Abstract: To address the problem that high speed sampling and large scale data storage are encountered by a high-resolution and wide-swath synthetic aperture radar(SAR), a novel method of multiple azimuth beams(MAB) SAR imaging via compressed sensing is proposed. The raw data is collected with lower sampling rate than the required Nyquist sampling rate in the fast and slow time dimension respectively, then the measurement matrix is established in range and azimuth dimension respectively, and two-dimensional reconstruction of a target using compressed sensing is carried out. The simulation results indicate that this method can not only solve the contradiction between azimuth high resolution and wide swath with less sampled data, but also reconstruct the target image under the noise.