Abstract:Nonlinear distortions in passive optical network (PON) systems degrade signal transmission performance. Joint linear and nonlinear pre-equalizers based on the Volterra series, memory polynomials and artificial neural networks (ANNs) are utilized to mitigate linear and nonlinear impairments. Centralized pre-equalization at the transmitter side of the optical line terminal (OLT) keeps the receiver side digital signal processing (DSP) at the optical network unit (ONU) simple. In this study, performance assessments of these three different pre-equalization schemes in intensity modulation and direct detection (IM/DD)-based 25-GBaud PON systems were evaluated. The joint linear and nonlinear pre-equalization schemes were based on the weight coefficients and the model obtained by channel estimation on the receiver side. We studied and compared the performance of three pre-equalization methods under various implementation conditions, such as received optical power (ROP), number of equalizer taps, and semiconductor optical amplifier (SOA) current. Besides, we studied the equalization performance under different modulation orders, which were applied for data rate optimization to support different power budgets, being the potential solutions for future PON systems with various fiber-reach and split-ratio distributions. As a proof-of-concept, the equalization performance of the 25-Gbaud signals modulated with non-return-to-zero (NRZ), four-level pulse amplitude modulation (PAM-4) and PAM-8 formats, corresponding to 25-, 50-, and 75-Gbps data rates, respectively, were tested. Both numerical simulations and experimental demonstrations were conducted to assess the performance of different pre-equalization schemes. With ANN pre-equalization, more than 4dB improvement in power budget was achieved for 25-GBaud PAM4 and PAM8 signals after 20 km fiber transmission. ANN-based pre-equalizer performed better than ANN-based post-equalizer even when the received signals were in low SNR. The power budgets of the 25-Gbaud NRZ, PAM4 and PAM8 signals after 20 km fiber transmission were 36.2 dB, 29.5 dB and 18 dB, respectively.

Fixed-Point Analysis and FPGA Implementation of Deep Neural Network Based Equalizers for High-Speed PON

Numerical Study of Parallel Optoelectronic Reservoir Computing to Enhance Nonlinear Channel Equalization

Complex principal component analysis-based complex-valued fully connected NN equalizer for optical fibre communications

Achieving High Throughput with a Trainable Neural-Network-Based Equalizer for Communications on FPGA

OAM mode-division multiplexing IM/DD transmission at 4.32 Tbit/s with a low-complexity adaptive-network-based fuzzy inference system nonlinear equalizer

Efficient FPGA Implementation of an Optimized SNN-based DFE for Optical Communications

Fully-blind Neural Network Based Equalization for Severe Nonlinear Distortions in 112 Gbit/s Passive Optical Networks

Real-Time FPGA Demonstrator of ANN-Based Equalization for Optical Communications

Low Complexity Neural Network Equalization Based on Multi-Symbol Output Technique for 200+ Gbps IM/DD Short Reach Optical System

CNN-Based Equalization for Communications: Achieving Gigabit Throughput with a Flexible FPGA Hardware Architecture

Performance Assessments of Joint Linear and Nonlinear Pre-Equalization Schemes in Next Generation IM/DD PON

A Novel Machine Learning-based Equalizer for a Downstream 100G PAM-4 PON

Parallel Adaptive Equalizer for Alamouti-Coded Signals Recovery in Simplified Coherent PON

Neural Turbo Equalization: Deep Learning for Fiber-Optic Nonlinearity Compensation

Deep Learning Equalizer Connected with Viterbi-Viterbi Algorithm for PAM D-Band Radio over Fiber Link

Advanced Equalization in 112 Gb/s Upstream PON Using a Novel Fourier Convolution-based Network

Low Computationally Complex Recurrent Neural Network for High Speed Optical Fiber Transmission

AdaNN: Adaptive Neural Network-based Equalizer via Online Semi-supervised Learning for High-speed Optical Fiber Communication

Deep Belief Network-Hidden Markov Model Based Nonlinear Equalizer for VCSEL Based Optical Interconnect

FPGA implementation of power-lite Volterra-inspired neural network equalizer in 180-Gb/s net bitrate IMDD short-reach optical system

Demonstration of photonic neural network for fiber nonlinearity compensation in long-haul transmission systems