Fluctuation‐Enhanced True Random Number Generators and Physical Unclonable Functions in a Synaptic Hybrid Ferro‐Ionic Tunnel Diode

ABSTRACT

Secure edge hardware demands compact device platforms that can integrate low‐power information processing with hardware‐level security. However, this convergence of computation and security remains challenging because accurate analog operation and stochastic security functionality rely on fundamentally different device characteristics. Here we demonstrate that transient fluctuations in a synaptic hybrid ferro‐ionic tunnel diode (HTD) can be harnessed as an in materia physical resource for unified hardware security, enabling both true random number generators (TRNGs) and physical unclonable functions (PUFs) within the same device platform. To achieve this functional separation, low‐frequency noise is introduced as a new degree of freedom for hardware security, serving either as a direct stochastic source or as a guide to the optimal operation condition. Therefore, the HTD supports both static identity and dynamic randomness in distinct operating regimes, while separating information processing in the conductance domain from security generation in the fluctuation domain. In the TRNG mode, the tunneling regime activates shot noise‐dominant fluctuations that provide a broadband entropy source with enhanced thermal stability for TRNG operation. In the PUF mode, voltage‐driven stochastic switching produces device‐specific power spectral density fingerprints, overcoming the limited dimensionality of conventional conductance‐based fingerprints.