LAUSR

In this paper, a non-stationary three-dimensional (3-D) wideband geometry-based stochastic model (GBSM) is proposed for low altitude unmanned aerial vehicle (UAV) multiple-input multiple-output (MIMO) communication systems. The proposed GBSM is a combination of line-of-sight (LoS) components, local multipath components (MPCs) scattering from the scatterers around the receiver (Rx), named as local scatterers, and far MPCs scattering from far scatterers, defined as not local scatterers, and uses two-dimensional (2-D) one-ring and 3-D cylinder to mimic local scatterers as well as 3-D multiple confocal elliptic-cylinders to mimic far scatterers. In this paper, two-state continuous-time Markov chains (CTMCs) are introduced to model appearances or disappearances of the LoS components, local MPCs at the transmitter (Tx) installed on UAV, and far MPCs at the Tx and Rx, and the evolution process of the far MPCs is also investigated. The concept of visibility region (VR) is introduced to model the birth and death processes of the local MPCs at the Rx, and the effect of the size of the VR on channel statistics is also considered. In the proposed GBSM, the inherited nature of the LoS components, the local MPCs and far MPCs is considered, and the corresponding statistical properties are derived. The proposed non-stationary 3-D GBSM is validated by the measured results in terms of temporal correlation, and numerical results show that the proposed 3-D GBSM is suitable for describing non-stationarity of 3-D UAV-MIMO channel.