Abstract: High range-Doppler sidelobe interference may severely deteriorate the accurate detectability of weak targets near strong targets, and the performance for non-grid targets is not considered in the existing algorithms. To solve these problems, a fast iterative range-Doppler sidelobe suppression algorithm based on two-dimensional echo is proposed in this paper. Under the reiteration minimum mean square error framework, the optimal weight vectors are estimated adaptively for each range-Doppler cell based on the gridded two dimensional echo. Besides, the inverse covariance matrix of different Doppler cells in the same range bin is converted to inverse covariance matrix of a certain range cell, which can significantly reduce the computational complexity without losing any accuracy and detection performance. The effectiveness of the proposed algorithm is validated by simulation and experimental results.