Contributions to massive MIMO: power allocation, user selection, and cell-free communications

Massive multiple-input multiple-output (MIMO) is seen as a key enabler for nextgeneration wireless communication systems, being capable of delivering very high throughput and reduced inter-cell interference by using spatial multiplexing and simple linear processing. In this thesis, we present new contributions to the massive MIMO field, covering aspects related to power allocation, user selection, and cell-free systems. Regarding power allocation, we propose a new initialization scheme for the max-min fairness power allocation algorithm used in massive MIMO systems under line-of-sight (LoS) propagation. The proposed scheme guarantees convergence to the optimal point. Additionally, we propose the bound test procedure as an even more efficient way of initializing the search interval. As for the user selection, we provide a review of user selection algorithms for massive MIMO systems under LoS propagation, highlighting the role of the favorable propagation condition in this scenario and the design of common linear precoding techniques. Although the LoS propagation is extremely important to some promising technologies, like in millimeter-wave communications, massive MIMO systems are rarely studied under this propagation model. Moreover, we propose two new user selection algorithms. The first one, called inter-channel interference-based selection (ICIBS), takes the global interference in the system into account, whereas the second one, namely fading-ratio-based selection (FRBS), considers both the global interference and the large-scale fading. These new algorithms present advantages over the algorithms in the literature, outperforming them in some scenarios, such as clustered and crowded scenarios. Simulation results show that when the ICIBS and FRBS are compared to the competing algorithms, the proposed approaches provided improved maximum throughput and 95%-probability throughput in some scenarios. For the cell-free massive MIMO systems, we evaluate the uplink performance of cell-free massive MIMO systems corrupted by quantization noise with correlated Rician fading channels. We present an expression for the uplink spectral efficiency for any linear combiner. Furthermore, we derive the optimal minimum mean squared error channel estimator and present a suboptimal scalable combiner.