Research Article
Hardware Trojan Detection Using XGBoost Algorithm for IoT with Data Augmentation Using CTGAN and SMOTE
@INPROCEEDINGS{10.1007/978-3-030-79276-3_10, author={C. G. Prahalad Srinivas and S. Balachander and Yogesh Chandra Singh Samant and B. Varshin Hariharan and M. Nirmala Devi}, title={Hardware Trojan Detection Using XGBoost Algorithm for IoT with Data Augmentation Using CTGAN and SMOTE}, proceedings={Ubiquitous Communications and Network Computing. 4th EAI International Conference, UBICNET 2021, Virtual Event, March 2021, Proceedings}, proceedings_a={UBICNET}, year={2021}, month={7}, keywords={Hardware Trojan SMOTE Machine learning CTGAN XGBoost}, doi={10.1007/978-3-030-79276-3_10} }- C. G. Prahalad Srinivas
S. Balachander
Yogesh Chandra Singh Samant
B. Varshin Hariharan
M. Nirmala Devi
Year: 2021
Hardware Trojan Detection Using XGBoost Algorithm for IoT with Data Augmentation Using CTGAN and SMOTE
UBICNET
Springer
DOI: 10.1007/978-3-030-79276-3_10
Abstract
Internet of things is the next blooming field in engineering. Seamless connectivity has become the new normal when devices are manufactured. This demand for IoT (Internet of Things) also results in various threats. Hardware trojans are one such threat which is encountered. Hardware trojan, being sneak circuits are overlooked when a new chip is manufactured. This further solidifies the threat posed by the hardware trojans. Applying machine learning model directly to the dataset will prove futile, since the amount of trojan infected nets are very minimal compared to the regular nets. In this work, this problem is addressed by creating a machine learning model, which can handle the data imbalance, using CTGAN (Conditional Tabular Generative Adversarial networks) and SMOTE (Synthetic Minority Oversampling Technique) algorithms, and detect the trojan infected nets.
