Projeto de Codebook para Sistemas MIMO Analógicos em Ondas Milimétricas

Abstract

The use of MIMO (Multiple-Input Multiple-Output) in millimeter wave systems allows the implementation of sets of antennas with a high number of elements, both in the transmitter and in the receiver. Current hardware limitations make the use of analog beamforming a natural design choice for these modern communication systems. Thus, analog beamforming is often implemented through a codebook (or set of vectors) with predefined phase adjustments, which must be carefully chosen in order to improve system performance. In order to address the problem of modeling a codebook for millimeter-wave MIMO systems, this work proposes a methodology that takes into account propagation channel data based on semi-deterministic estimation techniques, which are more accurate than statistical techniques. Such techniques make it possible to approximate the representation of the “real world” of the propagation environment through computational systems (or “virtual world”) using electromagnetic theory and numerical techniques. Based on this information, this work proposes the modeling of a codebook linked to the environment that provides a better beamforming gain. In this way, Lloyd’s generalized algorithm is applied to a set of samples of training channels generated from a dataset of vehicular networks. The approach also considers the hardware constraints of analog beamforming. In a next step, with a set of test channels, the estimated codebook is tested and compared with the DFT codebook with oversampling factor. The simulation results highlight the ability of the estimated codebook to provide better beamforming gains with a reduced size codebook.