Low-Complexity Adaptive Selection Beamforming for IRS-Assisted Single-User Wireless Networks
Muteen Munawar, Kyungchun Lee
Abstract
To maximize the signal strength in an intelligent reflecting surface (IRS)-assisted wireless network, semidefinite relaxation (SDR) and alternating optimization (AO) methods are widely used in the literature. The AO is more appealing owing to its low computational cost than SDR. This correspondence proposes an adaptive selection beamforming scheme for the IRS-assisted single-user communication systems, which requires a substantially lower computational complexity compared to the AO algorithm. Specifically, we ignore the channel gains of the IRS-user link, determine two low-cost sub-optimal active-beamforming vectors at the transmitter, and calculate the corresponding two passive-beamforming solutions at the IRS. Then, we calculate their total channel gains and the solution that yields a higher gain is selected as the final solution. The proposed scheme offers approximately a 60% reduction in computational complexity compared to the AO at the cost of slight performance degradation only for limited system configurations.