Statistical device simulations of III-V nanowire resonant tunneling diodes as physical unclonable functions source

Abstract

In this paper, utilising the non-equilibrium Green’s function (NEGF) formalism within the new device simulator NESS (Nano-Electronic Software Simulator) developed at the University of Glasgow’s Device Modelling Group, we present quantum mechanical simulations of current flow in double-barrier III-V GaAs-AlGaAs nanowire resonant tunneling diodes (RTDs). NESS is a fast and modular Technology Computer Aided Design (TCAD) tool with flexible architecture which can take into account various sources of statistical variability in nanodevices. The aim of this work is to show that, in the RTD devices with nano-scale dimensions, there is a direct correlation between the position and the numbers of random dopants and the key device parameters, e.g., position of the resonant-peak (VR) variations as well as the shape and number of peaks in the output current–voltage (I-V) characteristics. Such VR variability can be used as a quantum fingerprint which can provide robust security and hence can be used to deliver Physical Unclonable Functions (PUFs).