Study On New Motion Compensation Methods For SAR/ISAR Imaging

The imaging radar, which can provide an all-weather, day/night capability to generate a high-resolution two-dimensional representation of the radar reflectivity of the targets over a long range, has received intensive attention because of it potential applications in both civilian and military applications. In order to obtain high-resolution radar images, it is necessary to maintain the precise phase relationship between return signals. This require precise motion compensation, because the unplanned perturbations of the relative motion between the radar and the scatterers causes phase errors, which may destroy the phase coherence of the return signals and blur the image severely. Even though the modem electronic navigation systems are onboard in the SAR platform, it is difficult to determine the platform position to the required tolerance over the entire synthetic aperture under the performance limitation of the measurement instruments. This is especially true for high-resolution image systems. So it is necessary to adopt a data-driven technique to eliminate these phase errors, i.e. autofocus. In an ISAR system, the motion of the target measured by the radar system is usually too coarse to be used for high-resolution image application. Autofocus must be used to compensate the motion errors. Except for the phase error caused by uncompensated motion errors, the phase errors that affect the image quality can be caused by the propagation effects and system instability. The dissertation is mainly concerned with the autofocus of the spotlight SAR and ISAR images using the information obtained from the received radar signals themselves, when the motion measurement is not so precise, and even there are no motion measurement instruments. Range alignment of ISAR data is also discussed and a new range alignment algorithm is devoleped. The following is the summarization of the main work: <~ Many SAR/ISAR autofocus algorithms are based on synthesis of the multiple dominant scatterers, that is to synthesize some dominant scatterer bins to increase the signal to cutter plus noise ratio (SCNR), consequently, improve the precision of the phase errors estimation. In chapter 2, several phase error estimation algorithms, -Iv- which are all based on the multiple dominant scatterer synthesis, are discussed based on the concept of using accumulation (coherent/incoherent) to increase the SCNR. The performance of the algorithms is compared using simulation and real measured radar data. ~ In chapter 3, with respect to multiple dominant scatterers in one range bin, a autofocus algorithm based on parameter estimation of scatterers is proposed. Based on the signal model of multiple scatterers, the proposed algorithm estimates the parameters of the scatterers using RELAX. In order to estimate the phase error precisely, it adopts an iterative procedure to estimate the phase error and the scatterers?parameters alternatively. The algorithm is validated by using real- measured radar data. <~ The phase errors caused by unplanned perturbations of the platform in spotlight SAR systems are mainly low frequency phase errors, especially quadratic phase error and high order polynomial phase errors. These phase errors are the main sources to degrade the image quality. In chapter 4, with respect to low frequency high-order polynomial phase errors, a new autofocus algorithm to comp...