Indoor Visible Light Positioning Based on Image Denoising and Phase Difference Vector Fingerprint

Visible light positioning (VLP) technology has become a promising technology in the field of indoor positioning benefits from its greenness and low cost. However, the traditional fingerprint construction scheme in radio frequency cannot be effectively applied in the VLP system and the positioning accuracy of the fingerprint-based VLP system suffers from its specific propagation channel noise. Therefore, a novel fingerprint construction method based on phase difference vector (PDV) and fingerprint denoising network (FDNet) is proposed to improve the stability and positioning accuracy of the VLP system. Specifically, the arrival phase difference information is considered to construct a fingerprint image, and the improved denoising network is designed to obtain relatively high signal-to-noise ratio fingerprint information. Moreover, the positioning is achieved by mapping the constructed fingerprint database to location information through a back propagation (BP) network. The simulation results show that the proposed PDV-based FDNet positioning method achieves an average positioning accuracy of 1. 43cm, which outperforms the existing benchmark methods.