An I-Q-Time 3-dimensional post-equalization algorithm based on DBSCAN of machine learning in CAP VLC system

In this paper, we propose a novel I-Q-Time 3-dimensional post-equalization scheme utilizing density-based spatial clustering of applications with noise (DBSCAN). The system performance can be successfully improved for a carrier-less amplitude and phase modulated (CAP) visible light communication (VLC) systems with the amplitude and phase erratic fluctuations. Furthermore, we experimentally demonstrate a 1.13 Gb/s 16-QAM CAP VLC system with a red chip LED over 1m indoor free space transmission under the hard decision-forward error correction (HD-FEC) threshold of 3.8 ×10−3, which is affected by severe fluctuations. The key parameters of the algorithm in 16-QAM CAP VLC systems are evaluated. By employing our proposed method, Q factor improves by 1.5∼2.5 dB in the system with fluctuations amplitude peak is 70% of the signal. As far as we know, this is the first time that the DBSCAN has been successfully applied to QAM modulated system.