Identification of wireless transceiver devices using radio frequency (RF) fingerprinting based on STFT analysis to enhance authentication security

Wireless transceiver devices radio frequency (RF) fingerprinting (RFF) distinct native feature is a physical layer based security authentication method that enhances wireless security authentication, which is especially useful for the power and computing limited devices. However, the high authentication rate is necessary for the security purpose. In this paper, we propose a new approach where the radio frequency finger-printing is based on the short time Fourier transform (STFT) analysis. Specifically, in order to identify four wireless transceiver devices of same manufacture and batch, we combine the STFT with the support vector machine (SVM) algorithms. Our simulation results show that the average correct identification probability is higher than 80.5% even when the signal-to-noise ratios (SNR) is 0 dB. It is demonstrated that the novel method has high capability to against the noise.