A Fast Fourier-Based Near-Field 3-D Imaging Algorithm for MIMO Array

Multiple-input–multiple-output (MIMO) array near-field imaging technology, due to its high resolution, has great potential for development in the fields of nondestructive testing, medical imaging, and security inspection. However, currently available imaging algorithms present challenges, as they either require excessive hardware processor memory capacity or lead to prolonged imaging processing time. In this article, a fast Fourier-based near-field 3-D imaging algorithm for the MIMO array is proposed, and it requires less hardware processor memory capacity. Based on the feature that the target image can be obtained by integrating the data corresponding to all transceiver arrays, the data corresponding to all transceiver elements are split into the data corresponding to individual transmitting elements in a targeted manner and then imaged. After that, the final 3-D image is obtained by integrating the images corresponding to all transmitting elements. It effectively reduces the requirement for hardware processor memory capacity. The horizontal and vertical dimensional Fourier transforms (FTs) of all receiving element data corresponding to a single transmitting element are performed, which converts the spatial-domain data to the wavenumber domain. It effectively improves the computational efficiency. At the same time, the time-domain data are converted to the frequency domain by FT, and the phase compensation can be performed only on the peak data corresponding to the target. Therefore, it also effectively improves the computational efficiency. Simulation and experimental results show that compared with the commonly used algorithms, the proposed algorithm can not only obtain similar image quality but also has higher computational efficiency.