Top Banner
What the future holds for few-mode fiber transmission? William Shieh Centre for Energy-Efficient Telecommunications National ICT Australia Department of Electrical and Electronic Engineering The University of Melbourne, Melbourne, Australia
26

Tsinghua visit

May 26, 2015

Download

Technology

wtyru1989
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Tsinghua visit

What the future holds for few-mode fiber transmission?

William Shieh

Centre for Energy-Efficient Telecommunications National ICT Australia

Department of Electrical and Electronic Engineering The University of Melbourne, Melbourne, Australia

Page 2: Tsinghua visit

澳大利亚墨尔本大学电机与电子工程系教授Prof. Shieh目前正在招收优秀博士生(本科或者研究生平均分达到80分及以上, 以及排名30% 以上),可以选择的

研究方向包括 (1)正交频分复用技术(OFDM)在光或者无线网络中的应用 (2)相干光通信,光信号处理以及信道均衡 (3)波导设计和非线性的特性描述 (4)全光数据包交换,波长转换以及新型光网络结构 (5)射频(RF)光子技术,包括RF信号的产生、特性描述、传输以及处理 William Shieh教授是澳洲杰出青年(Australian Future fellow)获得者以及美国光学协会院士(Fellow of Optical Society of America)。他的个人主页是 http://people.eng.unimelb.edu.au/shiehw/ 如果您对以上研究方向感兴趣,请与2012年1月15日之前递交申请。欢迎您访问如下网页了解申请信息以及步骤 http://www.ee.unimelb.edu.au/future_students/PhD_in_Engineering.html 或者直接通过电子邮件跟Shieh教授联系,他的邮箱地址是 [email protected]

PhD Program at Melbourne Uni

Page 3: Tsinghua visit

Centre for Energy Efficient Telecommunications (CEET)

Page 4: Tsinghua visit

• Capacity limit in the current SMF fiber

Analytical expression of fiber capacity

• Two-mode fiber (TMF) based transmission

LP10 / LP11 transmission

Two degenerate LP11 modes transmission

• Challenges in FMF fiber based systems

• Conclusion

Outline

Page 5: Tsinghua visit

Motivation Capacity crunch

H. Kogelnik

•A. R. Chraplyvy, “The coming capacity crunch,” ECOC’09 Plenary Talk, 2009 •R. W. Tkach, “Scaling optical communications for the next decade and beyond,” Bell Labs Tech. J. vol. 14, pp. 3-10, 2010 •M. Nakazawa, “Hardware paradigm shifts in the optical communication infrastructure with three “M technologies” OECC’2010

Page 6: Tsinghua visit

Degrees of Freedom for Multiplexing and Modulation

(3) Complex constellation or I/Q modulation

(1) Time

(2) Frequency

(4) Polarization

(5) Space

Page 7: Tsinghua visit

Nonlinearity Noise Density in SMF Fibers

f

2

NLc

II II

= ( )

20 2

0

81 ,3 ln / 2c

s e

I BN h B B B

πα β αγ π β

≡ =

2

2( 1 ) 1( 1)

sLN L Ls s

e Ls

N e N e ehN e

αζ αζ αζ

αζ

− − −

− + −≡ +

•X. Chen, and W. Shieh, Opt. Express 18, 19039-19054 (2010). •W. Shieh and X. Chen, IEEE Photon. Journal, vol. 3, 158 – 173, (2011).

B

The maximum capacity can be achieved by filling the spectrum with signals

Assume that the input signal density is I, what is the INL?

Page 8: Tsinghua visit

Information Spectral Efficiency in Presence of Fiber Nonlinearity

Shannon Capacity Theory: Spectral Efficiency Parameter Dependence: To increase 1 bit/s/Hz: (i) Reach is to be reduced to half (ii) Nonlinearity coefficient is to be reduced by 2.8 (iii) Chromatic dispersion is to be increased by 8

•X. Chen, and W. Shieh, Opt. Express 18, 19039-19054 (2010). •W. Shieh and X. Chen, IEEE Photon. Journal, vol. 3, 158 – 173, (2011).

( )( )2 2 2

0

log 1 log 1/ c

IS SNRn I I I

= + ≅ + +

( )2/32 0

1log 1 /3 cS I n = +

( ) ( )( ) 1/31/3 2 22 2 0 0log 8 3 ln /

3s

eN N h B Bπα β γ

− ≅

Page 9: Tsinghua visit

Why Few-mode Fiber or Two-mode Fiber?

In theory, for N-mode fibers, if we have N transmitters and N receivers, we only need compute NxN H matrix and perform the matrix inversion H-1. Complexity of H-1

scales faster than N2

It is sensible to start with the two mode fiber (TMF).

