During the System-on-chip (SoC) design process, untrusted intellectual property (IP) cores are purchased from various external vendors in the form of soft-IPs (RTL code or gate-level netlist). While an adversary at the third-party IP vendor’s facility could introduce a hidden malicious functionality within the RTL code or gate-level netlist, it is extremely hard to verify the trust of a hardware design since the IP integrator does not have a golden model to compare with. We have developed a trust verification framework that takes the untrusted IP as an input and identifies a list of suspect nets that could be part of a hardware Trojan.
Hoque, T., Cruz, J., Chakraborty, P., & Bhunia, S. (2018). A Systematic Machine Learning Framework for IP Trust Verification. In SRC TECHCON.
Hoque, T., Cruz, J., Chakraborty, P., & Bhunia, S. (2018). Hardware IP Trust Validation: Learn (the Untrustworthy), and Verify. In 2018 IEEE International Test Conference (ITC).
Hoque, T., Mishra, P., & Bhunia, S. (2017). A Systematic Feature Selection Methodology for Machine Learning Based Hardware Trojan Detection. In SRC TECHCON.