Counterfeit electronics are a prevailing supply chain issue, which have continued to become more widespread in recent years. Critical systems have long life cycles and diminishing manufacturing source issues, making them a prime target for economically motivated counterfeiters. Counterfeit electronics create risks and life-threating issues, especially when incorporated into such critical systems. Since most of the chips in the market today are not equipped with mechanisms to aid in counterfeit detection, physical inspection is the most relied upon approach. During physical inspection, defects related to the exterior, interior, and material composition are extracted from the component under test, and these anomalies are used to determine if it is counterfeit. Physical inspection does not rely on electrical functionality, making it widely applicable to the diverse range of electronic chips in the market (analog, digital, mixed signal, large, small, etc.)
There are many longstanding challenges associated with physical inspection. First and foremost, in order to evade detection counterfeiters evolve and improve their methods over time. This can result in new defects and/or changes in the frequency of the defects found in counterfeit parts. There are few (if any) initiatives aimed at tracking such changes, making it difficult to develop accurate assessment of risk. Second, counterfeit testing is typically performed by subject matter experts (SMEs), which is time consuming, results in inconsistent/subjective results, and is difficult to adapt. Automated approaches that could relieve these issues need to be developed. Third, there are a diverse set of counterfeit IC types within the supply chain, all of which have their own unique defects that require a gamut of expensive tests/equipment to detect. Any research that could reduce the time and cost associated with physical inspection would be an immense help.
The main barrier towards overcoming the above challenges is a lack of data. Developing and improving counterfeit detection through physical inspection requires images and measurements from a variety of counterfeit ICs. In general, researchers in academia do not have access to the state-of-the-art equipment, counterfeit samples, and experience necessary to perform physical inspection on suspect parts. On the other hand, the test labs with access do not have the time, resources, and motivation to devote to exhaustive data collection. Our goal in developing this repository is to provide the community with access to such data. To this end, the team at UF/ FICS Research has spent countless hours collecting measurements and will continue to maintain Counterfeit-IC.org over the course of this NSF sponsored project. We hope that researchers will take advantage of this resource to push the boundaries of counterfeit research as well as to share their own data with the rest of the community.