Abstract: Multiuser scheduling for spatially correlated channels is investigated in multiuser multiple-input multiple-output (MIMO) systems. The impact of user scheduling algorithms on the orthogonality between the scheduled users is analyzed by characterizing the statistics of the angle between users’ channel vectors. It is shown that existing successive searching based schedulers reduce the orthogonality between scheduled users in a statistic sense in spatially correlated channels, which leads to a large performance gap between successive searching based schedulers and the optimal exhaustive searching scheduler. In order to improve the performance of multiuser MIMO systems in spatially correlated channels, an alternating searching based user scheduling method is then proposed and its low-complexity implementation is studied. Simulation results show that the proposed user scheduler effectively recovers the performance loss caused by successive searching based schedulers, which performs close to the optimal exhaustive searching but with much lower computational complexity in both spatially independent and correlated channels.