A measurement-based random-cluster MIMO channel model

We present a novel geometry-based stochastic MIMO channel model based on the concept of multipath clusters. The so-called random-cluster model (RCM) represents the channel in a propagation-based stochastic way, in which the geometry enters solely by statistical means. The starting point is the recently published parametric COST 273 MIMO Channel Model. However, instead of using physical parameters (e.g. rooftop height or scatterer distance), our model requires only cluster parameters. As another major feature beyond COST 273, the model can now be parametrized directly from measurements. The model presently supports multi-user packet access, i.e. the channel changes completely from one time instant to another. Simple future extensions will incorporate smoothly time-variant channels as well.