Minimum Image Size for Phase Error Correction Using Phase Gradient Autofocus on Synthetic Aperture Radar Images

Abstract

Synthetic aperture radar (SAR) is used for imaging large patches of earth�s surface; however, these images suffer from phase errors induced by demodulation errors. These demodulation error are caused by uncompensated fluctuations in the flight path and atmospheric irregularities which cause unknown delays for the returned signal. Phase errors cause the resulting image to blur along the dimension parallel to the flight path, and must be corrected in order to produce a high resolution radar image. This paper attempts to find a minimum image size for which phase gradient autofocus (PGA) can effectively be used to correct phase errors. To measure the accuracy of the correction two metrics are used: average coherence after coherent change detection (CCD) and mean absolute error (MAE) of the estimated phase error compared to the applied phase error.