TMF fiber contains three spatial modes including LP01 mode and two degenerate LP11 modes. B.Y. Kim, et al., Opt. Lett., 11, 389-391 (1986). B.Y. Kim, et al., Opt. Lett., 12, 729 (1987)

Complex optical design & electronic DSP design.

Nevertheless, this still leads to 3 times bandwidth of a SMF fiber.

Tx1

Tx2

TxN

Rx1

Rx2

RxN

[Hij]

Page 10: Tsinghua visit

Some Scenarios of Mode Multiplexed Systems

(1) Short-reach or interconnect

High differential-mode-delay (DMD) low mode crosstalk, optical demulitpelxing F. Yaman, et al., Opt. Express, 18, 21342 (2010).

(2) Long-reach

Low DGD, high mode crosstalk, electronic demulitpelxing

(3) Higher-order constellation such as 16-QAM and beyond

Electronic demulitpelxing

N. Hanzawa ‘Demonstration of mode-division multiplexing transmission over 10 km two-mode fiber with mode coupler’, OFC’2011, Paper TWA4.

Page 11: Tsinghua visit

For Corning's SMF-28e® SMF, V = 2.0396

For Corning's Infinicor® MMF, V =~ 20.8453.

TMF Fiber Parameters

( ) ( )V= 2 a/ NA= 2 5.935 /1.55 0.1505 3.6218π λ π⋅ × × =

_ - _ 0.0054_

n core n cladnn core

∆ = =

2 2 _ - _ 0.1505NA n core n clad= =

_ 1.4518n core =_ 1.4440n clad =

Refractive Index for core and cladding

Refractive Index Difference

Numerical Aperture

Page 12: Tsinghua visit

Two-mode or Three-mode Fiber?

1.420

1.430

1.440

1.450

1.460

1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 Wavelength (μm)

Mod

al In

dex

n ef

f core LP01 LP11 cladding

LP11 LP01

cutoff 2323nm

2 LP modes: LP01 and LP11 3 Spatial modes: LP01 and 2 degenerate LP11: and 6 Fiber modes: 2 (polarization) x 3 (spatial )

a11LP b

11LP

Page 13: Tsinghua visit

Two-mode LP10 / LP11 Transmission Systems

Tx 1

Tx 2

Tx 3

Tx 4LP01-LP11 Mode

MultiplexingLP01-LP11 Mode Demultiplexing

LP01-LP11 Dual-mode amplifiers

TMF

Polarization Multiplexing

Polarization DemultiplexingControlled coupling to LP01 mode

Rx1

Rx2

Rx3

Rx4

SMF SMF

Page 14: Tsinghua visit

Mode converter 2 nominal 100%

conversion ratio. LP 01 + LP 11

LP 11 + LP 01 TMF SMF

Deformation

LP 01

Mode stripper

R. C. Youngquist, J. L. Brooks, and H. J. Shaw, "Two-mode fiber modal coupler," Opt. Lett. 9, 177-179 (1984)

Mode Converter

Grating period = Beating length 01 11 2 /( - )

500BL

mπ β βµ

==

LP 01

TMF SMF

Mode stripper LP 01 + LP 11

Deformation Mode converter 1

nominal 50% conversion ratio

Page 15: Tsinghua visit

107 Gb/s dual-mode dual polarization transmission over 4.5-km TMF fiber. ‘X’ indicates controlled coupling between LP01 modes of SMF and TMF by

center splicing.

PolarizationDiversity

90° hybrid

PolarizationDiversity

90° hybrid

LOLO

EDFA Offline2x2 MIMO

OFDMdetection

Offline2x2 MIMO

OFDMdetection

LOLO

EDFA Offline2x2 MIMO

OFDMdetection

Offline2x2 MIMO

OFDMdetection

4.5 km TMF

Center Splicing

EDFA

One-symbol delay

50:50 PBC

50 % MC

100 % MC

MC1MS1

MC2MS2

MS3

LP11 Rx

λBand:Band: 1 2 3

Optical OFDM Tx

PBC: Polarization-beam-combinerMS: Mode stripper MC: Mode converterLO: Local oscillator TMF: Two-mode fiberTx/Rx: Transmitter/Receiver

LP01 Rx

PD ADCPD ADCPD ADC

PD ADC

PD ADCPD ADCPD ADC

PD ADCPolarization

Diversity90° hybrid

PolarizationDiversity

90° hybrid

CO-OFDM Experiment Setup

Page 16: Tsinghua visit

Y. Ma, Y. Tang, and W. Shieh, "107 Gbit/s transmission over multimode fibre …" Electron. Lett. 45, 848-849 (2009).

Transmission Parameters Parameters Value for OFDM Transmission Unit

Polarization 2

Band 3

Mode 2 (LP01+LP11)

