Modern society, revolutionized by the Internet of Things (IoTs), is witnessing exponential growth in the number of connected devices and the volume of data being generated and shared, raising significant concerns about safeguarding classified information against various cyber threats. Here, we introduce a lightweight, robust hardware security primitive based on the electromagnetic physical unclonable function (PUF) for cryptographic identification and authentication of wireless devices. Unlike traditional digital-based PUFs, the proposed electromagnetic PUF keys are generated using graphene-based harmonic transponders, of which the inherent variations in electronic properties of ambipolar graphene field-effect transistors (GFETs) result in highly stochastic, mixed modulations of radio frequency (RF) signals (i.e., unique electromagnetic fingerprints). Our experimental results demonstrate that this electromagnetic PUF exhibits excellent PUF performance metrics in terms of randomness, uniqueness, reliability, and resistance to machine learning-based modeling attacks. Moreover, the PUF keys can be reconfigured by altering the RF excitation frequency or through the electrostatic gating effect, further strengthening the security and resilience against modeling attacks. The proposed electromagnetic PUF may be well-suited for a variety of wireless authentication, encryption, and anticounterfeiting applications, and supports cryptographic key generation.
Keywords: graphene field-effect transistors (GFET); hardware security; low-dimensional nanomaterials; physical unclonable function (PUF); radio frequency (RF) oscillator.