Principal Component Analysis for Robust Region-of-Interest Detection in NLOS Optical Camera Communication

In Optical Camera Communication (OCC), the transmitted data sent through the brightness variation of a Light-Emitting Diode (LED) is detected by a camera rather than a conventional photodiode. Thus the major issue in the process of recovering the transmitted signal recorded in images is the detection of the Region-Of-Interest (ROI), corresponding to pixels that are actually containing information. To achieve this, the Principal Component Analysis is for the first time to the authors best knowledge, applied to images obtained through rolling shutter effect in order to extract the transmitted signal. Obtained results clearly show an improvement in the robustess of ROI detection compared to other methods, increasing by 123.7 % the achievable communication distance considering a raw Bit Error Rate (BER) of 10
⁻³
.