Three-dimensional geometry-based stochastic modeling and performance of 4×4 space-polarization mobile-to-mobile wideband MIMO channels

In this paper, we propose extending conventional co-polarized (CP) MIMO architectures for mobile-to-mobile communications to 4×4 space-polarized (SP) MIMO architectures. SP-MIMO systems employ spatially-separated dual-polarized antennas (where each dual-polarized antennas has two co-located orthogonally-polarized elements) in place of spatially separated CP antennas, effectively doubling the number of antenna ports without dramatically increasing the antenna system form factor. To analyze the potential gains that can be achieved with a SP-MIMO architecture, a mobile-to-mobile (M2M) wideband channel model is developed using a three-dimensional (3D) geometry-based stochastic channel model (GSCM). GSCM channel realizations are characterized by several parameters including the angular distributions of the scatterers, angles of arrival (AoA), angles of departure (AoD), antenna gain patterns, the Rician K-factor, the maximum Doppler frequency, the cross-polar discrimination (XPD), and the co-polar ratio (CPR). Using the SP-MIMO channel model, we study and quantify performance shifts that are possible when the CP antennas in a CP-MIMO systems are each replaced with a dual-polarized antenna, leading to a SP-MIMO system. Using simulation, we compare the performance of the resulting 4×4 SP-MIMO system relative to the conventional 2×2 CP-MIMO system in terms of capacity and diversity measures. Results show that, in the case of identical total transmit powers, 4×4 SP-MIMO systems dramatically outperform 2×2 CP counterparts in a variety of non-isotropic scattering and multipath fading propagation environments.