Fourier-Based Near-Field Three-Dimensional Image Reconstruction in a Multistatic Imaging Structure Using Dynamic Metasurface Antennas

Due to physical layer compression, and consequently, the failure to produce uniform radiation patterns, it is not possible to apply fast Fourier-based image reconstruction algorithms to raw measurements collected from coded metasurface antennas in computational imaging. An effective solution in the literature is a sub-wavelength sampling of the aperture. However, this solution is currently limited to a panel-to-probe model which requires a mechanical raster scan. On the other hand, existing works are based on time-division multiplexing. This means that only a single transmit/receive channel is active in each time slot. In this paper, we introduce a panel-to-panel model in a multistatic structure. Based on this model, two pre-processing (for two different individually measured signal and combined measured signal (CMS) scenarios) are derived to convert the raw measurements into the space-frequency domain. Then, by using the output data from the pre-processing stage and according to the configuration of the introduced imaging system, the range migration algorithm is adapted. The importance of the proposed solution for the CMS scenario is that, for the first time, it demonstrates the capability of using dynamic metasurface antenna diversity to achieve simultaneous data acquisition. The performance of the proposed approach is compared with state-of-art works in terms of reconstructed image quality and computational complexity using numerical simulations and analytical discussions.