Nonlinear equalization based on pruned artificial neural networks for 112-Gb/s SSB-PAM4 transmission over 80-km SSMF

Abstract

A pruning method of artificial neural network based nonlinear equalizer (ANN-NLE) is proposed and validated for single-sideband 4-ary pulse amplitude modulation (SSB-PAM4) in IM/DD system. As a classifier, ANN is capable to form a complex nonlinear boundary among different classifications, which is considered as an appropriate way to mitigate the nonlinear impairments in optical communication system. In this paper, first, we introduce the operation principle of the traditional linear equalizer (LE) and NLE such as volterra equalizer (VE). Then we make an analogy among the LE, VE and ANN-NLE. After that, a novel pruning method is applied to reduce the complexity of ANN. The BER performance of ANN-NLE outperforms VE after fiber transmission. After 60 km fiber transmission, ANN-NLE decreases the BER by about one order of magnitude compared to VE. By implementing the proposed pruning method, the connections of ANN reduced by a factor of 10x while keeping the BER under the threshold of 3.8x10.

Recent Advances in Equalization Technologies for Short-Reach Optical Links Based on PAM4 Modulation: A Review

“…In recent years, it has been proposed to mitigate the nonlinear impairments in optical communication system [89][90][91]. For short-reach PAM4 optical links, various research concerning the NN method has been performed to improve transmission performance [43][44][45][46][47][48][49][50][51][52][53]. The schematic of NN based nonlinear signal processing is presented in Figure 19a, where the leftmost part consists of a set of neurons representing the input features and the rightmost part is a non-linear activation function [48].…”

“…Thanks to the NN method, about one order of magnitude is decrease compared to the BER with VNLE after 60 km transmission. An example for BER performance of NN is shown in Figure 20a [49]. Thanks to the NN method, about one order of magnitude is decrease compared to the BER with VNLE after 60 km transmission.…”

“…Some new equalization techniques, such as direct detection-faster than Nyquist (DD-FTN) algorithm [33][34][35], intensity directed FFE/DFE (ID-FFE/ID-DFE) algorithm [36,37] and joint clock recovery and FFE (CR-FFE) algorithm [38,39], have recently been proposed to solve different problems in practical implementation. In addition, machine learning methods like support vector machine (SVM) [40][41][42] and neural network (NN) [43][44][45][46][47][48][49][50][51][52][53] have been investigated to further eliminate system impairments for PAM4 modulation.In this paper, an overview is provided on the important topic of advanced DSP for short-reach PAM4 optical links. The rest of this paper is organized as follows.…”

“…(a) BER vs. fiber length with NN for 112 Gbps SSB-PAM4 transmission. (b) BER vs. OSNR after 80 km dispersion uncompensated SSMF (reprinted from[49] with open access from OSA).…”

Computational complexity comparison of feedforward/radial basis function/recurrent neural network-based equalizer for a 50-Gb/s PAM4 direct-detection optical link

“…Many nonlinear compensation methods have been proposed for short-reach DD systems, such as the well-known Volterra-series-based nonlinear equalizer [7,8], and various neural network (NN)-based nonlinear equalizers [9][10][11][12][13][14][15][16][17]. Traditional equalization methods such as feedforward equalization (FFE) can also be applied for short-reach DD systems, however, their performance is quite limited compared with the nonlinear methods such as NNs [16].…”