NOMA-Based MISO Visible Light Communication Systems With Optical Intelligent Reflecting Surface: Joint Active and Passive Beamforming Design

The burgeoning technology of optical intelligent reflecting surfaces (OIRS) offers significant potential within the realm of visible light communication (VLC) systems. This is primarily attributed to the capability of OIRS to exploit reflected propagation paths, which in turn mitigates the inherent signal blockage challenges encountered in VLC. In this paper, a non-orthogonal multiple access (NOMA)-based multiple-input single-output (MISO) VLC system with the aid of OIRS is investigated, in which multiple light-emitting diodes (LEDs) serve multiple users equipped with a photodetector (PD) simultaneously. To this end, an analysis of the VLC channel gains is undertaken, encompassing the evaluation of both the line-of-sight (LoS) and the OIRS-reflected paths. Subsequently, the problem is formulated with the objective of jointly optimizing the active beamforming at LEDs and the passive beamforming at the OIRS to maximize the achievable sum data rate, while considering the constraints of OIRS and the successive interference cancellation (SIC) process. However, the formulated problem is non-convex. To address this challenge, a block coordinate descent (BCD) algorithm is introduced to decompose the original joint beamforming problem into two sub-problems. Specifically, relaxed iterative algorithms based on semi-positive definite relaxation and Taylor expansion are employed to solve these sub-problems. Additionally, a permutation-based genetic algorithm is proposed to tackle the decoding order problem. Meanwhile, the simulation results illustrate the improvement in the sum data rate achieved by the proposed algorithm, providing valuable insights for future research on OIRS-aided VLC systems.