Cluster-based geometrical dynamic stochastic model for MIMO scattering channels

This paper proposes a novel cluster-based geometrical dynamic stochastic model for multi-input and multi-output (MIMO) communication environments. It is assumed that the base station (BS) has an adaptive learning ability and the BS preprocesses the received data in advance. In the proposed channel model, scatterers can be clustered into different clusters based on a machine learning cluster algorithm. Subsequently, the scattering channels that correspond to different geometric ellipses can be established. On this basis, the authors derived the closed-form expression for the marginal probability density function (PDF) of the angle of arrival and the expression of the space time cross-correlation functions for different scatterer clusters. Close agreement is achieved between the results obtained in this study and previously reported results, thereby validating the generalization of the proposed channel model.