MIMO-SAR Tomography

Abstract

MIMO SAR employs multiple transmit and receive channels to improve the imaging performance and to acquire novel geo-information products. One example is SAR tomography, where the simultaneous transmission and reception with multiple antennas can provide a large number of baselines with a small number of antennas. In the limit, an appropriately designed MIMO-SAR configuration with N_Tx transmitters and N_Rx receivers can provide in total N_Tx∙N_Rx independent phase centers and therefore N_Tx∙N_Rx-1 independent baselines for SAR tomography. The other extreme is provided by uniform linear arrays with co-located transmitters and receivers. Such configurations are characterized by a large number of overlapping effective phase centers and are therefore regarded as highly redundant. In this paper, we will show that such a redundancy is nevertheless well suited to resolve an inherent challenge of conventional SAR tomography, which is limited in providing unambiguous 3-D scatterer position estimates in case of multiple scattering. For this, we show that redundant MIMO arrays allow not only an a posteriori beamforming on receive, but, at the same time, also a comparable a posteriori (i.e. after data acquisition) beamforming on transmit. This means that one can emulate, from one and the same recorded MIMO-SAR data set, different illumination scenarios on transmit and receive. By evaluating the 2-D spectrum provided by the independent Tx and Rx beams, it becomes then possible to differentiate between single- and multiple-bounce scattering. The separation between single- and double-bounce scattering has also been successfully demonstrated in a ground-based radar experiment and is presented in another EUSAR paper.