Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability
Low-complexity nonlinear equalization is critical for reliable high-speed short-reach optical interconnects. In this paper, we compare the complexity, efficiency and stability performance of pruned Volterra series-based equalization (VE) and neural network-based equalization (NNE) for 112 Gbps verti...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/17/4680 |
| _version_ | 1827708853293154304 |
|---|---|
| author | Wenjia Zhang Ling Ge Yanci Zhang Chenyu Liang Zuyuan He |
| author_facet | Wenjia Zhang Ling Ge Yanci Zhang Chenyu Liang Zuyuan He |
| author_sort | Wenjia Zhang |
| collection | DOAJ |
| description | Low-complexity nonlinear equalization is critical for reliable high-speed short-reach optical interconnects. In this paper, we compare the complexity, efficiency and stability performance of pruned Volterra series-based equalization (VE) and neural network-based equalization (NNE) for 112 Gbps vertical cavity surface emitting laser (VCSEL) enabled optical interconnects. The design space of nonlinear equalizers and their pruning algorithms are carefully investigated to reveal fundamental reasons of powerful nonlinear compensation capability and restriction factors of efficiency and stability. The experimental results show that NNE has more than one order of magnitude bit error rate (BER) advantage over VE at the same computation complexity and pruned NNE has around 50% lower computation complexity compared to VE at the same BER level. Moreover, VE shows serious performance instability due to its intricate structure when communication channel conditions become tough. Moreover, pruned VE presents more consistent equalization performance within varying bias values than NNE. |
| first_indexed | 2024-03-10T17:11:36Z |
| format | Article |
| id | doaj.art-edf5af515da049f387cb5440cfecdf01 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T17:11:36Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-edf5af515da049f387cb5440cfecdf012023-11-20T10:39:11ZengMDPI AGSensors1424-82202020-08-012017468010.3390/s20174680Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and StabilityWenjia Zhang0Ling Ge1Yanci Zhang2Chenyu Liang3Zuyuan He4State Key Laboratory of Advanced Optical Communication Systems and Networks, Shanghai Jiao Tong University, Shanghai 200240, ChinaState Key Laboratory of Advanced Optical Communication Systems and Networks, Shanghai Jiao Tong University, Shanghai 200240, ChinaState Key Laboratory of Advanced Optical Communication Systems and Networks, Shanghai Jiao Tong University, Shanghai 200240, ChinaState Key Laboratory of Advanced Optical Communication Systems and Networks, Shanghai Jiao Tong University, Shanghai 200240, ChinaState Key Laboratory of Advanced Optical Communication Systems and Networks, Shanghai Jiao Tong University, Shanghai 200240, ChinaLow-complexity nonlinear equalization is critical for reliable high-speed short-reach optical interconnects. In this paper, we compare the complexity, efficiency and stability performance of pruned Volterra series-based equalization (VE) and neural network-based equalization (NNE) for 112 Gbps vertical cavity surface emitting laser (VCSEL) enabled optical interconnects. The design space of nonlinear equalizers and their pruning algorithms are carefully investigated to reveal fundamental reasons of powerful nonlinear compensation capability and restriction factors of efficiency and stability. The experimental results show that NNE has more than one order of magnitude bit error rate (BER) advantage over VE at the same computation complexity and pruned NNE has around 50% lower computation complexity compared to VE at the same BER level. Moreover, VE shows serious performance instability due to its intricate structure when communication channel conditions become tough. Moreover, pruned VE presents more consistent equalization performance within varying bias values than NNE.https://www.mdpi.com/1424-8220/20/17/4680VCSELneural network-based equalizationVolterra series-based equalization |
| spellingShingle | Wenjia Zhang Ling Ge Yanci Zhang Chenyu Liang Zuyuan He Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability Sensors VCSEL neural network-based equalization Volterra series-based equalization |
| title | Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability |
| title_full | Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability |
| title_fullStr | Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability |
| title_full_unstemmed | Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability |
| title_short | Compressed Nonlinear Equalizers for 112-Gbps Optical Interconnects: Efficiency and Stability |
| title_sort | compressed nonlinear equalizers for 112 gbps optical interconnects efficiency and stability |
| topic | VCSEL neural network-based equalization Volterra series-based equalization |
| url | https://www.mdpi.com/1424-8220/20/17/4680 |
| work_keys_str_mv | AT wenjiazhang compressednonlinearequalizersfor112gbpsopticalinterconnectsefficiencyandstability AT lingge compressednonlinearequalizersfor112gbpsopticalinterconnectsefficiencyandstability AT yancizhang compressednonlinearequalizersfor112gbpsopticalinterconnectsefficiencyandstability AT chenyuliang compressednonlinearequalizersfor112gbpsopticalinterconnectsefficiencyandstability AT zuyuanhe compressednonlinearequalizersfor112gbpsopticalinterconnectsefficiencyandstability |