針對平衡控制特性之硬體木馬偵測方法

Abstract

近年隨著積體電路產業的外包以及全球化，第三方供應商很容易在積體電路中植入惡意的硬體木馬，成為相關產業的重大隱患。為了避開功能驗證，硬體木馬通常在邏輯閘層網表上至少擁有一個具非常低觸發機率之觸發信號，基於這種特性，以往的研究使用不平衡的可控性作為檢測硬體木馬之特徵，假定具有不平衡可控性的信號總是伴隨著低觸發機率。然而，本論文發現了一種新型硬體木馬插入之方法，經此方法插入後的木馬具有低觸發機率但平衡的可控性，這使得當前基於不平衡可控性的偵測方式在此種情況下無法偵測到插入之木馬。為了解決這個問題，我們提出了一種基於COP的檢測方法，並使用無監督的異常分析來檢測硬體木馬。該方法不僅能成功檢測到所提出的新型硬體木馬，且能檢測Trusthub上的580個木馬電路。實驗結果顯示，本論文提出的硬體木馬檢測器優於其他檢測器，在580個測試電路上實現了整體100％的真陽性率和0.37％的假陽性率。

In recent years, the security threat posed by Hardware Trojans (HTs) has escalated to a critical level within the integrated circuits (ICs) industry. The widespread outsourcing and globalization of semiconductor design and manufacturing phases have made ICs highly vulnerable to hardware Trojan insertion by malicious third-party vendors, which in turn exposes them to significant security risks. To evade functional verification, HTs tend to include at least one trigger signal at the gate-level netlist with a very low transition probability. Earlier research has utilized imbalanced controllability as a feature to detect HTs, assuming that signals with imbalanced controllability are always accompanied by low transition probability. However, this study has found out a way to create a new type of HT that has low transition probability but balanced controllabilty, thereby thwarting the previous detection methods. Consequently, existing imbalanced controllability detectors are inadequate in this scenario. To address this limitation, we propose a COP-based detection method that uses unsupervised anomaly analysis to detect HTs. Our proposed method can effectively detect not only the proposed HT but also the 580 Trojan benchmarks on Trusthub. Empirical results demonstrate that our proposed detector outperforms other detectors, achieving an overall 100% TPR and 0.37% FPR on the 580 benchmarks.