Stripmap autofocus of short range Ku-band synthetic aperture radar data

An autofocus scheme for stripmap synthetic aperture radar (SAR) data using dominant scatterers is given. The focus of the paper is on assessing the performance on real data from an experimental Ku-band gated frequency modulated continuous wave (FMCW) system. The measurements are performed from a rooftop rail of 33 m length. An inertial navigation system consisting of a satellite navigation receiver and an inertial measurement unit (IMU) produces input to a post processing software. The satellite receiver only gives centimeter relative accuracy, while including the IMU data we get millimeter relative accuracy. The result when applying the millimeter accuracy navigation data serves as a reference for the proposed autofocus scheme. A time domain backprojection is applied to the data, with the centimeter accuracy navigation data as input. At Ku-band this will not be sufficient to get focused SAR images. Dominant scatterers are chosen from the partly focused image, and spatial filters are applied to the image before an inverse wavenumber algorithm is applied to get clean range profiles. The phase history of the clean range profiles are then used in the autofocus algorithm. The result from the autofocus scheme is compared to the image produced when including the IMU data in the navigation post processing software.