Investigations on Massive MIMO Performance with Multi-Antenna Users by Ray-Tracing

The downlink data rate in a single-cell urban massive MIMO network is evaluated where the scenario comprises a base station with 256 antennas and 64 users each having two antennas. The downlink performance according to three different precoding algorithms, which can exploit multiple antennas at user terminals to improve the performance, are presented. Two different methods are used to generate the channel matrix, in particular, ray-tracing simulations and the WINNER II stochastic channel model, and the resulting downlink performances are compared.