We consider indoor positioning based on visible light where the receiver adopts a commercial smartphone camera. Two positioning approaches are proposed, either the light emitting diode (LED) positions are known or not. When the LED positions are known, the LED light signal intensities are measured to identify the identity of each LED in the image. Triangular similarity is adopted to estimate the receiver position. When the LED positions are unknown, we develop a shift rotation model on the receiver movement and further propose a novel indoor positioning algorithm. The algorithm estimates the rotation center in the image instead of treating the image center as the rotation center, leading to reduced positioning error. To achieve high accuracy, a location reference point is set such that the positioning errors of multiple LEDs have little effect. According to the experimental results, the average positioning error can be reduced to 1 cm, which outperforms the reported experimental results with receiver-ceiling distance larger than 2 m.
展开▼
