Order-Adaptive Fractional-ORDER Moment-Based Estimation of Shape Parameter of K-distribution

In the paper, an order-adaptive fractional-order moment-based (OAFOM) estimator is proposed to estimate the shape parameter of K-distribution. The performance of a fractional-order moment-based (FOM) estimator depends on both shape parameter and the order of the fractional-order moment. From numerical experiments, an empirical formula of the best order as a function of the shape parameter is found. An initial estimator of the shape parameter is obtained from one moment-based estimator and the best order is determined from the initial estimate and empirical formula. And then the FOM estimator with the best order is the final estimate of the shape parameter. Simulated results show that the proposed estimator attains better performance than other estimators.