Achieving large spectral efficiencies from MU-MIMO with tens of antennas: Location-adaptive TDD MU-MIMO design and user scheduling

Abstract
We consider schemes for joint scheduling training and downlink MU-MIMO in multi-cell deployments that can yield high cell and cell-edge throughputs with tens of antennas per base-station and single antenna terminals. TDD schemes are considered as they are able to support MU-MIMO over large antenna arrays despite the inherent training dimensionality bottleneck. The proposed schemes operate by dividing users into bins based on their relative location (pathloss) to different stations. MU-MIMO is then performed by splitting transmission resources among such bins, and optimizing the MU-MIMO design on a per-bin basis. We demonstrate the viability of these schemes in the context of a one-dimensional uniform cellular topology and show that high cell and cell-edge rates can be expected, provided the palette of per-bin signaling options is sufficiently large.

This publication has 6 references indexed in Scilit: