Polynomial Regression Predistortion for Phase Error Calibration in X-Band SAR

Abstract

Predistortion is used to compensate for a waveform or phase of a signal in advance. Although the signal that applies predistortion exhibits a nonideal shape at the baseband, it has an improved phase value after the signal is upconverted in the actual hardware. Most synthetic aperture radar studies have focused on image reconstruction using single-look complex or ground-range detected data. Other studies have only compensated for the peak phase error; thus, discontinuity remains. In this letter, we present a novel method to reduce the phase error by applying a polynomial regression predistortion based on system calibration. The new approach is formulated to minimize the overall pulse error in a signal generator. We use the second-order polynomial regression for the phase error and predistort the phase error using the regression value. The proposed method is applied to the baseband signal, which exhibits an improved phase value at the X-band (9.5-10 GHz).