Efficient Wideband MRCS Simulation for Radar HRRP Target Recognition Based on MSIB and PCA

摘要

In this paper, efficient wideband monostatic radar cross-section (MRCS) simulation is presented for radar high range resolution profile (HRRP) target recognition. Firstly, an efficient numerical approach is proposed for the wideband MRCS. The well-conditioned integral equation and the higher-order hierarchical divergence-conforming vector basis functions are utilized for the scattering field. The adaptive cross approximation (ACA) based matrix compression method is applied for efficient analysis of MRCS at a specific frequency point. The geometric theory of diffraction (GTD) based scattering model is utilized for MRCS over a wide frequency band. Secondly, the radar HRRP target identification is performed by using principal component analysis (PCA) on modified surrounding-line integral bispectrum (MSIB). The HRRP of target can be obtained by inverse fast Fourier transform (IFFT) of the spectral domain backscattering field within a certain frequency range. The one-dimensional MSIB features of HRRP are extracted to constitute eigenvectors for radar target recognition. To enhance the separation ability of radar target recognition, the MSIB is projected onto PCA space before recognition. Numerical examples prove that the proposed algorithm is feasible and efficient.