İç ortam çok kullanıcılı VLC sistemlerde tek katmanlı RSMA performans analizi

Visible Light Communication (VLC) has emerged as a promising complement to traditional radio-frequency (RF) wireless technologies, particularly for high-capacity indoor networks. However, the dense deployment of users and luminaries in indoor environments introduces significant interference and resource management challenges. Rate-Splitting Multiple Access (RSMA) has recently been advocated as a robust multiple access paradigm capable of managing interference and enhancing spectral efficiency, yet its potential in realistic VLC scenarios remains underexplored. This thesis investigates the applicability and performance of one-layer RSMA in downlink multi-user MISO VLC systems under practical hardware constraints. A physically accurate Lambertian line-of-sight (LOS) channel model is adopted, explicitly considering LED current linearity and illumination requirements. The joint precoder and power allocation problem is formulated and solved using numerical optimization, with two simulation frameworks: (i) a random user placement configuration benchmarking RSMA against power-domain NOMA (PD-NOMA) under varying transmit powers and user loads, and (ii) a spatial sensitivity analysis mapping spectral efficiency with respect to user locations. Comprehensive simulation results demonstrate that RSMA consistently outperforms PD-NOMA across all evaluated user densities and power budgets. At low optical power, RSMA yields relative spectral efficiency gains up to 50% for two users and over 300% for seven users; at higher power, substantial improvements persist, especially for denser deployments. Detailed analyses reveal the impact of optical transmit power, user density, spatial distribution, and optical hardware parameters on system performance. Spatial heatmap analyses further highlight the significance of user placement and luminaire geometry in practical VLC deployments. The findings confirm RSMA as a flexible and effective multiple access strategy for next-generation indoor VLC systems, with substantial gains under realistic constraints. Future work should address channel uncertainty, non-line-of-sight conditions, experimental validation, and the integration of machine learning for adaptive resource management.