Multiuser (MU) - multiple input multiple output (MIMO) techniques are attractive to increase the downlink spectrum efficiency since they allow a small number of antennas at each user station (STA). In MU-MIMO systems, the transmission performance is affected by the STA combination that is spatially multiplexed in the same frequency and same time slot. Thus, the AP needs to determine the optimum STA subset and the number of spatially multiplexed STAs using the channel state information. In this paper, we propose a user selection scheme that uses an iterative algorithm to determine the number and the optimal combination of multiplexed STAs. The proposed scheme chooses STAs one by one so as to improve the throughput of all STAs in the subset compared to that in single-user MIMO transmission. Thus, all STAs have a fair chance at being selected even when their SNRs vary widely. Furthermore, the iterative algorithm has lower calculation complexity than the exhaustive search approach. The index for employed for user selection is also simplified by using the signal-space vectors of the channel matrices. Computer simulations confirm that the effectiveness of the proposed user selection scheme in a wide-range of SNR values. The results show that the user selection proposal almost matches the performance of exhaustive search.