Chao Wang, Jianhua Zhang, Lei Tian, Mengmeng Liu and  Ye Wu
Publish Year: 2017

Abstract: It is necessary to study the spatial evolution of clusters in massive multiple-input multiple-output (MIMO) mobile channel, which is one of the important features in clusterbased models. In this paper, we analyze the data collected from the massive MIMO mobile measurement campaign, with a 256-element virtual array at 3.5 GHz in line of sight (LoS) and non line of sight (NLoS) conditions, respectively. Firstly, the clusterlevel angular power spectrums (APS), azimuth of departure (AoD) and azimuth of arrival (AoA), are shown to display the variation of clusters along with the measurement route. Secondly, the evolution of cluster number is presented, showing the changes of appearance and disappearance of the clusters, and the corresponding model based on birth-death process is built. Finally, the cumulative density functions (CDF) of radii of visibility regions (VR) are calculated, which are used to study the variations of VRs along with the measurement route, and lognormal distribution fits the CDFs well.