Study On Motion Compensation Technique Of Slow Platform SAR In Near-Space

Near Space is the region between about 20km and 100km. It is significant for intelligence gathering, surveillance & reconnaissance and communications. As far as slow platform SAR in Near Space is concerned, it will provide better performance and more agility, such as detection precision, loiter capability, the area of overlay, and so on, than the existing airborne and spaceborne SAR systems for its unique characteristic. Trajectory deviation of platform causes phase errors of radar signal and results in the region fuzzy. So the motion errors must be compensated in order to obtain high resolution imaging.Based on the research way of motion compensation in airborne SAR, We study the phase error's influence to SAR image in Near Space, and indicate that the motion errors of Slow Platform's APC in Near Space are emphasis and difficulty of motion compensation. By having set up the model of the SAR platform motion errors, the compensation precision for the motion errors of the platform is acquired according to the given guide line of slow platform SAR imaging in Near Space.Motion compensation based on measurement instrument and the SAR real data is the key technique to realize high-resolution imaging. The former is particularly studied in this dissertation. First, we analyzed the effects of trajectory deviation of LOS and space-variant phenomena. It is pointed out that the range error can be separated into space-invariant and variant contributions, which can be compensated by the first-order and second-order motion compensation processing. Then, we studied the principle & limitation of improved RD algorithm and present a motion compensation algorithm based on SCFT processing in two-dimension frequency for Slow platform SAR in Near Space. At the same time, the method is validated by the analysis of the simulated data. Finally, the processing results of the two difference algorithm are evaluated and the conclusion is obtained that motion compensation algorithm based on SCFT processing is superior to improved RD algorithm.The error of inertial navigation system was corrected by the weighted least square and kalman filter in integrated navigation system for achieving the precise position measurement of platform and ensuring the motion compensation precision in this dissertation. Then, we designate the limitation of weighted least square and the superiority of CDGPS/INS. The result shows that the accuracy of three-dimensional location can reach centimeter's level when it is corrected by CDGPS/INS, which is superior to the weighted least square.