Abstract: With regard to the range-Doppler (R-D) imaging algorithm of inverse synthetic aperture radar (ISAR) using stepped frequency chirp signals in low signal to noise ratio (SNR) environment, envelope alignment and phase focusing must be completed accurately. But in case of unknown target motion parameters, general method is difficult to solve the problem. Because the coarse distance signals of stepped frequency chirp radar are linear frequency modulation (LFM) signals in slow time domain, the fractional Fourier transform (FRFT) can be used to analyze these signals. While the FRFT spectrogram of single sub pulses train may be drowned with noise in low SNR, FRFT spectrograms of several consecutive sub-pulse trains can be overlaid by shifting a certain distance to search the target爷s motion parameters under intense noise. When the motion parameters are obtained, compensation functions can be constructed to accomplish envelope alignment and phase focusing. Because FRFT has fast algorithm and its computing speed is similar to that of fast Fourier transform (FFT) algorithm, the parameter estimation part of the algorithm is efficient and easy to implement. The simulation results show that the method can get ideal imaging scenarios and prove the effectiveness of the algorithm.