Subcarrier Power Loading for Coherent Optical OFDM Optimized by Machine Learning

Enhanced power loading based on machine learned subcarrier interactions and nonlinear SNR progression is proposed to improve BER performance in optical coherent OFDM systems, achieving a gain of 0.5 dB in OSNR vs. classical schemes.

[1]  Octavia A. Dobre,et al.  Optimal bit and power loading for OFDM systems with average BER and total power constraints , 2012, 2012 IEEE Global Communications Conference (GLOBECOM).

[2]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[3]  Robert Hecht-Nielsen.  III.3 – Theory of the Backpropagation Neural Network* , 1992 .

[4]  Kwang Bok Lee,et al.  Transmit power allocation for BER performance improvement in multicarrier systems , 2004, IEEE Transactions on Communications.

[5]  J. Armstrong,et al.  OFDM for Optical Communications , 2009, Journal of Lightwave Technology.

[6]  Rafael Rios-Müller,et al.  Experimental estimation of optical nonlinear memory channel conditional distribution using deep neural networks , 2017, 2017 Optical Fiber Communications Conference and Exhibition (OFC).

[7]  Robert Hecht-Nielsen,et al.  Theory of the backpropagation neural network , 1989, International 1989 Joint Conference on Neural Networks.

[8]  Kaiming Liu,et al.  Adaptive power loading based on unequal-ber strategy for OFDM systems , 2009, IEEE Communications Letters.