A radio frequency fingerprinting identification method based on energy entropy and color moments of the bispectrum

Abstract

Network security is a vital and essential problem in wireless communication system. There are two methods to ensure network security, one is based on bit-level credentials, and the other is based on radio frequency fingerprinting (RFF). RFF is getting more and more attention, for it is rather difficult to imitate and replicate RFF features by software. In this paper, the energy entropy and color moments of the bispectrum, as well as the support vector machine, are proposed or identifying different devices. Simulation results demonstrate that the proposed method outperforms the previous ones, especially when signal-to-noise ratio (SNR) is low. The identification accuracy achieves nearly 80% when SNR=0dB. Experiment is also conducted, further proving the effectiveness.