As the signal bandwidth and the number of channelsincrease, the synthetic aperture radar (SAR) imaging system produceshuge amount of data according to the Shannon-Nyquisttheorem, causing a huge burden for data transmission. This paperconcerns the coprime sampling and nested sparse sampling,which are proposed recently but have never been applied to realworld for target detection, and proposes a novel way which utilizesthese new sub-Nyquist sampling structures for SAR samplingin azimuth and reconstructs the data of SAR sampling bycompressive sensing (CS). Both the simulated and real data areprocessed to test the algorithm, and the results indicate the waywhich combines these new undersampling structures and CS isable to achieve the SAR imaging effectively with much less datathan regularly ways required. Finally, the influence of a little samplingjitter to SAR imaging is analyzed by theoretical analysis andexperimental analysis, and then it concludes a little sampling jitterhave no effect on image quality of SAR.