Mitigation of nonlinearities in analog radio over fiber links using machine learning approach

Machine learning (ML) techniques are looked upon as an innovative and realistic direction to cope up with nonlinearity issues in fiber optics communication. In this paper, a 64-quadrature amplitude modulation (QAM) based radio over (RoF) system is demonstrated for 10 km of standard single mode length utilizing support vector machine (SVM) method indicate effective mitigation front-hauls. The comparison SVM drawn conventional ML classifiers optimize symbol decision boundary that will reduce the RoF link impairments. results reported terms BER, Eye-linearity Quality factor.