Deep Neural Network Equalization for Optical Short Reach Communication

Abstract

Nonlinear distortion has always been a challenge for optical communication due to the nonlinear transfer characteristics of the fiber itself. The next frontier for optical communication is a second type of nonlinearities, which results from optical and electrical components. They become the dominant nonlinearity for shorter reaches. The highest data rates cannot be achieved without effective compensation. A classical countermeasure is receiver-side equalization of nonlinear impairments and memory effects using Volterra series. However, such Volterra equalizers are architecturally complex and their parametrization can be numerical unstable. This contribution proposes an alternative nonlinear equalizer architecture based on machine learning. Its performance is evaluated experimentally on coherent 88 Gbaud dual polarization 16QAM 600 Gb/s back-to-back measurements. The proposed equalizers outperform Volterra and memory polynomial Volterra equalizers up to 6th orders at a target bit-error rate (BER) of 10 − 2 by 0.5 dB and 0.8 dB in optical signal-to-noise ratio (OSNR), respectively.

Coupled Transceiver-Fiber Nonlinearity Compensation Based on Machine Learning for Probabilistic Shaping System

“…Another observation from Fig. 9 is that the constellation remains Gaussian-shaped distortion after NLC-ANN, which is not the case for some previous ANN's works [11], [17]. This may be explained by the observation that the nonlinear regression is smoother for higher order QAM targets, i.e.…”

“…Under the impact of fiber nonlinearity, other nonlinear equalizers based on inverse Volterra series transfer functions may also be deployed to partially invert the nonlinear distortion induced by the transmission link. However, the Volterra-based nonlinear compensation (NLC) has shown worse performance than an optimized machine learning and their complexity is also high [10], [11]. Recently, a supervised machine-learningbased technique, namely artificial neural network (ANN), has been proposed and studied for uniform 64-QAM as a predistortion compensator for impairments induced by a low resolution DAC, but ignoring other nonlinear effects [12].…”

Experimental Verification of Complex-Valued Artificial Neural Network for Nonlinear Equalization in Coherent Optical Communication Systems

“…1) against the dynamic neural network (DNN), suggested for NLC in [11] , and the classic DBP [21] . To ensure a fair comparison, the employed DNN had two layers with 192 neurons each, as in [7] , the same length of input symbol sequences as our NN, and operated on the symbols sequences from both polarizations Y H k , Y H k , as in [22] . The DBP operated on 2 samples/symbol signal with 2 steps per span (StPS).…”

Performance Versus Complexity Study of Neural Network Equalizers in Coherent Optical Systems

“…The first and, perhaps, simplest and well-studied NN-based equalizer that we consider is the MLP, proposed for the shorthaul coherent system equalization in [22] and the long-haul systems in [30]. The MLP is a deep feed-forward densely connected NN structure that handles the I/Q components for each polarization jointly, providing two outputs for each processed symbol: its real and imaginary parts.…”