A stochastic model of the temporal and azimuthal dispersion seen at the base station in outdoor propagation environments

TLDR

A simple statistical model of azimuthal and temporal dispersion in mobile radio channels is proposed. The model incorporates delay and azimuth PDFs and power conditioned on them, derived from macrocellular measurements, and in challenging urban settings employs a multicluster approach with Laplacian PAS and exponential PDS. In typical urban environments, the power azimuth spectrum follows a Laplacian distribution, the azimuth PDF is Gaussian, and both the power delay spectrum and delay PDF are exponential; in more challenging urban settings, a multicluster model with summed Laplacian and exponential components better captures dispersion.

Abstract

A simple statistical model of azimuthal and temporal dispersion in mobile radio channels is proposed. The model includes the probability density function (PDF) of the delay and azimuth of the impinging waves as well as their expected power conditioned on the delay and azimuth. The statistical properties are extracted from macrocellular measurements conducted in a variety of urban environments. It is found that in typical urban environments the power azimuth spectrum (PAS) is accurately described by a Laplacian function, while a Gaussian PDF matches the azimuth PDF. Moreover, the power delay spectrum (PDS) and the delay PDF are accurately modeled by an exponential decaying function. In bad urban environments, channel dispersion is better characterized by a multicluster model, where the PAS and PDS are modeled as a sum of Laplacian functions and exponential decaying functions, respectively.