Comment on "Estimate of channel capacity in Rayleigh fading environment

Lee (see ibid., vol.39, p.187, no.8, 1990) proposed a very simple and elegant model for estimating the capacity of slowly fading channels. A small error in his paper, however, has led to an inexact picture. We correct that inaccuracy. As a consequence, the loss in capacity of a Rayleigh fading channel with respect to the additive white Gaussian noise (AWGN) channel turns out to be negligible for all values of the signal-to-noise ratio that are of interest. Furthermore, we extend the analytical results to the situation with M-branch diversity and Ricean fading channels. The lower moments of the associated distributions, i.e., of the /spl chi//sup 2/-distribution (Nakagami with integer m-parameter) and Rice distribution agree when the parameters of the distribution are suitably chosen. With that choice of parameters, the capacities of the M-branch diversity and of the Ricean fading channel become almost identical.