ACCURATE APPROXIMATION TO THE OPTIMUM PARAMETER ESTIMATE FOR K-DISTRIBUTED CLUTTER

The authors analyse the suboptimal performance of simple texture measu res for estimating the order parameter of K-distributed radar clutter. A noncommittal neural net has been applied to the parameter estimatio n task and it shows that improved error estimates are obtained when mu ltiple moments are used to characterise the texture. A new estimator i s proposed which combines the mean normalised log intensity and the am plitude contrast moments of the imaged data to provide a more accurate measure of the texture information, which results in lower errors in the parameter estimates. The relative weighting in which the two momen ts are combined determines the error performance of the estimator. Usi ng a constant weight value, an estimator has been derived which gives close to maximum likelihood performance on the estimates over a wide r ange of the parameter values which are of interest. Thus it is shown t hat the use of multiple moments in a texture measure produces a closer approximation to the optimum parameter estimate for a K-distributed p rocess.