Multiuser MIMO in Distributed Antenna Systems With Out-of-Cell Interference

TLDR: Several downlink multiuser multiple input multiple output (MIMO) DAS strategies are proposed and compared in terms of per-user throughput and area spectral efficiency and approximations of the ergodic rate are proposed to facilitate rapid simulation and design space exploration.

Abstract: Distributed antenna systems (DAS) augment the base station's transmit capability by adding multiple remote radio units, connected to the base station via a high bandwidth and low latency link. With DAS, the base station operates as if it had multiple antennas, but the antennas happen to be in different geographic locations. DAS have been shown to enhance coverage and capacity in cellular systems, in a variety of different configurations. This paper proposes, analyzes, and compares several downlink multiuser multiple input multiple output (MIMO) DAS strategies in terms of per-user throughput and area spectral efficiency. Zero-forcing transmit beamforming is used for transmission, the remote radio units may have one or more antennas, and the subscriber has a single receive antenna. Techniques considered include beamforming across all remote radio units (full transmission), using the same beamforming vector for each remote radio unit (simplified transmission), and selecting a subset of remote radio units. To facilitate rapid simulation and design space exploration, approximations of the ergodic rate are proposed for each technique assuming path-loss, small-scale Rayleigh fading, and out-of-cell interference. Simulations accounting for multiple interfering cells are used to compare the different transmission techniques. Full transmission is found to have the best performance even accounting for out-of-cell interference, though gains diminish for higher numbers of active users. Simplified transmission improves over no DAS but performance degrades with more active remote radio units.