Degrés de Liberté des Communications MIMO en Visibilité Directe

Future wireless networks are envisioned to support new applications such as the industrial Internet of Things (IIoT), which demands unprecedented levels of throughput, reliability, number of supported devices, and low-energy consumption. The current trends for addressing some of these requirements include cell densification, advanced multiple-input multiple-output (MIMO) communication, and the use of higher frequency bands, such as the millimeter waves (mmWaves) and terahertz (THz) frequencies. The integration of these technology enablers implies that the practical use cases are primarily short-range links between electrically large multi-antenna transceivers in which the line-of-sight (LoS) components are predominant. In such cases, the conventional frameworks to analyze and design the MIMO communication systems based on the far-field propagation are no longer applicable. Conversely, it is required the development of new frameworks and designs for near-field communication based on spherical wavefront propagation. This thesis evaluates the spatial multiplexing capabilities of near-field MIMO communication systems.Firstly, it introduces an analytical framework for analyzing a LoS communication link between two holographic surfaces (HoloSs). This framework is obtained by using a quartic approximation for the wavefront propagation for paraxial deployments. It is also extended for non-paraxial deployments by partitioning the large surfaces into sufficiently smaller sub-surfaces that fit in the paraxial regime. Based on Landau's eigenvalue theorems, a closed-form expression for the number of degrees of freedom (DoF), i.e., the number of highly-coupled communication modes, is obtained, and the intensity of those modes is characterized. Furthermore, for specific practical deployments, analytical expressions for the optimal waveforms for encoding and decoding the data are obtained.Secondly, the resulting analytical expressions for the communication between HoloSs are utilized to develop a low-complexity design to optimize a reconfigurable intelligent surface (RIS)-aided MIMO communication system. Implementations based on diagonal and non-diagonal RISs are considered. Over line-of-sight channels, the performance obtained by the proposed schemes is close to the one obtained by using complex numerical methods.Finally, the framework used for spatially continuous HoloS is adapted for the case of discrete antenna arrays, specifically for planar and linear movable antenna arrays. It allows the optimization of the antenna array design to maximize the spatial multiplexing gain without increasing the energy consumption. For paraxial setups, the framework provides sufficient conditions for the optimal design expressed in terms of key design parameters, which are easier to interpret than those available in the literature. The framework is extended to non-paraxial setups by adopting an array of sub-arrays architecture for non-paraxial setups. In the case of linear arrays, it demonstrates that the optimal design can be achieved for some specific setups if a set of sufficient conditions are fulfilled.