A few years ago, researchers at MIT developed a cryptographic ID tag that is smaller and more cost-effective than traditional radio frequency tags (RFIDs) used to verify product authenticity.
This new tag, which provides enhanced security compared to RFIDs, utilizes terahertz waves that are smaller and faster than radio waves. However, this terahertz tag had a security flaw similar to RFIDs – a counterfeiter could remove the tag from a genuine item and attach it to a fake one without detection by the authentication system.
To address this vulnerability, the researchers used terahertz waves to create an antitampering ID tag that maintains the advantages of being small, affordable, and secure.
By incorporating microscopic metal particles into the glue that adheres the tag to an object, the researchers can use terahertz waves to detect the unique pattern formed by these particles on the item’s surface. This random pattern acts as a fingerprint to authenticate the item, as explained by Eunseok Lee, an electrical engineering and computer science graduate student and lead author of the paper on the antitampering tag.
The researchers developed a light-powered antitampering tag measuring about 4 square millimeters and created a machine-learning model to identify tampering by matching similar glue pattern fingerprints with over 99% accuracy.
Due to its low production cost, the terahertz tag could be implemented across a large supply chain. Its small size also allows it to be attached to items too small for traditional RFIDs, such as certain medical devices.
The research, to be presented at the IEEE Solid State Circuits Conference, is a collaboration between Han’s group and the Energy-Efficient Circuits and Systems Group led by Anantha P. Chandrakasan, MIT’s chief innovation and strategy officer, dean of the School of Engineering, and the Vannevar Bush Professor of EECS. Co-authors include graduate students Xibi Chen, Maitryi Ashok, and Jaeyeon Won.
Preventing tampering
This project was inspired by Han’s experience at his favorite car wash, where an RFID tag was attached to his windshield for car wash membership authentication. To enhance security, the tag was made of fragile paper that would be destroyed if removed and reattached to another windshield.
Instead of authenticating the tag, the researchers focused on authenticating the item itself by targeting the glue at the interface between the tag and the item’s surface.
The antitampering tag contains tiny slots that allow terahertz waves to pass through and interact with microscopic metal particles mixed into the glue. Terahertz waves are sensitive enough to detect these particles, unlike larger radio waves. A chip with multiple slots enables waves to capture more information on the distribution of particles, making it difficult to replicate the response if the glue interface is compromised.
A vendor would initially read the antitampering tag once attached to an item and store the data in the cloud for verification.
AI for authentication
Lee encountered difficulties in taking precise measurements to determine glue pattern matches for testing the antitampering tag. With assistance from a colleague in the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), they developed a machine-learning model to compare glue patterns with over 99% accuracy.
While the authentication system has limitations due to terahertz wave transmission loss, the researchers are working to improve the system and overcome technical challenges to unlock the potential of terahertz waves for ID, security, and authentication applications.
This research is supported by the U.S. National Science Foundation and the Korea Foundation for Advanced Studies.
