Algoritmi și proceduri inovative în sisteme de comunicații digitale 5G / Innovative algorithms and procedures for 5G digital communications systems

In telecommunications domain, 5G is the 5th generation of cellular network technologies. The communication systems with multiple antennas [Multiple-Input-Multiple-Output (MIMO)] are used in 5G networks to increase the capacity of the radio link. At the very core of the MIMO systems is digital signal processing. Due to that, the main objective of the thesis is to optimize the performance of a hybrid precoding scheme through the reduction of the complexity required by the identification of the spatially separable users. This led to the implementation of a statistical user grouping algorithm, using the spatial compatibility of the users. The proposed algorithm only uses the direction of the users, and not their channel quality. This way, the quantity of information required by channel estimation and channel quality feedback, is reduced. As such, the method is compatible with Massive MIMO 5G systems that can accurately determine the direction of the users through the embedded beam management procedures, and that try to minimize the quantity of information required for channel status computation. For an accurate study of the precoding scheme and of the grouping algorithm performance, a method of generating a realistic user traffic distribution, based on 3D geolocated call traces, is used. It is shown that using a realistic user traffic distribution, the computational complexity of the algorithm grows linearly with the number of users, and the user grouping optimizes the spectral efficiency of the hybrid precoding scheme, which is now similar to that of the nonlinear precoding scheme, but has a reduced complexity.