We propose a method for estimating the parameters of the K- distribution on a manner which significantly reduces computational requirements when compared with existing approaches. additionaly, the method yields paramter estimates with lower variance when compared with standard techniques.
展开▼