Modulation Classification Method Based on Clustering and Gaussian Model Analysis for VLC System

In the letter, a novel automatic modulation classification (AMC) method is proposed and experimentally demonstrated in orthogonal frequency division multiplexing (OFDM) visible light communication (VLC) system. For the proposed AMC method, the classification feature values are obtained by clustering and Gaussian model analysis. Then these classification feature values are used to classify the modulation format. In addition, the recognition accuracy at different optical signal-noise-ratio (OSNR) is investigated in OFDM-VLC system. The experimental results show that, for 4QAM, 16QAM, 32QAM, and 64QAM with the proposed method, it can be accurately recognized with 2000 symbols at OSNR of 4 dB, 11 dB, 13 dB, and 18 dB, respectively.