Mutant Altimetric Parameter Estimation Using a Gradient-Based Bayesian Method

Abstract

This paper proposes an advanced Bayesian algorithm for estimation of mutant altimetric parameters. A sparse prior is introduced to enforce a mutant evolution of the altimetric parameters. A <i>maximum a posterior</i> (MAP) estimator based on an alternating optimization algorithm is carried out to fulfill our proposed hierarchical Bayesian model. The proposed Bayesian method and the corresponding estimation algorithm are evaluated using both synthetic and real altimetric data associated with a delay/Doppler altimetric model. The experimental results show that the proposed method brings an improvement on mutant parameter estimation and tracking when compared to smooth estimation and other state-of-the-art estimation algorithms.