The spectrum crunch is a global phenomenon, where wireless networks constrained by scarce spectrum resource cannot keep pace with the explosion in mobile broadband use, particularly at a time when smartphones and tablets are becoming even more prevalent and heavily used. Every new opportunity has to be maximally exploited to cope with this spectrum deficit and meet the demands of explosive broadband usage by pushing more data through existing spectrum. Massive multiple-input multiple-output (MIMO), an advanced antenna technology only developed in 2010 offers one such opportunity.
Massive MIMO enables a sparse infrastructure network, whereby a single base station (BS) is powerful enough to eliminate inter-cell interference through highly directional beamforming, and hence avoid the need for any cell-to-cell coordination. Initial work, particularly the experiments in have demonstrated the feasibility of massive MIMO. However, there is still lack of insightful understanding of the fundamental limits of massive MIMO, and also there is a large gap in the performance evaluation of massive MIMO under ideal and non-ideal practical conditions. The aim of this project is to establish a unified theoretical framework for the fundamental limits of massive MIMO with various practical constraints, and develop sophisticated signal processing algorithms to realize the concept of massive MIMO in realistic environments. The novelty of this project lies in the fact that advanced mathematical tools, such as random matrix theory and stochastic geometry, will be used to capture the dynamic nature of multi-user wireless channels. Sophisticated signal processing methods, such as game theoretic algorithms and compressed sensing, will be applied to massive MIMO in order to combat the practical constraints, such as frequency selective channel fading and limited channel feedback.
