scispace - formally typeset
Journal ArticleDOI

10 $\times$ 224-Gb/s POLMUX-16QAM Transmission Over 656 km of Large- ${\rm A}_{\rm eff}$ PSCF With a Spectral Efficiency of 5.6 b/s/Hz

Reads0
Chats0
TLDR
In this article, the authors demonstrate the successful transmission of 10 channels with 224-Gb/s POLMUX-16QAM modulation (28 GBaud) on a 37.5 GHz wavelength grid.
Abstract
The authors demonstrate the successful transmission of 10 channels with 224-Gb/s POLMUX-16QAM modulation (28 GBaud) on a 37.5-GHz wavelength grid. Using large-Aeff pure-silica-core fibers they show a 656-km transmission distance with a spectral efficiency of 5.6 b/s/Hz. They report a back-to-back performance penalty of 3.5 dB compared to theoretical limits at the forward-error correction (FEC) limit (bit-error rate of 3.8·10-3), and a margin of 0.5 dB in Q-factor with respect to the FEC limit after 656 km of transmission.

read more

Content maybe subject to copyright    Report

10x224-Gb/s POLMUX-16QAM transmission over 656 km of
large-Aeff PSCF with a special efficiency of 5.6 b/s/Hz
Citation for published version (APA):
Sleiffer, V. A. J. M., Alfiad, M. S., Borne, van den, D., Kuschnerov, M., Veljanovski, V., Hirano, M., Yamamoto,
Y., Sasaki, T., Jansen, S. L., Wuth, T., & Waardt, de, H. (2011). 10x224-Gb/s POLMUX-16QAM transmission
over 656 km of large-Aeff PSCF with a special efficiency of 5.6 b/s/Hz.
IEEE Photonics Technology Letters
,
23
(20), 1427-1429. https://doi.org/10.1109/LPT.2011.2162722
DOI:
10.1109/LPT.2011.2162722
Document status and date:
Published: 01/01/2011
Document Version:
Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)
Please check the document version of this publication:
• A submitted manuscript is the version of the article upon submission and before peer-review. There can be
important differences between the submitted version and the official published version of record. People
interested in the research are advised to contact the author for the final version of the publication, or visit the
DOI to the publisher's website.
• The final author version and the galley proof are versions of the publication after peer review.
• The final published version features the final layout of the paper including the volume, issue and page
numbers.
Link to publication
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal.
If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please
follow below link for the End User Agreement:
www.tue.nl/taverne
Take down policy
If you believe that this document breaches copyright please contact us at:
openaccess@tue.nl
providing details and we will investigate your claim.
Download date: 26. Aug. 2022

