Approximating bivariate Suzuki distribution with bivariate Lognormal distribution

Abstract

In this paper, we present a simple method to approximate the bivariate Suzuki distribution by a bivariate Lognormal distribution. In particular, each Suzuki random variable (RV) is approximated by a new Lognormal RV, and the correlation coefficient between two Suzuki RVs is approximated by a new correlation coefficient between two Lognormal RVs. Numerical results validate the accuracy of this approximating method, and this method will facilitate the future analysis and calculation of the diversity system over correlated Suzuki channels.