Efficient range migration algorithm for near-field MIMO array imaging

This paper proposes an efficient range migration algorithm (ERMA) for near-field multiple-input multiple-output (MIMO) array-based microwave imaging. For the usual imaging algorithms implemented in the spatial frequency domain for MIMO arrays, the multi-dimensional Fourier transforms (FTs) are first performed with respect to transmitters and receivers, respectively, to transform the echo data in the spatial domain to those in the spatial frequency domain. After performing the matched filtering and the Stolt interpolations usually in a five-dimensional (5-D) domain for common two-dimensional (2-D) planar MIMO arrays, the high dimensional data needs to shrink into a three-dimensional (3-D) domain. The processing in a 5-D domain is time-consuming and requires huge memory storage. The proposed algorithm moves the dimension reduction in advance of the matched filtering and interpolations. To do this, we first perform data completion in respect of the undersampled data. A phase calibration method is employed to preserve precision. In so doing, the higher dimensional matched filtering and interpolations are eliminated, which results in a significant increase in imaging speed. Besides, the ERMA can enhance the resolution of the imaging results, because the dimension reduction discards the convolution between the transmit and receive spectra. Numerical simulations are demonstrated to verify the efficacy and performance of ERMA.