Array synthetic aperture radar (A-SAR) 3-D imaging is an emerging technique capable of producing a 3-D map of scattered electric fields. Commonly, A-SAR based on the classical matched filter technique requires a large full-sampled 2-D antenna array, which may cause very high system cost. To reduce the number of the antenna elements, this paper surveys the use of compressed sensing recovery and sparse measurement strategies for A-SAR 3-D imaging. Exploiting the spatial sparsity of the underlying scene, we pose the A-SAR imaging as finding sparse solutions to under-determined linear equations. Further, a sparse recovery algorithm based on orthogonal matching pursuit (OMP) is presented to effectively recover the 3-D images with the large-scale echoes data. Lastly, the performance of the approach is verified by an X-band ground-based A-SAR system. The experimental results demonstrate that the approach can produce a better quality 3-D image compared with the conventional method with the very sparse array.