User Selection Methods for Multi-User Downlinks with Multiple-Antenna Terminals

The need for high downlink throughput has led to the consideration of multiple-input multiple-output (MIMO) systems with multi-antenna user terminals. Spatial multiplexing via orthogonal space division multiplexing (OSDM) may be achieved using block diagonalization (BD) techniques. User subset selection is required when the potential user pool exceeds the available transmission resources. Judicious user subset selection will help ensure high system throughput. However, optimum selection requires exhaustive search. We demonstrate how existing antenna selection algorithms may be modified to operate on blocks of the overall composite channel matrix to serve as user selection algorithms for BD-OSDM. Four algorithms are derived. The approach is efficient because no BD pre-coding related computations are needed. Next, BD may employ additional user selection or receive antenna selection on the chosen user subset to help increase the sum rate or minimize the transmission power. We will show how three of the proposed algorithms lend themselves for this purpose with low additional complexity. Numerical results show good sum rate performance for the proposed algorithms especially when receive antenna selection is done on the chosen user subset