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