Indoor localization for mobile devices can employ beacons inserted in known positions in the environment. Light-emitting diodes (LEDs) are becoming popular beacons because of their simplicity; however, using the light for localization presents problems due to unexpected obstacles.
This paper proposes a localization framework that addresses the main problem of visual light communication, namely removing information coming from LEDs that are blocked by obstacles. The solutions of a set of equations and inequalities on the geometric relationships between the position of the mobile device and the received signal strength of the LEDs produce the 3D coordinates of the mobile device. To reduce the localization error, triangle inequalities are applied to detect the LEDs that are blocked by obstacles, and those LEDs are ignored.
Simulation results using five LEDs are analyzed with various disturbance levels. The error reduction is significant; however, the computation time has an acceptable twos value only when the step_size to enumerate the possible locations is 0.1 m.
The paper is of interest to researchers in cyber-physical systems (CPS), and perhaps in autonomous robotics. The algorithm is described in such a way that students can reproduce and try it.