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

Sang, Bohan; Zhou, Wen; Tan, Yuxuan; Kong, Miao; Wang, Chen; Wang, Mingxu; Zhao, Li; Zhang, Jiao; Yu, Jianjun

doi:10.1109/jlt.2022.3146863

Cited by 42 publications

(14 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Schemes of (a) VNE and (b) NNE with single-symbol output and multisymbol output, and the corresponding window schemes of (c) VNE and (d) SSO-NNE and proposed MSO-NNE, and the architectures of (e) the LSTM layer and (f) the GRU layer in NNE115. Reprinted with permission from ref . Copyright 2022 Optica.…”

Section: Research Progress On Short Reach or Access Networkmentioning

confidence: 99%

See 1 more Smart Citation

High-Speed Optical Fiber Communication in China

2022

ACS Photonics

Self Cite

View full text Add to dashboard Cite

In the past decade, China has made great investments in studying photonics and photocommunication with larger communication capacity, better performance, and lower cost. After years of heavy investment, great progress has been made in academic research, manufacturing capabilities, and optical network scale. The number of papers written and published by researchers in China in the world’s top optical fiber communication conferences has increased year by year, accounting for about 10% of the total number of papers. These high-level articles show that scientists in China have made great strides in photonics research in devices, systems, and networks. On the other hand, photonics industrialization in China is getting more competitive with its large market scale and strong production capacity.

show abstract

Section: Research Progress On Short Reach or Access Networkmentioning

confidence: 99%

Section: Research Progress On Short Reach or Access Networkmentioning

confidence: 99%

High-Speed Optical Fiber Communication in China

2022

ACS Photonics

Self Cite

View full text Add to dashboard Cite

show abstract

“…To address this performance-complexity trade-off, two well-known approaches can generally be considered. First, we can modify the original NN equalizer architecture, which recovers just one symbol at a time from a multisymbol input, so that multiple symbols can be recovered at a time [11], [18]. This may be achieved by using multidimensional regression predictive modeling (or a multidimensional classification when a soft demapper is coupled to the NN equalizing structure [18]).…”

Section: Introductionmentioning

confidence: 99%

“…First, we can modify the original NN equalizer architecture, which recovers just one symbol at a time from a multisymbol input, so that multiple symbols can be recovered at a time [11], [18]. This may be achieved by using multidimensional regression predictive modeling (or a multidimensional classification when a soft demapper is coupled to the NN equalizing structure [18]). In the case when the resulting multi-output NN architecture is similar to the original one (that recovered just one symbol at a time), the overall complexity per recovered symbol is reduced.…”

Section: Introductionmentioning

confidence: 99%

Reducing Computational Complexity of Neural Networks in Optical Channel Equalization: From Concepts to Implementation

Freire

Napoli

Costa

et al. 2023

J. Lightwave Technol.

View full text Add to dashboard Cite

In this paper, a new methodology is proposed that allows for the low-complexity development of neural network (NN) based equalizers for the mitigation of impairments in highspeed coherent optical transmission systems. In this work, we provide a comprehensive description and comparison of various deep model compression approaches that have been applied to feed-forward and recurrent NN designs. Additionally, we evaluate the influence these strategies have on the performance of each NN equalizer. Quantization, weight clustering, pruning, and other cutting-edge strategies for model compression are taken into consideration. In this work, we propose and evaluate a Bayesian optimization-assisted compression, in which the hyperparameters of the compression are chosen to simultaneously reduce complexity and improve performance. Next, this paper presents four distinct metrics (RMpS, BoP, NABS, and NLGs) that are discussed here that can be used to evaluate the amount of computing complexity required by various compression algorithms. These measurements can serve as a benchmark for evaluating the relative effectiveness of various NN equalizers when compression approaches are used. In conclusion, the trade-off between the complexity of each compression approach and its performance is evaluated by utilizing both simulated and experimental data in order to complete the analysis. By utilizing optimal compression approaches, we show that it is possible to design an NN-based equalizer that is simpler to implement and has better performance than the conventional digital back-propagation (DBP) equalizer with only one step per span. This is accomplished by reducing the number of multipliers used in the NN equalizer after applying the weighted clustering and pruning algorithms. Furthermore, we demonstrate that an equalizer based on NN can also achieve superior performance while still maintaining the same degree of complexity as the full electronic chromatic dispersion compensation block. We conclude our analysis by highlighting open questions and existing challenges, as well as possible future research directions.

show abstract

“…However, there exist difficulties around the implementation of an RNN-based equaliser in hardware, specifically the timing requirements of the RNN feedback mechanism, as noted in [12], [19]. Parallel multisymbol output schemes for neural network-based equalisers have been investigated in [20], and also implemented on Field Programmable Gate Arrays (FPGA) in [12], [19], [21], with the authors of [19] realising NNE based on long short-term memory (LSTM) with competitive complexity for coherent transmission. Here we investigate such multi-symbol techniques and their impact on overall system DR and equaliser complexity in a PON context.…”

Section: Introductionmentioning

confidence: 99%

High Dynamic Range 100G PON Enabled by SOA Preamplifier and Recurrent Neural Networks

Murphy¹,

Jamali²,

Townsend³

et al. 2023

J. Lightwave Technol.

View full text Add to dashboard Cite

In recent years the PON research community has focused on future systems targeting 100 Gb/s/λ and beyond, with digital signal processing seen as a key enabling technology. Spectrally efficient 4-level pulse amplitude modulation (PAM4) is seen as a cost-effective solution that exploits the ready availability of cheaper, low-bandwidth devices, and Semiconductor Optical Amplifiers (SOA) are being investigated as receiver preamplifiers to compensate PAM4's high signal-to-noise ratio requirements and meet the demanding 29 dB PON loss budget. However, SOA gain saturation-induced patterning distortion is a concern in the context of PON burst-mode signalling, and the 19.5 dB loudsoft packet dynamic range expected by the most recent ITU-T 50G standards. In this paper we propose a recurrent neural network equalisation technique based on gated recurrent units (GRU-RNN) to not only mitigate SOA patterning affecting loud packet bursts, but to also exploit their remarkable effectiveness at compensating non-linear impairments to unlock the SOA gain saturated regime. Using such an equaliser we demonstrate > 28 dB system dynamic range in 100 Gb/s PAM4 system by using SOA gain compression in conjunction with GRU-RNN equalisation. We find that our proposed GRU-RNN has similar equalisation capabilities as non-linear Volterra, fully connected neural network, and long short-term memory based equalisers, but observe that feedback-based RNN equalisers are more suited to the varying levels of impairment inherent to PON burst-mode signalling due to their low input tap requirements. Recognising issues surrounding hardware implementation of RNNs, we investigate a multi-symbol equalisation scheme to lower the feedback latency requirements of our proposed GRU-RNN. Finally, we compare equaliser complexities and performances according to trainable parameters and real valued multiplication operations, finding that the proposed GRU-RNN equaliser is more efficient than those based on Volterra, fully connected neural networks or long short-term memory units proposed elsewhere.

show abstract

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

Cited by 42 publications

References 20 publications

High-Speed Optical Fiber Communication in China

High-Speed Optical Fiber Communication in China

Reducing Computational Complexity of Neural Networks in Optical Channel Equalization: From Concepts to Implementation

High Dynamic Range 100G PON Enabled by SOA Preamplifier and Recurrent Neural Networks

Contact Info

Product

Resources

About