A Neural Network-Based VLC Indoor Positioning System for Moving Users
In this paper, we present an indoor visible light communication (VLC) system to estimate the position of a moving user. This system uses two approaches based on received signal strength, trilateration estimation, and neural network estimation. In the VLC system, each transmitter sends its position information via light. A photo-detector receiver supported with the moving user is used to receive the transmitted power from each transmitter. The receiver position is calculated using the estimation of trilateration and the prediction of the neural network. We consider the sight line (LOS) and non-line of sight (NLOS) cases in our simulation. The results showed that the neural network estimation approach offers more accurate positioning than the trilateration estimation in case of the normal of the receiver has a tilted angle with the normal of the transmitter with 94% accuracy. © 2019 IEEE.