Indoor visible light positioning based on 3D ROI constraints and the improved crested porcupine optimizer

Indoor three-dimensional (3D) visible light positioning (VLP) suffers from non-line-of-sight multipath reflections and receiver-height uncertainty, which degrade positioning accuracy. To address these challenges, we propose a weighted least-squares-improved crested porcupine optimizer (WLS-ICPO) positioning method. The method first employs WLS for coarse localization to construct 3D region-of-interest (ROI) constraints and then applies ICPO within the ROI for refined optimization. ICPO incorporates a three-tier elite stratification strategy, independent Gaussian perturbation along the z-axis, and simplex-based local search, enabling a balance between global exploration and local exploitation. In a 5 m3×5 m3×3 m3 environment, simulation results show that WLS-ICPO achieves an average positioning error of 3.38 cm, representing reductions of 90.54% and 78.50% compared with WLS and conventional crested porcupine optimizer, respectively. The method also maintains low positioning errors across the tested signal-to-noise ratio levels, indicating its potential for indoor VLP applications.