Abstract: The bistatic SAR, when transmitter and receiver move along nonparallel tracks with unequal velocities, has the potential for various applications. But the double-square-root term of range history and a large amount of echo storage are two basic challenges of traditional frequency domain imaging for bistatic SAR. Recent theory of Compressed Sensing (CS) suggests that exact recovery of an unknown sparse signal can be achieved from few measurements with overwhelming probability. In this paper, a novel bistatic SAR high resolution imaging algorithm is proposed based on CS theory and the model of bistatic SAR. In the novel algorithm, the 2-D random down-sampling echo data is as measurement value and the targets are reconstructed via CS in the range and azimuth direction, respectively. The simulation results show that the targets can be perfectly reconstructed by only using few down-sampling echo data instead of all numbers of measurements and also verify the validity of the proposed algorithm which is higher resolution, lower peak side-lobe ratio (PSLR) and integrated side-lobe ratio(ISLR), less sampled data than traditional bistatic SAR imaging algorithm.