The modulation classification methods in PPM–VLC systems

Intelligent methods have been applied to many fields for a long time. Recently, Visible Light Communication (VLC) systems widely include learning and classification models improve their performances. The of L-Pulse Position Modulation (L-PPM) formats is crucial VLC since the modulation order L very effective providing energy efficiency increasing transmission capacity. In this paper, therefore, it reported first time, L-PPM schemes in by using Decision Tree, K-nearest neighbor (KNN), Support Vector Machine, Direct Decision-based Linear Model technique. A novel feature extraction model derived be able classify type schemes. comparison has given observe performance taking into account level Signal Noise Ratio distance between receiver transmitter. 2-PPM scheme highest success rate against other terms metrics at overall distance. Additionally, KNN method gives best accuracy 2.32 m more, while obtained 97.85% Tree 2.20 m.