A Feature Extraction Method for Hardware Trojans Detection
Zhixun Zhao, Lin Ni, Shaoqing Li, Yubo Shi · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2015
Recently Hardware Trojan has been widely studied for its threat to IC security.The method based on power side-channel information is an effective one in various Hardware Trojan detection methods proposed at present.However, there is a problem that Hardware Trojan power is difficult to be distinguished and easy to be drowned by the noise if using the power information to recognize Hardware Trojan through direct power difference during detection process.On the base of power component analysis, this paper proposes that we can extract the feature character of power information based on the feature difference between Hardware Trojan power and noise to manifest the power caused by Hardware Trojan.Then it gives out the power feature extraction algorithm and Trojan power identification model.In the experiment, the AES circuit is used as attack carrier embedded into the hardware Trojan circuit, then simulating the power, the results show that the feature extraction algorithm and convergence and divergence identification model can detect Hardware Trojans effectively.