Graf Research OpTrust HTH Benchmark

v1.2026.1

Graf Research Corporation is making a robust Hardware Trojan Horse (HTH) benchmark set available to select parties.  This benchmark enables production of statistically relevant empirical metrics for HTH detection methods.  Production of statistically relevant Probability of Detection (Pd) and Probability of False Alarm (Pfa) metrics requires evaluation of methods against a benchmark consisting of a sufficient quantity and diversity of both Trojans and circuit designs that both do and do not contain those Trojans.  The GRC HTH Benchmark contains a dataset that is large, as defined by statistical relevancy, and representative of populations of both circuit designs and Trojans, as expressed through coverage of varying ontologies.

The GRC HTH Benchmark set includes a repository of attacks, comprised of circuit designs that function as HTH activation mechanisms (triggers) and other circuits that produce desired malicious effects (payloads), used in combination with golden designs to produce exploited designs. The combination of golden designs and exploited designs with included attacks forms a viable benchmark dataset for determining the efficacy of HTH detection-based countermeasures. The 200 included golden circuit designs are selected to ensure coverage of design ontologies and the 12,296 HTH circuit designs are selected to ensure coverage of HTH ontologies.  Ontologies categorizing the designs are included in the distributed dataset along with an HTML-based overview to enable ready understanding of the collection.  Using this information, statistically relevant test sets can be sampled from the benchmark to guarantee coverage of a sufficiently broad set of designs and HTHs.

Please use the following citation for the GRC HTH Benchmark when it has contributed to your work:

  • J. Koiner, K. Paar, F. Archuleta, J. Graf, S. Harper, W. Batchelor, A. Cook, and M. Winslow, “Graf Research OpTrust Hardware Trojan Horse Benchmark v1.2026.1,” v. 1.2026.1, Graf Research, 2026. [Online]. Available: www.grafresearch.com/optrust-hth-benchmark-120261.

Register here for access to the benchmark

If you have any questions about accessing the benchmark, reach out to info@grafresearch.com.

References

  1. “Developing Hardware Trojan Benchmarks for Testing Bitstream-level Detection Methods,” F. Archuleta, J. Koiner, A.A. Sohanghpurwala, W. Batchelor, J. Graf, S. Harper, International Test Conference, October 2026.

  2. "Generating Statistically Relevant Trojan Benchmarks for Microelectronics Quantifiable Assurance," M. Winslow, W. Bachelor, J. Koiner, K. Paar, S. Harper, and J. Graf, GOMAC 2023. Link to PDF.

  3. “A practical application of game theory to optimize selection of hardware Trojan detection strategies,” J. Graf, W. Batchelor, S. Harper, R. Marlow, E. Carlisle, and P. Athanas, HaSS Journal, December 2019. Link to Springer website.