Influence of noise on a novel RMS delay spread estimation method

This paper describes the influence of noise on a previously proposed measurement method for wide-band, Rayleigh- and Ricean-fading radio channels. As the original procedure exploits the proportionality between the level crossing rate in the frequency domain (LCR/sub f/) and the RMS delay spread of the channel, the influence of noise on the LCR/sub f/ is investigated and evaluated here. A compact solution is given for the noise-influenced case, which requires only one parameter, additional to those from the original procedure. However, we observe that the continuous-time (and -frequency) model used to obtain this solution yields a bias of the theoretical result against the ones obtained from the measurements, since they are discrete in frequency. Therefore, for the Rayleigh fading case, the discrete-time LCR is derived to identify the source of this bias. It is seen that the analytical result is not very suitable for being applied to a measurement procedure. It can be used however, to quantify the influence of noise in a particular situation and to assess noise-reduction techniques that might be required.