Covariance matching estimation of ocean surface velocity and coherence time in ATI-SAR systems

Conventional along-track interferometric synthetic aperture radar (ATI-SAR) systems derive the ocean surface velocity from an estimate of the phase difference between the SAR echoes received from two displaced phase centres with a single time-lag. We propose an asymptotic maximum likelihood (ML) algorithm for jointly estimating the ocean surface velocity and ocean coherence time by using multibaseline ATI-SAR data collected from an array of phase centres with multiple time-lags. The method combines a covariance matching (COMET) algorithm with the use of the extended invariance principle (EXIP). Simulated results show that the proposed technique is highly efficient in a practical small sample regime, and provides better estimation accuracy than the conventional single-baseline system. Moreover, it provides unambiguous velocity retrieval and flexibility to varying ocean coherence time