A Throughput Balancing Problem between Uplink and Downlink in Multi-user MIMO-Based WLAN Systems

Collision mitigation is one of classical research issues for wireless local area networks (WLANs). Recently, multiple-input multiple-output (MIMO) transmission techniques have been widely deployed in wireless systems, while a multi-user MIMO-based collision mitigation scheme in uplink WLANs was proposed by authors, and we showed the scheme is very efficient for the uplink performance. However, for an infrastructure-based WLAN, we observe a significant performance unbalance problem between uplink and downlink, compared to the conventional WLANs. Moreover, access point (AP) yields lower throughput performance than each contending station(STA). In order to solve this unbalance problem between uplink and downlink, we adopt a modified minimum contention window (CWmin) adjustment scheme and a random piggyback scheme to the multi-user MIMO-based WLANs. We also develop an analytical model to evaluate the performance of multi-user MIMO-based WLANs in a saturated traffic environment. The result shows that the random piggyback scheme performs more efficiently for the multi-user MIMO-based WLANs.