MAC/PHY Layer Strategies for High Efficiency WLANs

Abstract

Along with the steady increase in mobile data traffic, wireless local area network (WLAN) technology has been developed to support heavy traffic for various mobile devices. The-state-of-art IEEE 802.11 specifications such as 802.11n and 802.11ac have focused on improving physical layer (PHY) rate by enabling multiple stream transmission via multiple-input multiple-output (MIMO) technology, wide bandwidth transmission via channel bonding, and high order modulation via 256-QAM and short guard interval. While the emerging technologies greatly increase PHY rate over 1~Gb/s, the achievable throughput is much limited due to the low reliability with high PHY rates and medium sharing among the nodes operating on the same channel. In this dissertation, we tackle three different strategies to enhance the achievable throughput in IEEE 802.11 WLANs. Firstly, we study a cost-effective approach, namely antenna subset selection, to enhance reliability even for the high PHY rates. There are practical challenges to employ antenna subset selection in WLANs such as the lack of channel state information at the transmitter and multiple retry chain utilization. Only few researches have addressed those practical challenges, which result in a limited employment of antenna subset selection in WLANs. We propose a practical antenna subset selection system considering those practical challenges, and evaluate the performance of the proposed system via prototype implementation and extensive experiments. Secondly, we focus on the clear channel assessment (CCA) of IEEE 802.11 WLAN which is too conservative to exploit spatial reuse. The problem is arise due to a limitation of the current CCA mechanism. Only the received signal strength (RSS) of an ongoing transmission is used to determine the status of the medium, i.e., busy or idle. We propose a novel CCA mechanism which utilizes the information delivered in PHY header of the ongoing transmission so that it can stimulate concurrent transmissions for better spatial reuse. Through simulations, we evaluate our proposed approach and demonstrate throughput gain in various scenarios. Lastly, we investigate transmit power and data rate control method to further exploit spatial reuse. Along with our proposed CCA mechanism, more concurrent transmissions become feasible by adapting transmit power and data rate depending on the ongoing transmission. Accordingly, we propose a joint transmit power and data rate control algorithm which operates dynamically depending on the existence of ongoing transmission. We evaluate our proposed algorithm under various scenarios through extensive simulations. In summary, we propose three different methodologies for high efficiency WLANs, one for reliability enhancement and the other two for better spatial reuse. The operation and performance gain of each methodology are verified by testbed experiments or network level simulations.