Trojan Detection System

Abstract

Hardware security has become an important concern in recent years. A hardware Trojan (HT) is also a hardware virus. It is a malicious modification of a circuit so it can manage, modify, disable, monitor or affect the operation of the circuit. HTs can be inserted into IC during RTL or during manufacturing, through manipulation of the layout masks and varying the doping concentration. As adversaries would want access to foundries to insert Trojans during the fabrication process, the likelihood of them being inserted at design time is far higher. it is vital to detect the hardware Trojans from a viewpoint of security. Due to the outsourcing in hardware production, malicious circuits (or hardware Trojans) are easily inserted into hardware products by attackers. Since hardware Trojans detection is difficult. Under the circumstances, numerous hardware-Trojan detection methods are proposed. During this project, it elaborates the overview of hardware-Trojan detection and reviews the hardware-Trojan detection methods using machine learning (both supervised and unsupervised learning). In supervised learning, applying KNN algorithms to resolve this problem. In unsupervised learning, applying neural networks to resolve this problem. a way to detect Hardware Trojan with the help of Deep Learning algorithm by feeding it features extracted from the gate-level netlist of the circuit. The proposed method doesn’t require any golden circuits (reference circuits) circuits. The features that are extracted from the gate-level netlist are accustomed to train the Deep learning algorithm. This method doesn’t require the simulation of the circuit so on classify genuine nodes and Trojan affected nodes.