SVM for Constellation Shaped 8QAM PON System

doi:10.12142/ZTECOM.2022S1009

ZTE Communications ›› 2022, Vol. 20 ›› Issue (S1): 64-71.DOI: 10.12142/ZTECOM.2022S1009

收稿日期:2021-07-25 出版日期:2022-01-25 发布日期:2022-03-01

SVM for Constellation Shaped 8QAM PON System

LI Zhongya^1,^3,, CHEN Rui^2,^3,, HUANG Xingang², ZHANG Junwen^1,^3,, NIU Wenqing^1,^3,, LU Qiuyi^1,^3,, CHI Nan¹()

^1.Key Laboratory for Information Science of Electromagnetic Waves, Fudan University, Shanghai 200433, China
^2.ZTE Corporation, Shenzhen 518057, China
^3.Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, Shanghai 200433, China
the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China

Received:2021-07-25 Online:2022-01-25 Published:2022-03-01
About author:LI Zhongya is with the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China.|CHEN Rui is with the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China.|HUANG Xingang is a senior expert of technical pre-research of ZTE Corporation. He received the M.S. degree in physics from Xi’an Jiaotong University, China in 2008. He is engaged in the research of optical access technology, especially in WDM-PON, TWDM-PON, NG-EPON, 50 G PON.|ZHANG Junwen is with the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China.|NIU Wenqing is with the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China.|LU Qiuyi is with the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China.|CHI Nan (nanchi@fudan.edu.cn) is with the Department of Communication Science and Engineering, School of Information Science and Technology, Fudan University, China.
Supported by:
the NSFC project(61925104);ZTE Industry?Academia?Research Cooperation Funds(HC?CN?20191231006)

摘要/Abstract

Abstract:

Nonlinearity impairments and distortions have been bothering the bandwidth constrained passive optical network (PON) system for a long time and limiting the development of capacity in the PON system. Unlike other works concentrating on the exploration of the complex equalization algorithm, we investigate the potential of constellation shaping joint support vector machine (SVM) classification scheme. At the transmitter side, the 8 quadrature amplitude modulation (8QAM) constellation is shaped into three designs to mitigate the influence of noise and distortions in the PON channel. On the receiver side, simple multi-class linear SVM classifiers are utilized to replace complex equalization methods. Simulation results show that with the bandwidth of 25 GHz and overall bitrate of 50 Gbit/s, at 10 dBm input optical power of a 20 km standard single mode fiber (SSMF), and under a hard-decision forward error correction (FEC) threshold, transmission can be realized by employing Circular (4, 4) shaped 8QAM joint SVM classifier at the maximal power budget of 37.5 dB.

Key words: passive optical networks, support vector machine, geometrically shaping, constellation classification, digital signal processing

. [J]. ZTE Communications, 2022, 20(S1): 64-71.

LI Zhongya, CHEN Rui, HUANG Xingang, ZHANG Junwen, NIU Wenqing, LU Qiuyi, CHI Nan. SVM for Constellation Shaped 8QAM PON System[J]. ZTE Communications, 2022, 20(S1): 64-71.

图/表 11

Figure 1 Constellations of shaped 8 quadrature amplitude modulation (8QAM): (a) Circular (4, 4), (b) rectangular and (c) triangular

Figure 2 (a) Flow chart of SVM classifier, (b) nonlinear distortions of the three constellations and (c) their influences on hyperplanes of the SVM classifier

Figure 3 (a) Simulation setup of constellation shaped 8QAM PON system, (b) spectrum of the transmitted signal and (c) electro-absorption modulator (EAM) transmission characteristic

Figure 4 BER of Circular (4, 4) versus (a) roll off of shaping filter, (b) wavelength of LD, (c) line width of LD and (d) bias of laser

Figure 5 BER performance versus amplitude of the Tx signal (a) Circular (4, 4), (b) Rectangular and (c) Triangular

Figure 6 Distortions under different levels of nonlinearity

Figure 7 The classification results of SVM and ED decision: (a) Circular (4, 4), (b) Rectangular and (c) Triangular

Figure 8 BER performance versus bandwidth limitation of fiber system: (a) Circular (4, 4), (b) Rectangular and (c) Triangular

Figure 9 BER performance versus received power: (a) Circular (4, 4), (b) Rectangular and (c) Triangular

Figure 10 BER performance versus fiber input power: (a) Circular (4, 4), (b) Rectangular and (c) Triangular

Table 2 Computational complexity

Methods	Training		Prediction
Methods	Additions	Multiplications	Additions	Multiplications
LMS	$(M 1 + ? M 1 + 1) T 1$	$(M 1 + ? M 1 + 1) T 1$	$M 1$	$M 1$
LMS-VOT	$(2 M 1 + M 2 M 2 + 1 2$ $+ M 2 2 + 2) T 1$	$(2 M 1 + M 2 M 2 + 1$ $+ M 2 2 + 2) T 1$	$M 1 + M 2 M 2 + 1 2 + 1$	$M 1 + M 2 M 2 + 1$
SVM^[19]	$O (N s 2 + N s d T 2)$		$O (N s d)$

Table 2 Computational complexity

Methods	Training		Prediction
Methods	Additions	Multiplications	Additions	Multiplications
LMS	$(M 1 + ? M 1 + 1) T 1$	$(M 1 + ? M 1 + 1) T 1$	$M 1$	$M 1$
LMS-VOT	$(2 M 1 + M 2 M 2 + 1 2$ $+ M 2 2 + 2) T 1$	$(2 M 1 + M 2 M 2 + 1$ $+ M 2 2 + 2) T 1$	$M 1 + M 2 M 2 + 1 2 + 1$	$M 1 + M 2 M 2 + 1$
SVM^[19]	$O (N s 2 + N s d T 2)$		$O (N s d)$

