irs assisted rayleigh fading miso systems beamforming and asymptotic analysis

• DOI: 10.14288/1.0391010
• Publication Date: 2020-01-01
• Authors: Bayan Alnahhas

The recent concept of intelligent reflecting surfaces (IRS) enabled wireless communication can overcome the adverse effects of the wireless channel by manipulating the propagation of the radio waves in the environment. This emerging technology, which relies on low-cost passive reflecting elements, is shown to yield promising performance results in recent studies that often focus on Rayleigh fading environments and assume the availability of perfect channel state information (CSI) at the base station (BS) and the IRS. This work considers the practical setting where only imperfect CSI is available at the BS and the IRS, and studies the downlink sum-rate performance of a multi-user IRS-assisted multiple-input single-output (MISO) system in the asymptotic regime where the number of BS antennas, IRS elements and users grow large. We first derive the MMSE estimates of the IRS-assisted and direct channels under a protocol that ensures that the IRS element stay passive. Then we formulate the signal-to-interference-plus-noise ratio (SINR) and sum rate expressions under maximum ratio transmission (MRT) precoding, and obtain their asymptotically tight deterministic approximations using tools from random matrix theory. The asymptotic performance limits are also derived that reveal that in Rayleigh fading environments, IRS only yields an array gain and loses on the reflect beamforming gain promised by existing works. Simulation results show the tightness of the asymptotic expressions for moderate system dimensions and reveal important insights into how beneficial the IRSs are when the environment has rich scattering as represented in Rayleigh fading. It turns out that IRS introduces high sum- rate gains in noise-limited scenarios whereas the gains start to diminish in interference-limited scenarios.