Neural Network DPD for Aggrandizing SM-VCSEL-SSMF-Based Radio over Fiber Link Performance

This paper demonstrates an unprecedented novel neural network (NN)-based digital predistortion (DPD) solution to overcome the signal impairments and nonlinearities in Analog Optical fronthauls using radio over fiber (RoF) systems. DPD is realized with Volterra-based procedures that utilize indirect learning architecture (ILA) direct (DLA) becomes quite complex. The proposed method NNs evades issues associated ILA utilizes NN first model RoF link then trains NN-based predistorter by backpropagating through model. Furthermore, experimental evaluation carried out for Long Term Evolution 20 MHz 256 quadraturre amplitude modulation (QAM) 850 nm Single Mode VCSEL Standard Fiber establish a comparison between Memory Polynomial Generalized ILA. efficacy of examined reporting Adjacent Channel Power Ratio Error Vector Magnitude. findings imply NN-DPD convincingly learns which may not suit model, hence offer favorable trade-off terms computational overhead performance.