An Efficient mmW Frequency-Domain Imaging Algorithm for Near-Field Scanning 1-D SIMO/MIMO Array

Abstract

In recent years, millimeter-wave (mmW) 3D imaging technology for scanning 1D single-input multiple-output (SIMO) or multiple-input multiple-output (MIMO) array has been extensively studied due to its unique advantages in near-field applications. However, current imaging algorithms for this kind of scheme are not satisfactory, either of low quality since irrational approximations are introduced, or too slow as the back projection method is employed. In this paper, an efficient frequency-domain imaging algorithm is developed to fetch up these shortages. The precise spectral expression of the echo signal within SIMO data is derived through spherical-wave decomposition and fast Fourier transform (FFT) operations. By properly approximating the nonlinear phase, the wavefront curvature is compensated after employing a 1D Stolt interpolation, and then the reflectivity map of the target can be obtained through several multiplications and inverse FFT operations. Image reconstruction for a MIMO array can be realized by coherently summing up all the SIMO focusing results. Both simulation analysis and experimental results demonstrate the effectiveness of the proposed algorithm in imaging quality and computational efficiency.