Nonlinear Equalization Based on Artificial Neural Network in DML-Based OFDM Transmission Systems

• Published in: Journal of lightwave technology Vol. 39; no. 1; pp. 73 - 82
• Main Authors: Huang, Wei-Hsiang, Nguyen, Hong-Minh, Wang, Chung-Wen, Chan, Min-Chi, Wei, Chia-Chien, Chen, Jyehong, Taga, Hidenori, Tsuritani, Takehiro
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Cost function; Equalization; Equalizers; Fading channels; intensity modulation; Neural network; Neural networks; Nonlinear distortion; Nonlinear optics; OFDM; Orthogonal Frequency Division Multiplexing; Pins; Signal to noise ratio; Waveforms
• ISSN: 0733-8724; 1558-2213
• DOI: 10.1109/JLT.2020.3025163

This article reports the application of an equalizer based on an artificial neural network (ANN), in the form of nonlinear waveform regression, to mitigate nonlinear impairments in directly modulated laser (DML)-based orthogonal frequency-division multiplexing (OFDM) optical transmission. Experiments involving transmission over 0-200 km demonstrate that using an ANN with one hidden layer can greatly reduce nonlinear distortion. The proposed scheme outperformed a Volterra nonlinear equalizer at transmission distances exceeding 25 km. Using a 10G-class DML, the proposed scheme achieved the following data rates: 39.2 Gbps at 100 km (an improvement of 59%) and 33.5 Gbps at 150 km (an improvement of 57%). We also modified the cost function of the ANN during the training procedure to overcome the poor signal-to-noise ratio of the original ANN at low frequencies. This resulted in <inline-formula><tex-math notation="LaTeX">></tex-math></inline-formula>30-Gbps transmission over 0-200 km.