Iterative maximum likelihood and zFlogz estimation of parameters of compound-Gaussian clutter with inverse gamma texture

Abstract

Compound-Gaussian model with the texture of inverse gamma distribution is regarded as one of the best models to characterize the high-resolution sea clutter at low grazing angles. The model parameters are usually estimated through moments or maximum likelihood function of sample. The former one is of low precision since the using of high order moments. The latter is difficult to implement due to its high computing complexity. In this paper, a zFIogz estimator is proposed to decrease the order of moments, and an iterative maximum likelihood (ML) estimator is constructed to simplify the computation of ML estimator. The experiments based on simulated and real sea clutter data show that the proposed methods perform better than the method of moments while computing more efficiently than the ML estimator.