Downlink Path-Based Precoding Space-Time Block Code in FDD Massive MIMO Systems Without CSI Feedback

For a massive multiple-input-multiple-output (MIMO) system operated under frequency-division duplex (FDD), downlink training was usually considered impractical due to huge amount of pilot signals and feedback overhead. Although some efforts have been spent to reduce such an overhead in recent works, the reduction is still limited. More recently, some precoding methods that do not require downlink training and channel state information (CSI) feedback have been proposed by recognizing some similarity between uplink and downlink channels. The base station may take advantage of such similarity and acquire partial knowledge of the downlink channels using previous uplink channel estimates. In this thesis, we propose to further exploit the diversity of a multipath channel in a downlink precoding method without CSI feedback. Specifically, space-time block code (STBC) is applied to enhance the robustness of downlink transmission based on the partial CSI at the transmitter (CSI-T). Simulation results suggest that the proposed method achieves a competitive performance to other methods with only the partial CSI-T.