accuracy across various modeling and algorithm parameters.

Abstract

Due to design and fabrication outsourcing to foundries, the problem of malicious modifications to integrated circuits known as hardware Trojans has attracted attention in academia as well as industry. To reduce the risks associated with Trojans, researchers have proposed different approaches to detect them. Among these approaches, test-time detection approaches have drawn the greatest attention and most approaches assume the existence of a "golden model". Prior works suggest using reverse-engineering to identify such Trojan-free ICs for the golden model but they did not state how to do this efficiently. In this paper, we propose an innovative and robust reverseengineering approach to identify the Trojan-free ICs. We adapt a well-studied machine learning method, one-class support vector machine, to solve our problem. Simulation results using state-of-the-art tools on several publicly available circuits show that our approach can detect hardware Trojans with high accuracy rate across different modeling and algorithm parameters.