Intelligent Reflecting Surface Enabled Wireless System with Antenna Selection at Source Under Transceiver Hardware Impairments

Abstract:
We elucidate the reliability of detection in a wireless system in which the source performs antenna selection (AS) and is aided by an intelligent reflecting surface (IRS) comprising of N elements in communicating to a user under transceiver hardware impairments. To this end, we deduce the probability density function of the effective channel gain for a source with L antennas and a single radio frequency (RF) chain. We derive a novel and tight approximation for the bit error rate (BER) of MPSK/MQAM for this system under distortion noise arising due to imperfect RF chains. We extend our study to evaluate the impact of subset AS and discrete phase shifts on BER. We show that BER improves either by increasing L or N while keeping number of RF chains fixed at one. We prove that by adding a few more passive IRS elements: savings of radiated power can be obtained, a larger transmit-receive distortion noise can be tolerated while keeping the BER fixed, and system with single RF chain can achieve BER identical to a source that performs subset AS. We also show that BER with 8 phase shift levels at IRS is close to the BER that can be obtained with an infinite precision IRS.