Abstract: The down-looking array 3D-SAR avoids the de-correlation problem caused by multiple flights and can well solve the problems of shadow, overlay and the like in side-looking SAR imaging.The number of conventional down-looking array 3D-SAR cross-track array elements is large, which increases the cost and complexity of the system. Moreover, radar data needs to be sampled in three dimensions, and the amount of echo data is large. In view of the above problems, this paper proposes a two-dimensional sparse 3D-SAR imaging algorithm on the cross-track and along-track. The random arrangement of the array antennas in the cross-track reduces the number of array elements of the array antenna, and randomizes the along-track to random sampling, reducing the number of along-track sampling points. In this paper, Orthogonal Matching Pursuit algorithm(OMP) is used to perform sparse reconstruction of two dimensional data. Finally, the effectiveness of the imaging algorithm is verified by point target simulation experiments.