Rate-Splitting Multiple Access for Multi-Antenna Broadcast Channels with Statistical CSIT

TL;DR

This paper explores the use of rate-splitting multiple access (RSMA) in multi-antenna broadcast channels with only statistical channel state information, showing its advantages over traditional methods in fairness and robustness.

Contribution

It introduces a novel optimization of RSMA-based beamforming with statistical CSIT, demonstrating significant fairness and robustness improvements over existing schemes.

Findings

RSMA outperforms SDMA in max-min fairness rate.

RSMA is more robust to CSIT inaccuracies.

Numerical results confirm the rate gains in different scenarios.

Abstract

Rate-splitting multiple access (RSMA) is a promising technique for downlink multi-antenna communications owning to its capability of enhancing the system performance in a wide range of network loads, user deployments and channel state information at the transmitter (CSIT) inaccuracies. In this paper, we investigate the achievable rate performance of RSMA in a multi-user multiple-input single-output (MU-MISO) network where only slow-varying statistical channel state information (CSI) is available at the transmitter. RSMA-based statistical beamforming and the split of the common stream is optimized with the objective of maximizing the minimum user rate subject to a sum power budget of the transmitter. Two statistical CSIT scenarios are investigated, namely the Rayleigh fading channels with only spatial correlations known at the transmitter, and the uniform linear array (ULA) deployment with only channel amplitudes and mean of phase known at the transmitter. Numerical results demonstrate the explicit max min fairness (MMF) rate gain of RSMA over space division multiple access (SDMA) in both scenarios. Moreover, we demonstrate that RSMA is more robust to the inaccuracy of statistical CSIT.