Convolutional Neural Network based Nonlinear Classifier for 112-Gbps High Speed Optical Link

Chuang, Chiao‐Lin; Liu, Lichun; Wei, Chia-Chien; Liu, Jun-Jie; Henrickson, Lindor; Huang, Wan-Jou; Wang, Chih-Lin; Chen, Young-Kai; Chen, Jyehong

doi:10.1364/ofc.2018.w2a.43

Cited by 48 publications

(31 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In references [77]- [82] that employ neural networks for equalization, usually a vector of sampled receive symbols act as the input to the neural networks with the output being equalized signal with reduced inter-symbol interference (ISI). In [77], [78], and [79] for example, a convolutional neural network (CNN) would be used to classify different classes of a PAM signal using the received signal as input. The number of outputs of the CNN will depend on whether it is a PAM-4, 8, or 16 signal.…”

Section: Nonlinearity Mitigationmentioning

confidence: 99%

An Overview on Application of Machine Learning Techniques in Optical Networks

Musumeci

Rottondi

Nag

et al. 2019

IEEE Commun. Surv. Tutorials

485

226

View full text Add to dashboard Cite

Today's telecommunication networks have become sources of enormous amounts of widely heterogeneous data. This information can be retrieved from network traffic traces, network alarms, signal quality indicators, users' behavioral data, etc. Advanced mathematical tools are required to extract meaningful information from these data and take decisions pertaining to the proper functioning of the networks from the network-generated data. Among these mathematical tools, Machine Learning (ML) is regarded as one of the most promising methodological approaches to perform network-data analysis and enable automated network self-configuration and fault management.The adoption of ML techniques in the field of optical communication networks is motivated by the unprecedented growth of network complexity faced by optical networks in the last few years. Such complexity increase is due to the introduction of a huge number of adjustable and interdependent system parameters (e.g., routing configurations, modulation format, symbol rate, coding schemes, etc.) that are enabled by the usage of coherent transmission/reception technologies, advanced digital signal processing and compensation of nonlinear effects in optical fiber propagation.In this paper we provide an overview of the application of ML to optical communications and networking. We classify and survey relevant literature dealing with the topic, and we also provide an introductory tutorial on ML for researchers and practitioners interested in this field. Although a good number of research papers have recently appeared, the application of ML to optical networks is still in its infancy: to stimulate further work in this area, we conclude the paper proposing new possible research directions.

show abstract

Section: Nonlinearity Mitigationmentioning

confidence: 99%

An Overview on Application of Machine Learning Techniques in Optical Networks

Musumeci

Rottondi

Nag

et al. 2019

IEEE Commun. Surv. Tutorials

485

226

View full text Add to dashboard Cite

show abstract

“…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].…”

Section: Neural Networkmentioning

confidence: 99%

“…Yang et al experimentally demonstrate a C-band 4 × 50 Gbit/s PAM4 transmission over 80 km SSMF employing a radial basis function ANN [44]. In [46], with the help of CNN, a 112 Gbit/s PAM4 transmission over 40 km SSMF is accomplished and the BER performance outperforms traditional VNLE. For more complicated DNN, the BER of 4.41 × 10 −5 is obtained for a 64 Gbit/s PAM4 transmission over 4 km MMF based on 850 nm VCSEL [47].…”

Section: Summary Of Recent Workmentioning

confidence: 99%

“…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.…”

mentioning

confidence: 99%

See 1 more Smart Citation

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

Zhou

Liu

et al. 2019

Applied Sciences

View full text Add to dashboard Cite

In recent years, short-reach optical links have attracted much more attention and have come to constitute a key market segment due to the rapid development of data-center applications and cloud services. Four-level pulse amplitude modulation (PAM4) is a promising modulation format to provide both a high data rate and relatively low cost for short-reach optical links. However, the direct detector and low-cost components also pose immense challenges, which are unforeseen in coherent transmission. To compensate for the impairments and to truly meet data rate requirements in a cost-effective manner, various digital signal processing (DSP) technologies have been proposed and investigated for short-reach PAM4 optical links. In this paper, an overview of the latest progress on DSP equalization technologies is provided for short-reach optical links based on PAM4 modulation. We not only introduce the configuration and challenges of the transmission system, but also cover the principles and performance of different equalizers and some improved methods. Moreover, machine learning algorithms are discussed as well to mitigate the nonlinear distortion for next-generation short-reach PAM4 links. Finally, a summary of various equalization technologies is illustrated and our perspective for the future trend is given. Appl. Sci. 2019, 9, 2342 2 of 22 multi-tone (DMT), and quadrature amplitude modulation (QAM) based on Kramers-Kronig (KK) receiver [18]. Considering the power consumption and implementation complexity, the most attractive format in short-reach optical links is PAM4, which is believed as a highly promising candidate for the next-generation passive optical network (NG-PON) and has been ratified by IEEE 802.3 bs for 400 Gbps Ethernet transmission [19][20][21][22][23][24][25].However, the low-pass filtering effects induced by the limited bandwidth of transmitter and receiver can cause severely inter-symbol interference (ISI). Low-cost devices such as lasers, modulators, photodiodes (PD) and trans-impedance amplifiers (TIA) also produce nonlinear distortions like level-dependent skew and level-dependent noise. Furthermore, the interaction between chromatic dispersion (CD) and direct detection will lead to a power-fading effect, where the detected signal may contain frequency notches after several kilometers transmission at a high symbol rate. Therefore, various equalization technologies based on digital signal processing (DSP) have been investigated for short-reach links over the years . Conventional equalizers, such as the feed-forward equalizer (FFE) or decision feedback equalizer (DFE), are employed to compensate for the linear impairments induced by limited bandwidth and CD [26][27][28], while the equalizer based on the Volterra series is utilized to mitigate nonlinear distortion [29][30][31][32]. 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...

show abstract

“…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].…”

Section: Introductionmentioning

confidence: 99%

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

Sun

et al. 2019

Opt. Express

View full text Add to dashboard Cite

The computational complexity and system bit-error-rate (BER) performance of four types of neural-network-based nonlinear equalizers are analyzed for a 50-Gb/s pulse amplitude modulation (PAM)-4 direct-detection (DD) optical link. The four types are feedforward neural networks (F-NN), radial basis function neural networks (RBF-NN), auto-regressive recurrent neural networks (AR-RNN) and layer-recurrent neural networks (L-RNN). Numerical results show that, for a fixed BER threshold, the AR-RNN-based equalizers have the lowest computational complexity. Amongst all the nonlinear NN-based equalizers with the same number of inputs and hidden neurons, F-NN-based equalizers have the lowest computational complexity while the AR-RNN-based equalizers exhibit the best BER performance. Compared with F-NN or RNN, RBF-NN tends to require more hidden neurons with the increase of the number of inputs, making it not suitable for long fiber transmission distance. We also demonstrate that only a few tens of multiplications per symbol are needed for NN-based equalizers to guarantee a good BER performance. This relatively low computational complexity signifies that various NN-based equalizers can be potentially implemented in real time. More broadly, this paper provides guidelines for selecting a suitable NN-based equalizer based on BER and computational complexity requirements.

show abstract

Convolutional Neural Network based Nonlinear Classifier for 112-Gbps High Speed Optical Link

Cited by 48 publications

References 1 publication

An Overview on Application of Machine Learning Techniques in Optical Networks

An Overview on Application of Machine Learning Techniques in Optical Networks

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

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

Contact Info

Product

Resources

About