Application Of GPS And IMU In High Resolution SAR

SAR system trends towards small-sized and high resolution in recent years, and motion error has more and more important influence on SAR imaging, therefore motion error compensation plays an extremely important role in high resolution SAR imaging.GPS and IMU are widely used in SAR, as they can directly provide the motion information of the SAR antenna. However, GPS has a low data rate and limited position accuracy, while its positioning error does not accumulate for a long time, and IMU' positioning error accumulates seriously and it cannot be used in positioning for a long time. So it is a good idea to combine the GPS and IMU for a better positioning result.This paper is aimed to find a way to get an accurate motion errors by using GPS and IMU-accelerometer. In order to find the best way to combine the GPS and IMU, the content including the SAR imaging algorithm, GPS and IMU data preprocessing method and combining method are discussed. The main work and achievements of the paper are as follows:1. First, the model of FMCW SAR echoes and motion error are established, and the demand precision of motion error compensation is analyzed. Based on the existing imaging algorithm, self-autofocus and motion parameters estimating are involved to compensate the remained phase error, and a new improved RMA is put forward.2. Then the data preprocessing procedure of GPS and accelerometer is discussed,so the performance of the GPS and accelerometer, which is the basement of following GPS and accelerometer fusion, can be analysed. Focus on the influence of gravity component, a method is developed to estimate the gravity component by using the GPS information.3. Finally, to combine the information of GPS and accelerometer, two interpolation methods are proposed. By theoretical analyzing and real data positioning, the effects of two interpolation methods are compared. In SAR imaging, Motion compensation using the proposed two interpolation methods and GPS data alone are compared. The results verify the validity and high performance of the proposed interpolation algorithm.