Fuzzy Neural Network Approach for Estimating The K-distribution Parameters

This paper provides a novel approach based on neuro-fuzzy inference system for the estimation problem of the K-distributed parameters. The proposed method is based on a network implementation with real weights and the genetic algorithm (GA) tool is applied for an off-line training of the fuzzy-neural network (FNN) shape parameter estimator. Moreover, the proposed estimator combines the Raghavan's and ML/MOM (maximum-likelihood and moments) methods and the experimental results are presented to demonstrate the validity of the approach. It is shown that such the FNN estimator is successful with a lower variance of parameter estimates when compared with existing Raghavan's and ML/MOM approaches.