Abstract
A new approximate recursive least square filtering algorithm is derived for a two-dimensional discrete random field. The state-space model due to Attasi12) is employed for describing the random field. The principle of orthogonal projection is applied for deriving the equations for the filtered and predicted estimates of the states. A closed form formula for the covariance equation is obtained by trancating, the covariance expression in terms of a double series. Digital simulation studies are also carried out to show the feasibility of the recursive algorithm presented here.
