Advancing multi-antenna technologies for 6G: rate-splitting multiple access, (cell-free) Massive MIMO, reconfigurable intelligent surfaces

Abstract

The advent of sixth-generation (6G) wireless communications is expected to significantly boost the key performance indicators (KPIs) of fifth-generation (5G) networks and ensure ubiquitous wireless connectivity. To that end, multi-antenna technologies are expected to be a fundamental enabler of the 6G networks. These technologies leverage multiple antennas to exploit the spatial dimensions of the wireless channel, thereby offering substantial improvements to spectral efficiency (SE), energy efficiency (EE), coverage etc., of the network. Multi-antenna technologies, however, are not without their challenges. The reliance on highly accurate channel state information at the transmitter (CSIT), energy inefficiency due to a high number of active radio-frequency (RF) chains to enable aggressive spatial multiplexing, and peculiarities of the use-cases themselves are a few to mention. Therefore, it is imperative to pursue transmission strategies or multiple access schemes that measure up to the aforementioned challenges and advance multi-antenna technologies. Following this, the thesis proposes pre-processing techniques for diverse multi-antenna technologies in downlink scenarios to enhance the SE of the network. First, we consider a multi-user multiple-input multiple-output (MIMO) downlink scenario and propose a transmission framework for a robust scheme named rate-splitting multiple access (RSMA). We investigate the sum degrees-of-freedom (DoF), ergodic sum-rate (ESR) and link- level throughput performance of RSMA under different CSIT quality regimes and network loads. Next, motivated by the robustness of RSMA to imperfect CSIT in MIMO settings, we investigate the performance of RSMA as a pilot contamination mitigation strategy in time- division duplex (TDD) Massive MIMO networks. Building on this work, we next integrate RSMA with TDD Cell-Free Massive MIMO to enable random access in massive machine type communications (mMTC). Finally, to reduce the number of active RF chains in Massive MIMO downlink transmission, we utilize a beyond-diagonal reconfigurable intelligent surface (BD-RIS), and design a space division multiple access (SDMA)-based transmission strategy.