Bit Rate (raw) 25 (per pol/band/mode) * 6 Gbit/s

Bit Rate (net) 107 Gbit/s

Symbol Period 7.2 ns

Bandwidth 6.5625 (per band) GHz

No. of Subcarriers 64

Total No. of Symbols 500

No. of Training Symbols 20

Cyclic Prefix (CP) 1/8 of observation window

Modulation Format QPSK

Fiber Length 4.5 km

Launch Power 5.5 dBm

Receive Power -0.5(LP01) / -5.3(LP11) dBm

Page 17: Tsinghua visit

Experiment Results

1549.21nm

0.1nm/div

19.69GHz 10dB

1549.21nm

0.1nm/div

19.69GHz

10dB

Optical Spectra LP01 LP11

Constellation

Band1 Band2 Band3 Band1 Band2 Band3

LP01 LP11

Page 18: Tsinghua visit

LP01 Band1 Band2 Band3 Avg. pol-x 19.5 18.4 18.1 18.7 pol-y 18.5 18.3 17.9 18.3 Avg. 19.0 18.4 18.0 18.5 LP11 Band1 Band2 Band3 Avg. pol-x 15.2 18.6 16.2 16.9 pol-y 14.7 17.0 16.5 16.2 Avg. 15.0 17.8 16.4 16.5

No error was measured out of 100,590 bits for each band, polarization and mode measured.

Q Factors for 12 Tributaries

Page 19: Tsinghua visit

Two Degenerate LP11 Modes Transmission

BSBS

CL

CL CL

SMF TMF

Mode stripper coiling

Mode converterV-groove

Core-centeraligned splice

SMF TMF

0.9 mm jacket

IR beam profiler

TMF

Rotating FC connector

(a)

LP11a

LP11b

BS

TMF V-groove

TMF

(b)

CL

Page 20: Tsinghua visit

‘Dream’ System of Multimode Fiber Link

Narrow Linewidth (<10 ~ 100 KHz) laser Array

MMF with low loss: ~ 0.2 dB/km

MMF Amplifier MMF MUX

MMF OADM

MMF DeMUX

Page 21: Tsinghua visit

Review of Progress of Few-mode Transmission

•J. Sakai, et. al., Trans. Micro. Theory & Techn. 26, 658-665 (1978). •K. Kitayama, et. Al., IEEE J. Quantum Electron. (Lett.), vol.QE-15, pp. 6-8, 1979.

Low-speed Short-reach Systems

High-speed Long-reach systems using Conventional MMF •Z. Tong, et. al., OECC’2008, paper PDP5. •Z. Tong, et. al., Electronics Letters, vol. 44, pp. 1373-1375, 2008.

High-speed Long-reach systems using FMF fiber OFC’2011, Postdeadline papers

•A. Li et al, Proc. OFC, 2011, p.PDPB8. •M. Salsi et al., Proc. OFC, 2011, p.PDPB9. •R. Ryf et al, Proc. OFC, 2011, p.PDPB10

ECOC’2011, 8 more postdeadline papers on few-mode/core fibers

Page 22: Tsinghua visit

Exponential Internet Traffic Growth

R. W. Tkach, Bell Labs Tech. J., vol. 14, 2010

Bandwidth needs to scale up ~ 30 dB for the next two decades!!

Page 23: Tsinghua visit

Implications of 30 dB of More Bandwidth

Power

Space

System Complexity Implications

Transponders Fiber cables ROADMs Optical Amplifiers

Page 24: Tsinghua visit

Trade off between Spectral and Energy Efficiency

Spec

tral

Effi

cien

cy [b

/s/H

z]

101

1000 5 10 15 20 25

Required SNR per bit (dB)

Shannon

4

16

64

256512 QAM

2010(PDM)

30 35

512QAM

OFDM/64QAM16QAM

36QAM

QPSK

64QAM

Spec

tral

Effi

cien

cy [b

/s/H

z]

101

1000 5 10 15 20 25

Required SNR per bit (dB)

Shannon

4

16

64

256512 QAM

2010(PDM)

30 35

512QAM

OFDM/64QAM16QAM

36QAM

QPSK

64QAM

(i) Strive for high spectral efficiency with low energy efficiency (ii) Strive for high energy efficiency with low spectral efficiency But few-mode transmission can achieve both high spectral and

energy efficiency

Page 25: Tsinghua visit

Spatial Mode Multiplexing: a Promise or a Curse?

•Will the system be too complex and never be practical?

•Need to be mindful of DSP complexity for MIMO processing

• Electronics is getting better, but not a panacea; Could be energy-hog.

Page 26: Tsinghua visit

• Spatial mode division multiplexing (SMDM) has recently been demonstrated to be an additional degree of freedom for achieving ultrahigh capacity beyond that of SSMF fiber.

• Should be always mindful of complexity involved when proposing devices and subsystems for SMDM based systems.

• But FMF or SMDM is a fertile ground for innovation.

Conclusion