Bistatic Terahertz Radar Azimuth-Elevation Imaging Based on Compressed Sensing

Abstract

Terahertz (THz) radar imaging can achieve high range resolution for transmitting of wideband signals. High cross-range resolution can be achieved simultaneously with small rotation angles for the tiny wavelength. However, the radar echo should be sampled according to the Nyquist theory and the imaging should ensure that the unambiguous range is large than the size of targets, which limits its application from both data collection and data storage perspectives. In this paper, a bistatic THz radar azimuth-elevation imaging scheme based on compressed sensing is proposed. By exploiting the sparsity of the space distribution in the imaging plane, it is sufficient to reconstruct images of targets with a small number of random measurements within the radar aperture. This means that the limitation for the high resolution THz imaging of large targets can be overcome. The available high resolution bistatic images of targets, which show the target as it is observed, can be very useful for scattering understanding and target recognition at THz frequencies. Simulation results by the electromagnetic calculation data are presented to demonstrate the validity of the proposed method.Index Terms-Terahertz (THz) radar, bistatic radar imaging, azimuth-elevation imaging, compressed sensing, block sparse.

Three-dimensional Imaging Method of Target Based on Time-domain Sparse Representation of Multi-view SAR Data

“…After solving every sub-model d * * q = Ψ * * q • s * * q + v in (20) respectively and mapping every s * * q to a sub-3-D image, the whole 3-D image is a superimposition of all sub-images. As Q > 2, Q ≥ 2, M * * q < M , N * * q < N , and M * * q × N * * q << M × N , the dimension of Ψ * * q is much lower than Ψ and Ψ * , let alone Ψ.…”

“…Fourier imaging methods require radar to collect data over densely sampled points in both azimuth and elevation [14]. In recent years, sparse-representation based 3-D radar imaging methods have been developed and show advantages and super performance on sparsely collections [15][16][17][18][19][20][21]. However, the huge dimension of separable dictionaries, as well as the costly computation to solve them, is an obstacle for sparse representation-based methods.…”