参考文献 19

1	IEEE. Physical layer specifications and management parameters for 25 Gbit/s and 50 Gbit/s passive optical networks: IEEE 802.3ca Task Force [S]. 2018
2	HOUTSMA V, DVAN VEEN. Optical strategies for economical next generation 50 and 100G PON [C]//Optical Fiber Communication Conference. OFC, 2019. DOI:10.1364/ofc.2019.m2b.1 DOI URL
3	JI H L, YI L L, LI Z X, et al. Field demonstration of a real-time 100-Gbit/s PON based on 10G-class optical devices [J]. Journal of lightwave technology, 2017, 35(10): 1914–1921. DOI: 10.1109/JLT.2016.2633482 DOI URL
4	HOUTSMA V, DVAN VEEN. Demonstration of symmetrical 25 Gbit/s TDM-PON with 31.5 dB optical power budget using only 10 Gbit/s optical components [C]//2015 European Conference on Optical Communication. IEEE, 2015: 1–3. DOI:10.1109/ECOC.2015.7341691 DOI URL
5	WEI J L, EISELT N, GRIESSER H, et al. Demonstration of the first real-time end-to-end 40-Gbit/s PAM-4 for next-generation access applications using 10- Gbit/s transmitter [J]. Journal of lightwave technology, 2016, 34(7): 1628–1635. DOI:10.1109/JLT.2016.2518748 DOI URL
6	IEEE. Ethernet amendment 10: media access control parameters, physical layers, and management parameters for 200 Gbit/s and 400 Gbit/s 802.3bs-2017: IEEE 802.3bs-2017 [S]. 2017
7	ZHANG J, YU J J, CHIEN H, et al. Demonstration of 100 Gbit/s/λ PAM-4 TDM-PON supporting 29 dB power budget with 50 km reach using 10 G class O-band DML transmitters [C]//Optical Fiber Communication Conference Postdeadline Papers. OSA, 2019. DOI:10.1364/ofc.2019.th4c.3 DOI URL
8	ZHANG J W, YU J J, LI F, et al. 11 × 5 × 93 Gbit/s WDM-CAP-PON based on optical single-side band multi-level multi-band carrierless amplitude and phase modulation with direct detection [J]. Optics express, 2013, 21(16): 18842. DOI:10.1364/oe.21.018842 DOI URL
9	WEI J L, INGHAM J D, CUNNINGHAM D G, et al. Performance and power dissipation comparisons between 28 Gbit/s NRZ, PAM, CAP and optical OFDM systems for data communication applications [J]. Journal of lightwave technology, 2012, 30(20): 3273–3280. DOI:10.1109/JLT.2012.2213797 DOI URL
10	YIN S, HOUTSMA V, DVAN VEEN, et al. Optical amplified 40-Gbit/s symmetrical TDM-PON using 10-Gbit/s optics and DSP [J]. Journal of lightwave technology, 2017, 35(4): 1067–1074. DOI: 10.1109/JLT.2016.2614767 DOI URL
11	WANG C, DU J B, CHEN G Y, et al. QAM classification methods by SVM machine learning for improved optical interconnection [J]. Optics communications, 2019, 444: 1–8. DOI: 10.1016/j.optcom.2019.03.058 DOI URL
12	NIU W Q, HA Y, CHI N. Support vector machine based machine learning method for GS 8QAM constellation classification in seamless integrated fiber and visible light communication system [J]. Science China information sciences, 2020, 63(10): 202306. DOI: 10.1007/s11432-019-2850-3 DOI URL
13	SAIN S R. The nature of statistical learning theory [J]. Technometrics, 1996, 38(4): 409. DOI: 10.1080/00401706.1996.10484565 DOI URL
14	HSU C W, LIN C J. A comparison of methods for multiclass support vector machines [J]. IEEE transactions on neural networks, 2002, 13(2): 415–425. DOI: 10.1109/72.991427 DOI URL
15	JOACHIMS T. Training linear SVMs in linear time [C]//The 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM, 2006: 217–226. DOI: 10.1145/1150402.1150429 DOI URL
16	KEERTHI S S, DECOSTE D. A modified finite Newton method for fast solution of large scale linear SVMs [J]. Journal of machine learning research, 2005, 6(12):341-361
17	CHEN K Z, CHEN L W, LIN C Y, et al. 224-Gbit/s transmission for next-generation WDM long-reach PON using CAP modulation [C]//2016 Optical Fiber Communications Conference and Exhibition (OFC). IEEE, 2016: 1–3
18	STOJANOVIC N, KARINOU F, QIANG Z, et al. Volterra and Wiener equalizers for short-reach 100 G PAM-4 applications [J]. Journal of lightwave technology, 2017, 35(21): 4583–4594. DOI: 10.1109/JLT.2017.2752363 DOI URL
19	BURGES C J C. A tutorial on support vector machines for pattern recognition [J]. Data mining and knowledge discovery, 1998, 2(2): 121–167. DOI: 10.1023/A: 1009715923555 DOI URL

SVM for Constellation Shaped 8QAM PON System

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 19

相关文章 0

编辑推荐

Metrics