Paper Info
Title
A Pseudo-Geometrical Channel Model for Dense Urban Line-of-Sight Street Microcell
Abstract
A channel model for line-of-sight (LoS) small street microcell in dense urban areas is proposed. The scattering power distri- bution (SPD) in the channel seen from the receiver (Rx) based on the physical phenomenon is obtained. The coefficients of the distribution are derived by approximation to a set of urban street microcell measurement data. The azimuth-power spec- trum (APS) for the proposed model is compared to those for conventional elliptical model as well as to the experimental re- sults obtained from measurements in three different streets of a dense urban area. It is shown that the proposed model in con- trast to the conventional models produces results that closely agree with the experimental results. I. INTRODUCTION Characterization and modeling of the radio propagation chan- nel are essential for mobile and wireless systems design. A detailed knowledge about mobile communication propagation channel leads to a more successful design of the communi- cation system. Especially to design and evaluate the multi- antenna systems understanding the spatial properties of the channel is a prerequisite. One of the most commonly used di- rectional channel models is based on a geometrical description of the scattering process, which focuses on the detailed inter- nal construction or realization of the channel. The major ad- vantage of geometrical channel models is their simplicity for simulation. The shape and size of the scattering power distri- bution (SPD), or equivalently scattering distribution plus scat- tering coefficients, required to achieve a reliable simulation of the propagation phenomenon however is subject to debate (1). Geometrical modeling of the propagation channel has always been attractive for the researchers due to its advantages. The very well known Jakes model is a geometrical channel model itself (2). In this model the scatterers are assumed to be uni- formly distributed over a circular ring. The model proposed in (3) is circular as well however the scatterers are assumed to be uniformly distributed within a circular disk while (4) pro- poses a circular gaussian scattering distribution around mobile station. In the models suggested in (5), (6) the scatterers are considered to be distributed over an elliptical disk. The major axis of the ellipse is assumed to be along the base station to the mobile axis in (5) while in (6) it has aligned to the street where mobile is located. In all of these models only the local scatter- ing cluster which is located around the mobile station is taken into account. Therefore these models are suitable for macrocell environments where base station antenna heights are relatively large and therefore there is no signal scattering from locations near the base station. As a result, for scenarios where scatterers are located around base station as well as mobile station, the model's performance degrades. For the small cells with rela- tively low height antennas the elliptical model called geometri- cally based single bounce elliptical model (GBSBEM) is more attractive (7), (8). The model assumes a uniform distribution of the scatterers within an ellipse where the base station and the mobile are foci of the ellipse and therefore scattering near the base station is as likely as near the mobile. Nonuniform distri- bution of the scatterers over the ellipse was suggested in (9) by dividing the main ellipse into a number of elliptical subregions where each subregion correspond to one range of excess delay. The elliptical models are particularly attractive for considera- tion of the locus of the scatterers with the same delay. They have been widely used for modeling and simulation of the mi- crocell and even macrocell scenarios (e.g. in (10)). The main drawback of the geometrical channel models is that only a single interaction is accounted for each path and multiple-bounces are not considered. In particular, for a street microcell, the single-bounce assumption is rather restrictive as by no means the street width is sufficient to match the ellipse of the maximum delay. To overcome this shortage the con- cept of effective street width was introduced in (11). In a street microcell scenario, the base station and the mobile are not far from each other. Moreover the antenna heights are lower than the surrounding buildings. In such a scenario the propagation channel is experiencing severe multiscatterings. Considering the locus of the scatterers with equal delays, as in the conven- tional elliptical models, is not sufficient. In fact, the prime im- portance in a propagation channel is the scattering power or path gain. In this paper we introduce an SPD based on major propaga- tion mechanisms in the street microcell. By the term scatter- ing we mean any kind of interaction of the propagation path to the channel including specular reflection, diffuse scattering or diffraction, and accordingly scatterer means the interacting object. In the proposed model the multiscatterings are taken into account however we visualize them as single-bounce scat- terings. By this we can provide the distribution of the scatter- ing power to develop the model. However as this is not ac- cording to the definition of the geometrical modeling, we call the proposed model a pseudo-geometrical channel model. The physical description of the model and the proposed SPD are described in section II. In section III. the coefficients for the proposed model are derived by using measurement data. The azimuth-power spectrum (APS) from the proposed model is compared to that of GBSBEM and the measurement data in
Year
DOI
Venue
2006
10.1109/PIMRC.2006.254033
Helsinki
Keywords
DocType
ISBN
electromagnetic wave scattering,microcellular radio,wireless channels,azimuth-power spectrum,dense urban line-of-sight street microcell,pseudo-geometrical channel model,receiver,scattering power distribution
Conference
1-4244-0330-8
Citations 
PageRank 
References 
3
0.52
6
Authors
3
Name
Order
Citations
PageRank
Mir Ghoraishi1557.69
Jun-ichi Takada223747.62
Tetsuro Imai313222.72