Simulation-Based Approach to Generating Golden Data for PCB-Level Hardware Trojan Detection Using Capacitive Sensor

Reports highlight the threat of Hardware Trojans (HTs) being inserted not only into Integrated Circuits (ICs) but also into items such as Printed Circuit Boards (PCBs) and cables. This is known as PCB-level HT. There is a risk that these PCB-level HTs could be inserted even after a product is shipped, which indicates the need for lifetime monitoring. One such method for lifetime monitoring has been proposed to detect the electrical characteristics of PCBs using capacitive sensors implemented in ICs. In the HT detection method using capacitive sensors, PCB-level HT is detected by comparing the sensing results with golden data that serves as a reference. Therefore, it is required to guarantee that PCB-level HT is not inserted when generating golden data that will be used as a reference. On the other hand, it is difficult to guarantee that HT is not inserted on the PCB using actual measurements when creating golden data, so generating golden data by simulation is reasonable. In this paper, we focus on the measurement process using capacitive sensors and show that golden data can be generated by using a simple model that includes the transmission lines on the PCB to be protected from PCB-level HT without incorporating the detailed physical structure on the PCB into the model. We also confirm that PCB-level HT can be detected by using the generated golden data.