Summary
International Symposium on Antennas and Propagation
2010
Session Number:2WC1
Session:
Number:2WC1-4
Adding Dense Multipath Components to Geometry-Based MIMO Channel Models
Juho Poutanen, Fredrik Tufvesson, Katsuyuki Haneda, Lingfeng Liu, Claude Oestges, Pertti Vainikainen,
pp.-
Publication Date:2010/11/23
Online ISSN:2188-5079
Summary:
In geometry-based stochastic channel models (GSCMs), the radio channel is created by placing clusters in the simulation environment to act as physical scattering objects. The clusters consist of groups of closely located multipath components (MPCs), and the directions, delays, and complex amplitudes of each MPC are directly computed based on the geometry of the simulation environment. Since GSCMs are inherently capable of modelling the spatial and temporal characteristics of radio channels realistically, they have become very popular in MIMO channel modelling. Examples of cluster-based GSCMs are the COST 259 [1], COST 273 [2], and WINNER [3] channel models. One of the main open issues in the existing GSCMs is the incorporation of the so called dense multipath components (DMC) into the models. Even if the DMC are known to contribute significantly to the radio channel characteristics, a serious effort has not been put in adding the DMC to the existing implementations of the GSCMs. A part of the reason why the DMC have been omitted from GSCMs has been the lack of thorough understanding on the propagation characteristics of the DMC. However, results reported recently in [4] ? [6] have brought the understanding on the underlying physical phenomena related to the propagation mechanisms of the DMC to a level at which it is possible to take further step in incorporating the DMC into the clusterbased GSCMs. In short, measurements have shown that for each cluster the DMC have an angular distribution similar to that of the specular components (SC), and, in addition, exhibit an exponential decay in the delay domain. Still it is somewhat of an open question how much the parameter estimation algorithm influences the specific characteristics of the DMC, but it can be concluded that the angular properties as well as the delay properties of the SC and the DMC are closely connected. In this paper, we present a method of including DMC to GSCMs. The basic idea of the proposed method is to add MPCs representing the DMC around the center of each cluster. In the delay domain, the added DMC MPCs are associated with an exponentially decaying power delay profile (PDP); in the angular domain, the MPCs of the DMC are generated according to the same distribution as the SC, but with a larger deviation. The remainder of this paper includes the following contents. Section 2 summarizes the recent findings on the characteristics of the DMC, thus forming a basis for the modelling approach proposed in this work. In Section 3, the methods of how the DMC are added to GSCMs are detailed. Finally, Section 4 concludes the work.