Abstract: ‍ ‍Compared with the traditional single-site linear array synthetic aperture radar （SAR）， the arc array bistatic SAR with an aerostat is a new imaging system with high efficiency and high resolution. This imaging mode combines the benefits of arc array antennas and aerostats， offering significant potential for applications in helicopter landing， emergency relief， and similar areas. In this imaging mode， due to the effect of the horizontal movement of the aerostat platform， the downward movement of the helicopter platform， and the special circular arc structure of the arc array antenna， the echo signal of the target is difficult to process directly， which brings difficulties to the imaging. To solve these problems， a method based on two-step motion compensation and Keystone transformation is proposed with the imaging geometry model and echo characteristics of the model of arc array bistatic SAR of the aerostat transmitter. Initially， Taylor series expansion is used to approximate the bistatic platform complex oblique range equation， and the range resolution and azimuth resolution in this imaging mode are briefly analyzed. In the second step， according to the oblique range error caused by the aerostat’s horizontal motion， the corresponding motion compensation factor is constructed to compensate for the phase of the range migration term caused by the aerostat’s translation. The third step is to construct the corresponding compensation factor to compensate for the range migration term caused by helicopter descent according to the slant range error caused by the downward movement of the helicopter platform. Subsequently， to correct the coupling between range and azimuth， a new angle variable is introduced to transform the azimuth angle according to the Keystone transformation idea to compensate for the coupling between range frequency and azimuth angle， and then azimuth pulse compression is conducted in the azimuth frequency domain. The accuracy of the imaging results and the effectiveness of the method are confirmed through a comparative analysis. This includes examining the simulation results of various lattice targets before and after imaging compensation， analyzing imaging performance indicators of these targets， and simulating scene distribution targets.