Parametric texture estimation and prediction using measured sea clutter data

In this study, the authors present a deterministic parametric sea clutter texture model for high-resolution radar backscatter at low-grazing angles in the open ocean. The clutter texture forms a component of the compound-Gaussian sea clutter model and they exploit the spatiotemporal relationships in the clutter by relating it to its physical source: sea swell. They present an efficient algorithm for the estimation of the spectral components for the parametric texture model through the estimation of two-dimensional (2D) 'tones' across contiguous range bins instead of a series of 1D estimates as is used elsewhere. Validation is performed by comparing the predictive fit for their estimator with a series of temporal estimators and a non-parametric estimator using measured sea clutter data from the Atlantic Ocean recorded by the Intelligent PIXel (IPIX) radar of McMaster University in Canada. Implementation of the spatiotemporal estimator results in a more parsimonious estimate with improved reliability due to the increased separation of the tones in 2D space. Their results are shown to agree with an established physical model for the sea swell.