Summary
International Symposium on Antennas and Propagation
2008
Session Number:3A05
Session:
Number:3A05-4
Improved Multipath Clustering of a Small Urban Macrocellular MIMO Environment at 4.5 GHz
Lawrence Materum, Jun-ichi Takada, Ichirou Ida, Yasuyuki Oishi,
pp.-
Publication Date:2008/10/27
Online ISSN:2188-5079
Summary:
Multipath clustering has been started to be considered in Multiple-Input Multiple-Output (MIMO) propagation channel models [1]. Availing the benefits of MIMO systems also includes knowing how these multipath clusters behave in order also to come up with better MIMO wireless system designs. Assuming the presence multipath clusters in a certain channel scenario, the knowledge of them includes identifying them in a satisfactorily accurate way. Cluster identification was done manually in most of previous studies, e.g. [2], thus affecting their subsequent channel models. However, as large data sets from channel sounding have made the manual identification of clusters cumbersome, the use of clustering algorithms has been a resort. Thus, clusters can be identified objectively and eventually modeled without visual bias. In this paper, using an optimized automatic clustering approach, results of the determination of the clustering of multipaths are presented. Our observations show that the distribution of cluster power proportion should also be taken into account when modeling clusters as unnecessary exclusion of least clusters may not represent the characteristics of clusters in the channel.