Parameter estimation of two-dimensional moving average random fields: algorithms and bounds

This paper considers the problem of estimating the parameters of two-dimensional moving average random elds. We rst address the problem of expressing the covariance matrix of a moving average random eld, in terms of the model parameters. Assuming the random eld is Gaussian, we derive a closed form expression for the Cramer-Rao lower bound on the error variance in jointly estimating the model parameters. A computationally e cient algorithm for estimating the parameters of the moving average model is developed. The algorithm initially ts a two-dimensional autoregressive model to the observed eld, then uses the estimated parameters to compute the moving average model.