Open AccessSeparation and feature extraction of micro‐motion signal of ballistic target
Author(s) -
Li Yuxi,
Feng Cunqian,
Xu Xuguang,
Han Lixun,
Wang Dayan
Publication year - 2021
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/tje2.12083
Subject(s) - signal (programming language) , computer science , feature extraction , feature (linguistics) , basis (linear algebra) , convergence (economics) , artificial intelligence , process (computing) , independent component analysis , separation (statistics) , pattern recognition (psychology) , trajectory , algorithm , computer vision , mathematics , physics , machine learning , philosophy , linguistics , geometry , astronomy , economics , programming language , economic growth , operating system
During the process of the ballistic target's mid‐flight, it is very important to accurately identify the target. The separation of target micro‐Doppler curves and the extraction of characteristic parameters are the key to accurate identification. Aiming at the slow convergence speed of the traditional Fast‐ICA algorithm, this paper proposes an improved Fast‐ICA algorithm, which realizes the separation of the micro‐Doppler curve of the scattering points. On this basis, the trajectory of scattering points in space is analyzed, the expression of the target feature value is established, and the target feature parameters are effectively extracted. According to the results of the simulation experiment, the algorithm can achieve better signal separation and the characteristic parameters of the target are effectively extracted.