Visible to the public Designing Subspecies of Hardware Trojans and Their Detection Using Neural Network Approach

TitleDesigning Subspecies of Hardware Trojans and Their Detection Using Neural Network Approach
Publication TypeConference Paper
Year of Publication2018
AuthorsInoue, T., Hasegawa, K., Kobayashi, Y., Yanagisawa, M., Togawa, N.
Conference Name2018 IEEE 8th International Conference on Consumer Electronics - Berlin (ICCE-Berlin)
KeywordsArtificial neural networks, Benchmark testing, Design Time, electric devices, electronic engineering computing, feature extraction, gate-level netlist, gradient methods, Hardware, hardware devices mass-production, hardware trojan, hardware trojan subspecies designing, high-performance hardware device, integrated circuit design, invasive software, learning (artificial intelligence), Logic gates, machine learning, machine-learning-based hardware-Trojan detection method, neural nets, Neural Network, neural network approach, neural-network based hardware-Trojan detection method, outsourcing, pubcrawl, third-party vendors, TPR, trigger circuits, trojan horse detection, Trojan horses, true positive rate

Due to the recent technological development, home appliances and electric devices are equipped with high-performance hardware device. Since demand of hardware devices is increased, production base become internationalized to mass-produce hardware devices with low cost and hardware vendors outsource their products to third-party vendors. Accordingly, malicious third-party vendors can easily insert malfunctions (also known as "hardware Trojans'') into their products. In this paper, we design six kinds of hardware Trojans at a gate-level netlist, and apply a neural-network (NN) based hardware-Trojan detection method to them. The designed hardware Trojans are different in trigger circuits. In addition, we insert them to normal circuits, and detect hardware Trojans using a machine-learning-based hardware-Trojan detection method with neural networks. In our experiment, we learned Trojan-infected benchmarks using NN, and performed cross validation to evaluate the learned NN. The experimental results demonstrate that the average TPR (True Positive Rate) becomes 72.9%, the average TNR (True Negative Rate) becomes 90.0%.

Citation Keyinoue_designing_2018