Opportunistic Beamforming using an Intelligent Reflecting Surface\n Without Instantaneous CSI

While intelligent reflecting surface (IRS) assisted wireless communication\nhas emerged as an important research paradigm, channel state information (CSI)\nacquisition remains a critical challenge to design the IRS phase-shifts and\nyield the promised coherent beamforming gains. In this paper, we propose an\nIRS-assisted opportunistic beamforming (OBF) scheme under proportional fair\nscheduling, which does not require instantaneous CSI to design the IRS\nparameters. In a slow-fading environment, we show that with only random\nrotations at the IRS, the proposed scheme can capitalize on the multi-user\n(MU)-diversity effect to approach the performance of coherent beamforming as\nthe number of users grows large. Next we study the sum-rate scaling of\nIRS-assisted OBF in the correlated Rayleigh fast fading environment under a\ndeterministic beamforming scheme that results in a considerable sum-rate\nimprovement.\n