Design and Performance Analysis of MIMO Based WLANs

Multiple Input – Multiple Output (MIMO) refers to the use of multiple antennas at receiver and transmitter in order to improve the performance of communication link and the whole system. It is receiving growing attention for application in wireless networks, reflected in the vivid activities
related to MIMO in standardization bodies for different systems, e.g. IEEE 802.11n and 802.16 (Worldwide Interoperability for Microwave Access (WiMAX)), High-Speed Packet Access Evolution (HSPA+) and 3rd Generation Partnership Project (3GPP) Long Term Evolution (LTE). Application of MIMO poses several challenges to system design, not only on Physical Layer (PHY), but also on Medium Access Control (MAC) layer, such as signaling methods, channel estimation and Channel State Information (CSI) feedback. Also, adapting channel aware scheduling algorithms to the layered structure of the MIMO channel promises high performance gains. MAC layer design and performance evaluation of MIMO based Wireless Local Area Networks (WLANs) are addressed in this thesis. Single-User-Distributed Coordination Function (SU-DCF) and Multi-User – Distributed Coordination Function (MU-DCF) are proposed as extensions of the IEEE 802.11 Distributed Coordination Function (DCF). Orthogonal Frequency Division Multiple Access (OFDMA) signaling is introduced in order to efficiently exchange control frames and reduce the overhead that MU-DCF suffers from. Performance evaluation of these two protocols, by means of achievable throughput, provided fairness, and delay characteristic, on both link and system level is presented, revealing the benefits and drawbacks of Single-User (SU) and Multi-User (MU) transmission strategies. Cases of both uninformed transmitter and informed transmitter that applies different channel aware scheduling algorithms are studied, investigating the impact of MU diversity and channel uncertainty on the performance of the scheduling algorithms.