Statistical 3-D Beamforming for Large-Scale MIMO Downlink Systems Over Rician Fading Channels

In this paper, we investigate a downlink transmission algorithm for single-cell multiuser systems with two-dimensional (2-D) large-scale antenna array at the base station (BS) over Rician fading channels. We first derive some properties of the channel's line-of-sight (LOS) component in the large-scale antenna array scenario. Next, based on these properties and under the assumption of only statistical channel state information (CSI), i.e., the LOS component and Rician K-factor, at the BS, we derive the optimal beamforming vector for each user in maximizing an approximation of the ergodic sum rate. The main guidelines for user scheduling are also presented. Then, a three-dimensional (3-D) beamforming downlink transmission algorithm exploiting only the statistical CSI of each user is proposed. For this algorithm, we derive an exact analytical closed-form expression for the achievable ergodic rate and present tractable approximations. Based on our analysis, we gain some valuable insights. The proposed algorithm is shown to perform well in achieving considerable sum rate while requiring much less CSI at the BS. It requires three scalar values at the BS for each user, and can achieve an ergodic sum rate closer to the ergodic sum rate achieved by the matched-filter precoding with perfect CSI at the BS.