Opportunistic Scheduling with Frame Aggregation for Next Generation Wireless LANs

The multiuser diversity phenomenon is exploited via opportunistic scheduling for increasing system throughput in wireless networks. Another method to enhance system throughput is through frame aggregation, which increases MAC efficiency. We consider opportunistic scheduling jointly with frame aggregation. In this paper, we argue that existing opportunistic schemes are not optimal when frame aggregation is used. We propose new scheduling approaches that combine channel states with queue states. Through detailed simulations, we show that our new algorithms offer significant improvement in network throughput over non-opportunistic and greedy schedulers. Our algorithms also provide a good compromise between throughput and fairness.

I. Introduction

Wireless access has been increasingly popular recently due to portability and low cost. Extensive research is being carried out to increase provided data rates of wireless networks to values comparable with wired networks. The emerging technologies for wireless local area networks (WLANs) are defined by the IEEE 802.11 standards, which started by 802.11b with Physical Layer (PHY) data rates of up to 11 Mbps and were enhanced in 802.11a/g to provide up to 54 Mbps with the introduction of Orthogonal Frequency Division Multiplexing (OFDM). In the newest WLAN standard group 802.11n, PHY data rates exceeding 200 Mbps are provisioned with the realization of multiple input multiple output (MIMO) techniques. MIMO systems are based on the presence of multiple antennas at the transmitter and receiver ends of the communication link, which make use of spatial diversity and yield significant increase in system capacity. On the other hand, the actual throughput experienced by the WLAN users is considerably lower than the PHY data rates. Since the transmission media is shared, efficient multi/Medium Access Control (MAC) is the key to provide desired high data rate services.