Visible light indoor positioning via an iterative algorithm based on an M5 model tree.

In this work, a novel iterative algorithm based on an M5 model tree (M5Tree) is proposed to realize indoor positioning using a single LED and multiple photodetectors. A visible light positioning system was implemented to evaluate the algorithm performance, including the effect of the maximum depth of the tree and the number of iterations on positioning accuracy. Finally, in the same single LED positioning system, the traditional machine-learning algorithms, being k-nearest neighbor and support vector regression, were compared with the M5Tree algorithm. The simulation results demonstrate that the proposed algorithm has good positioning performance on a single LED positioning system.