Tagging using radio-frequency identification (RFID) is getting increasing adoption throughout the supply chains of commercial products. The megahertz (MHz) and gigahertz (GHz) bands currently used for RFID will soon become congested, especially considering the interference from simultaneous interrogations of many co-located RFID devices and the limited bandwidth allocated for RFID. In addition, limited by their off-chip components for wireless powering and data transmission, current RFID devices have relatively high packaging cost and large form factor. The low available power also prevents the usage of high-security algorithms and hardware on the RFID devices. All these problems have hindered the effectiveness and application of the technology in areas like authentications of medicines, semiconductor chips, banknotes, etc. This project will apply low-terahertz (~0.3 THz in this project) wireless powering and back-scattering circuit techniques to explore a new electromagnetic spectrum for future RFID tagging. The 1000 times larger bandwidth and the highly-collimated downlink and uplink beams will accommodate orders of magnitude more tags than what we have today. The new THz RFID device realized on a semiconductor chip will also have a package-less ultra-small form, leading to dramatic cost reduction and wide range of applications. The proposal will also develop energy-efficient authentication hardware and algorithms for the proposed THz RFID chip to secure the storage and transmission of sensitive data related to, for example, financial and biometric information. The proposed research will be tightly integrated with a few outreach activities, including a SuperUROP undergraduate research program, a "Tagging the World" lecture series, an internship through MIT Summer Research Program (MSRP), a career workshop at MIT Rising Star, and a few "T-Ray Lab Day" events for K-12 students.
This project will utilize custom-designed, ultra-high-speed Schottky-barrier diodes to perform practical THz power rectification, with an expected THz-to-DC conversion efficiency of up to 50%. A novel antenna array, which is co-designed with the diode and adopts a multi-functional electromagnetics methodology, will also be developed to simultaneously perform large-aperture wireless powering and back-scattering-based data communication. In addition, low-energy authentication protocols and hardware will be applied to the system. In particular, the RFID chip will be equipped with a side-channel resistant elliptic curve processor, which uses dynamic voltage-frequency scaling (DVFS) to reduce its DC power to microwatt level. Novel CMOS-compatible memory for secret-key storage, as well as beam-steering of the RFID tag output wave for anti-eavesdropping, will also be implemented in the chip. The package-less and battery-less RFID chip is expected to have a volume smaller than one tenth of a cubic millimeter. Lastly, using high-speed indium phosphide (InP) heterojunction bipolar transistors (HBTs), the project will innovate a THz interrogator transceiver to pair with the RFID chips. With a scalable array architecture and high-versatility transceiver front-end circuit units, the interrogator will be able to radiate more than 100 mW of power with highly focused beam. The proposed approach will advance the utilization efficiency and diversity of electromagnetic spectrum, as well as the end-to-end solutions for secured wireless data authentication. The project will also generate new design methodologies for both high-frequency and high-security integrated circuits, especially under power-constrained conditions.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
