High-Mobility Emitter Identification Framework Using RF-Fingerprints in Orthogonal Time-Frequency Space System

Abstract

Next-generation wireless systems face unique challenges in identifying high-mobility emitters/vehicles (HME/V), especially in the delay-Doppler domain, where existing technologies fall short. The orthogonal time-frequency space (OTFS) technique has recently been proposed to deal with challenges in the delay-Doppler domain. Hence, for the first time, this letter introduces a novel emitter identification (EI) framework using radio-frequency (RF) fingerprints of HME/V using OTFS. A convolutional neural network (CNN) is exploited to extract the RF fingerprints from in-phase and quadrature-phase (IQ) components of the received OTFS-modulated signal to identify different emitters accurately. Further, we also analyze the impact of the same/different set of digitally modulated baseband signals such as binary phase shift keying (BPSK), on-off keying (OOK), 4-amplitude shift keying (4ASK), and 8-amplitude shift keying (8ASK) within the OTFS, employed by the emitters. Experimental results depict the efficacy of the proposed EI framework for HME/V even under low signal-to-noise ratios (SNRs).