Bidirectional mapping positioning method based on motion information in VLP systems

Abstract

Visible light positioning signals are susceptible to indoor environmental factors, leading to signal instability that hinders accuracy improvement. To mitigate the limitations of information obtained at a single location, a bidirectional mapping (BDM) positioning method based on motion information is proposed. This algorithm combines channel state information (CSI) received at multiple locations with motion information and uses bidirectional mapping to construct multiple location sets with spatial relationships. Combining the positioning information of the current position with that of the multiple previous positions, the estimation of the current position is inferred by iteratively calculating the cumulative deviation between multiple continuous positions, instead of relying on the single position information for positioning, which enhances the robustness and accuracy of positioning. In addition, to reflect the different contributions of positioning information at different locations to the cumulative deviation, a weighted BDM positioning method is presented to further improve positioning accuracy. In a 4 m×4 m×3 m indoor multipath simulation environment, the average positioning error and root mean square positioning error of the proposed method are 1.55 cm and 1.79 cm without IMU error, respectively, when using joint positioning information from 10 consecutive locations during motion. Simulation results show that this method significantly and robustly mitigates significant positioning errors introduced by environmental factors, offering a new potential approach to positioning optimization.