Simulation and Experimental Validation of Optical Camera Communication

While simulation tools for visible light communication (VLC) with photo detectors (PDs) have been widely investigated, similar tools for optical camera communication (OCC) with complementary metal oxide semiconductor (CMOS) sensors are lacking in this regard. Camera based VLC systems have much lower data rates owing to camera exposure times. Among the few extant OCC simulation tools, none allow for simulation of images when exposure time is greater than the signal period. An accurate simulation of the OCC system can be used to improve the data rate and quality of performance. We propose a simple simulation technique for OCC which allows to test for system performance at frequencies beyond the camera shutter speed. This will allow much needed data rate improvement by operating at the actual frequency a decoding algorithm ceases detection instead of the exposure limit used now. We have tested the accuracy of simulation by comparing the detection success rates of simulated images with experimental images. The proposed simulation technique was shown to be accurate through experimental validation for two different cameras.

Tree based Single LED Indoor Visible Light Positioning Technique

Visible light positioning(VLP) has gained prominence as a highly accurate indoor positioning technique. Few techniques consider the practical limitations of implementing VLP systems for indoor positioning. These limitations range from having a single LED in the field of view(FoV) of the image sensor to not having enough images for training deep learning techniques. Practical implementation of indoor positioning techniques needs to leverage the ubiquity of smartphones, which is the case with VLP using complementary metal oxide semiconductor(CMOS) sensors. Images for VLP can be gathered only after the lights in question have been installed making it a cumbersome process. These limitations are addressed in the proposed technique, which uses simulated data of a single LED to train machine learning models and test them on actual images captured from a similar experimental setup. Such testing produced mean three dimensional(3D) positioning error of 2.88 centimeters while training with real images achieves accuracy of less than one centimeter compared to 6.26 centimeters of the closest competitor.