In the high resolution SAR, the effect of motion error at the azimuth direction is more severe. Though the traditional Phase Gradient Autofocus (PGA) algorithm can be used to estimate the motion error, the iterative phase correction has to be done. The huge computation makes it difficult to be used in real time image processing. In order to meet the request of real time imaging of high resolution SAR, the Iterative Shift And Correlate (ISAC) algorithm, PGA algorithm and Range Doppler (RD) algorithm are combined in this paper. With the new combined algorithm, the iterative operation can be reduced. At the same time, the precision request for the initial motion parameter is decreased by adding appropriate iterations to the Shift And Correlate (SAC) algorithm. The theoretical analysis and the result of imaging for real SAR data can validate the new algorithm.