Optimized Orientation and Beamforming for Intelligent Omni-Surface: Meta-Heuristic Approach

Abstract

We consider a downlink multi-user non-line-of-sight communication setting, which is supported by an intelligent omni-surface (IOS) with adjustable orientation. IOS can serve users on both sides of the surface by independently controlling the reflecting and transmitting phase shifts of the IOS elements. We study a unified framework for maximizing the sum spectral efficiency of the system. In this framework, we optimize the azimuth and elevation angles of the IOS, the reflection and transmission coefficients of the IOS, and the base station beamforming. To that end, we propose a novel meta-heuristic approach to optimizing the orientation angles of the IOS using a particle swarm algorithm. Furthermore, we employ WMMSE and alternating optimization to adjust the hybrid beamforming and energy-splitting factor. Our numerical results show significant gains over conventional MIMO and IRS-assisted networks. The results show a gain of 2.5 b/s/Hz over the IOS-based system with a fixed orientation.