SVM-based detection in visible light communications

A support vector machine (SVM)-based data detection for 8-superposed pulse amplitude modulation and direct-current-biased optical orthogonal frequency division multiplexing in visible light communication is proposed and experimentally demonstrated. In this work, the SVM detector contains multiple binary classifiers with different classification strategies. The separating hyperplane of each SVM is constructed by means of the training data. The experiment results presented that the SVM detection offers improved bit error rate performance compared with the traditional direct decision method.