Optimizing multiuser MIMO for access point cooperation in dense wireless networks

Abstract

As the usage of wireless devices continues to grow rapidly in popularity, wireless networks that were once designed to support a few laptops must now host a much wider range of equipments, including smart phones, tablets, and wearable devices, that often run bandwidth-hungry applications. Improvements in wireless local access network (WLAN) technology are expected to help accommodate the huge traffic demands. In particular, advanced multicell Multiple-Input Multiple-Output (MIMO) techniques, involving the cooperation of APs and multiuser MIMO processing techniques, can be used to satisfy the increasing demands from users in high-density environments. The objective of this thesis is to address the fundamental problems for multiuser MIMO with AP cooperation in dense wireless network settings. First, for a very common multiuser MIMO linear precoding technique, block diagonalization, a novel pairing-and-binary-tree based user selection algorithm is proposed. Second, without the zero-forcing constraint on the multiuser MIMO transmission, a general weighted sum rate maximization problem is formulated for coordinated APs. A scalable algorithm that performs a combined optimization procedure is proposed to determine the user selection and MIMO weights. Third, we study the fair and high-throughput scheduling problem by formally specifying an optimization problem. Two algorithms are proposed to solve the problem using either alternating optimization or a two-stage procedure. Fourth, with the coexistence of both stationary and mobile users, different scheduling strategies are suggested for different user types. The provided theoretical analysis and simulation results in this thesis lay out the foundation for the realization of the clustered WLAN networks with AP cooperation.