Statistical modeling of the received signal envelope in a mobile radio channel

Propagation loss measurements taken in the Ottawa region in the 900-MHz frequency band are reported and discussed. Significant statistical parameters for a mobile radio channel, such as the mean, the standard deviation of the signals, the level crossing rates and the fading distance, are presented. The signal is first transformed from a time scale to a distance scale; it is then separated into slow and rapid variation components, a low-pass filter with a cutoff point corresponding to 0.125 cycle per wavelength being used to estimate the slow variation component. The statistical analysis of the rapid variation is related to the Rice probability law and to a direct-component-to-multipath-component ratio. For the medium-density urban area under consideration, this ratio is shown to change from around 5 dB to around 9 dB as one moves from medium-density urban to the surrounding open areas