Performance assessment of octagonal layout based integrated photonic reservoir computing for optical channel equalization

Compared to electronic equalizers, the equalizers based on integrated photonic reservoir computing exhibit the advantages of large bandwidth, short latency, and low power consumption. The existing rectangular reservoir layout employs internal nodes as input and output nodes, leading to inadequate signal exchange and extra noise due to waveguide crossing. We proposed an optical channel equalizer based on an octagonal reservoir layout and verified its equalization ability. The system-level simulation of the proposed layout is conducted to investigate the impact of layout shape, node quantity, input and output port number, and weight algorithm on channel equalization performance in a 25Gbaud 4-level pulse amplitude modulation optical communication system. Results indicate that, within a transmission distance of 5-30 km, the bit error rate (BER) of the equalized signal in the proposed octagonal layout is lower than that of a rectangular layout with the same number of nodes. Specifically, at 5 km, the octagonal layout's BER is 5 and 10 orders of magnitude lower than the rectangular layout and the feedforward equalizer, respectively. Meanwhile, the best readout algorithm, K-nearest neighbor regression, achieves an equalized BER of 7.012 × 10<sup>-14</sup>.

Authors

Institutions

Publication Details

Journal
Optics Express
Published
2025-03-11
DOI
https://doi.org/10.1364/oe.554334
Citations
1
Primary Topic
Neural Networks and Reservoir Computing
Type
article
Field-Weighted Citation Impact
2.48

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Performance assessment of octagonal layout based integrated photonic reservoir computing for optical channel equalization

Qi Qu, Bowen Bai, Zhouyi Hu, Jianshuai Wang et al.
1 citations
Optics Express
Neural Networks and Reservoir Computing
2.48
article

Performance assessment of octagonal layout based integrated photonic reservoir computing for optical channel equalization

Qi Qu, Bowen Bai, Zhouyi Hu, Jianshuai Wang, Bing Bai, Xiaoyan Zuo, Pei Li, Fei Dong
article en
1 citations

Abstract

Compared to electronic equalizers, the equalizers based on integrated photonic reservoir computing exhibit the advantages of large bandwidth, short latency, and low power consumption. The existing rectangular reservoir layout employs internal nodes as input and output nodes, leading to inadequate signal exchange and extra noise due to waveguide crossing. We proposed an optical channel equalizer based on an octagonal reservoir layout and verified its equalization ability. The system-level simulation of the proposed layout is conducted to investigate the impact of layout shape, node quantity, input and output port number, and weight algorithm on channel equalization performance in a 25Gbaud 4-level pulse amplitude modulation optical communication system. Results indicate that, within a transmission distance of 5-30 km, the bit error rate (BER) of the equalized signal in the proposed octagonal layout is lower than that of a rectangular layout with the same number of nodes. Specifically, at 5 km, the octagonal layout's BER is 5 and 10 orders of magnitude lower than the rectangular layout and the feedforward equalizer, respectively. Meanwhile, the best readout algorithm, K-nearest neighbor regression, achieves an equalized BER of 7.012 × 10<sup>-14</sup>.

Optics ExpressVol. 33(7)
Photonic Science (United Kingdom) (GB), Aerospace Institute (Germany) (DE)
National Natural Science Foundation of China, Beijing Jiaotong University
Affordable and clean energy
Openalex Percentile: Top 11%
Neural Networks and Reservoir Computing
2.48
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.