Phase Unwrap Using Nonlinear Kalman Filtering for SAR Systems

In this paper, we develop a novel phase unwrap algorithm with applications to Synthetic Aperture Radars (SAR). It is based on nonlinear Kalman filtering, which can be applied to both one dimensional data and two dimensional phase unwrap. To the algorithm design, the Unscented Kalman Filter (UKF) is superior in the sense of its accuracy. If we apply the UKF to the one dimensional data, a linear least square fitting may lead to an efficient Doppler parameter estimator. Furthermore, we also may apply the UKF to the interferometric SAR data which are two dimensional in nature. Computer simulations demonstrate that the proposed algorithm based on UKF achieves excellent performance. The algorithm is capable of estimating Doppler parameters for one dimensional data. We see that UKF may provide reliable two dimensional phase unwrap for interferometric mode SAR.