Open AccessCommunication emitter identification based on distribution of bispectrum amplitude and phase
Author(s) -
Han Jie,
Zhang Tao,
Ren Dongfang,
Zheng Xiaoyu
Publication year - 2017
Publication title -
iet science, measurement and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.418
H-Index - 49
eISSN - 1751-8830
pISSN - 1751-8822
DOI - 10.1049/iet-smt.2017.0024
Subject(s) - bispectrum , amplitude , pattern recognition (psychology) , cluster analysis , signal (programming language) , computer science , phase (matter) , artificial intelligence , feature extraction , feature (linguistics) , identification (biology) , physics , spectral density , optics , telecommunications , linguistics , philosophy , botany , quantum mechanics , biology , programming language
This paper develops a novel method for feature extraction of steady communication signals based on the distribution of bispectrum amplitude and phase (BAP) to solve the identification problem of the same type of emitters. First, a propagation model of communication signals is modeled by analyzing the mechanism of emitter fingerprint, and the irrelevance between bispectrum amplitude and phase is demonstrated. Then, onthe basis of estimating the signal bispectrum, bispectrum symmetry is used to adopt different methods to extract features according to their own characteristics of amplitude and phase spectrum. Finally, the simulated and actual communication signals from the same type of emitters are used for the experiment. The identification performance of BAP method is compared with another three methods in terms of antinoise performance, influence of the numberof training samples, and feature distribution. Theoretical analysis and experimental results show that the BAP method overcomes the shortcomings of SIB and SB methods, and the extracted features have good clustering and interclass separability, solvingthe identification problem of the same type of emitters under low SNR and small number of training samples.