Aimed at continuous scene imaging, this paper utilizes signal reconstruction method to eliminate random phases of scatter cells and reduce signal bandwidth, to achieve downwardlooking 3D imaging in the condition of sparse flight for array SAR. With the sampling method of repeat flight such as 7 dense flights and 4 sparse flights with sampling criteria of [1 1 1 0 0 1 0] (Barker code sequence), [0 1 1 0 1 0 1] and [1 0 1 0 1 0 1] (uniformly-spaced sequence), respectively, the 3D imaging simulation is implemented in this paper. In addition, we compare the imaging quality before and after signal reconstruction. Simulation results illustrate that the adoption of signal reconstruction can significantly improve image quality; the combination of Barker code sparse flight and signal reconstruction presents the best performance among all the implemented methods.