Statistically correct simulation models for the generation of multiple uncorrelated Rayleigh fading waveforms

Abstract

An ergodic simulation model for the generation of multiple time uncorrelated Rayleigh waveforms with correct ensemble averages is proposed. This model uses the inverse discrete Fourier transform to generate the transmission coefficients with only one random number generator. Two recent modified versions of the Jake's and Smith's models, with correct ensemble averages, are analysed and compared with the proposed model. Analytical and simulation results show that the proposed model is ergodic in mean and autocorrelation, as is the modified Smith's model. Simulation results also show that the modified Jake's model requires the highest computation effort among the three analysed models. As a result, this model become less efficient in simulation.