IEEE PHOTONICS TECHNOLOGY LETTERS, VOL. 23, NO. 20, OCTOBER 15, 2011 1427
10 224-Gb/s POLMUX-16QAM Transmission
Over 656 km of Large-
PSCF With a Spectral
Efciency of 5.6 b/s/Hz
V. A. J. M. Sleiffer, Student Member, IEEE,M.S.Alad, Student Member, IEEE, D. van den Borne, Member, IEEE,
M. Kuschnerov, Member, IEEE, V. Veljanovski, M. Hirano, Y. Yamamoto, T. Sasaki, S. L. Jansen, Member, IEEE,
T. Wuth, Member, IEEE, and H. de Waardt, Member, IEEE
Abstract—The authors demonstrate the successful transmission
of 10 channels w ith 224-Gb/s POLMUX-16QAM modulation
(28 GBaud) on a 37.5-GHz wavelength grid. Using large-
pure-silica-core bers they show a 656-km transmission distance
with a spectral efciency of 5.6 b/s/Hz. They report a back-to-back
performance penalty of 3.5 d B compared to theoretical limits
at the forward-error correction (FEC) limit (bit-error rate of
3.8
10 ), and a margin of 0.5 dB in Q-factor with respect to the
FEC limit after 656 km of transmission.
Index Terms—Coherent detection, polarization multip lexing
(POLMUX), quadrature amplitude modulation (QAM).
I. INTRODUCTIO N
T
HE rapid increase in popularity of bandwidth-consumin g
applications such as internet video, cloud storage and so -
cial networking requires large volumes of data to be transmitted
over long distances. This has fueled the historically exponential
growth of data trafc volumes in the worldwide telecommunica-
tion network, and based on current trends it is likely to continue
to drive an unsurpassed need for transmission capacity over
the next decade. The industry-wide drive to develop the optical
components, subsystems an d systems required to upgrade long-
haul and ultra long-haul networks to 100 G line rates is a key
development to facilitate the need for more capacity in core net-
works. Pola rization-multip lexed quadrature ph a se shif t keying
(POLMUX-QPSK) [1] has emerged in recent years as t he most
suitable modulatio n format for 100 G line rates, as it is compat-
ible with th e standardized 50-GHz channel spacing. This allows
for a spectral efciency of 2.0 b/s/Hz, scaling C-band transmis-
sion systems to a total capacity of betw een 8 and 10 Tb/s.
Manuscript received April 04, 2011; revised July 06, 2011; accepted July 16,
2011. Date of publication July 22, 2011; date of current version September 21,
2011.
V.A.J.M.Sleiffer,M.S.Alad, and H. de Waardt are with the COBRA In-
stitute, Eindhoven University of Technology, 5612A Z Eind ho v en, The Nether-
lands (e-mail: v.a.j.m.sleiffer@tue.nl; m.s.alad@ieee.org; h.d.waardt@tue.nl).
D. van den Borne, M. Kuschnerov, V. Veljanovski, S. L. Jansen,
and T. Wuth are with Nokia Siemens Networks, 81541 Munich, Ger-
many (e-mail: dirk.vandenborne@nsn.com; maxim.kuschnerov@nsn.com;
vladimir.veljanovski@nsn.com; sander.jansen@nsn.com; torsten.wuth@nsn.
com).
M. Hirano, Y. Yam amoto, and T. Sasaki are with S umitom o Electric
Industries, Ltd., Yokohama 244-8588, Japan (e-mail: masahirano@sei.co.jp;
yamamoto-yoshinori@sei.co.jp; sasaki-takashi@sei.co.jp).
Color versions of one or m ore of the gures in this letter ar e available online
at http://ieeexplore.i eee.org.
Dig
ital Object Identier 10.1109/LPT.2011.2162722
With 100 G transport solution now well established and
rst commercial deployments o ngo ing , there is strong need to
understand ho w long-haul transmission systems can evolve to
400 G line rates. Increasing the spectral e fciency by using
higher-order modulation formats is the most effective metho d
to realize higher transmission capacities at a com parable
or lower cost per transmitt ed bit. Transponder techn olo gies
developed for 100 G line rates such as the optical compo-
nents for dual-polarization QPSK modu lation can be largely
reused when scaling to d enser modulation formats. Based
on such technolog ies several transmission experiments h ave
been reported using 16-level quadrature amp litude modulation
(POLMUX-16QAM) [2], [3], or even denser constellations
such as POLMUX-32QAM [4], POLMUX-64QAM [5], [6]
or POLMU X -1 28Q AM [7]. However, this research show ed
as well that the increase in optical signal-to-noise (OSNR)
threshold coupled with a decreased nonlinear threshold makes
it tremendously challenging to transport such formats over
long-haul distances.
The strongly reduced transmission ma rgin for dense constel-
lations limit s the spectral efciency that can be used for long-
haul transmission [8], m akin g it very hard to transport 400 G
within a 50-GHz W DM grid [4]. This underlin es the need for
an optim ized infrastructur e to improv e the feasible transmis-
sion distance of next-generation transport systems while main-
taining the maxim um possible spectral efciency. One solu-
tion is the use of exi-grid technology, which allows a ex-
ible scaling o f channel spacing across the transmission band [9],
[10]. This permits 40 0 G line rates to use a spectral band broader
than 50 GHz, and enables a trade-off between transmission dis-
tance and spectral efciency depending on network topology
and physical layer i nfrastructure. D ual-carrier modulation, with
both carriers being 200 G POLMUX-16QAM m odulated, is
a particularly pr omising approach to realize 400 G line rates,
as this all ows for the use of components with a lower optical
and/or electrical bandwidth. For this purpose we investigated
in this letter the transmission of 224 -Gb/s POLMU X-16QAM
on a 37.5-GHz grid over 656 km of large-
pure-silica-core
ber [11]. This allows for a 5.6 b/s/Hz spectral efciency, and
demonstrates the feasib ili ty of next-generation 400 G transport
using exi-r a te technology on an optimized ber infrastructure.
II. E
XPERIMENTAL SETUP
Fig.
1 depicts the setup used in the t ransmission experiment.
At t
he transmitter (Fig. 1 (a)), two combs of 5 lasers w ith a
1041-1135/$26.00 © 2011 IEEE

1428 IEEE PHOTONICS TECHNO LOGY LETTERS, V OL. 23, NO. 20, OCTOBER 15, 20 11
Fig. 1. Experim e ntal setup. (a) Transmitter ( Tx ). (b) Generation of electrical 4-PAM dri vin g signal. (c) Spe ctrum and recovered back-to-back constellations for
both polarizations. (d) Recirculating loop setup.
wavelength spacing of 75 GHz are coupled together using a
passive coupler. The two combs consist of a mix of ECL an d
DFB lasers, and the relative wavelength spacing between both
combs is 37.5 GHz. The output signals of both combs are subse-
quently 112-Gb/s 16-QAM modulated with two I Q-m od ulators,
with a
of 2.2 V. Two outputs of a pulse pattern generator
(PPG), generating
PRBS sequences at a 28-Gbaud symbol
rate, are rst cleaned up using t wo D ip-ops (DFF). Each of
the DFF outputs is subsequently split, amplied and delayed for
de-correlation (Fig. 1(b)). One of the paths is 6 d B attenuated
and afterwards the two paths are com bined to create an elec-
trical 4-PAM signal [3]. Two 4-PAM signals are used to drive
the In-ph ase and Quadrature inputs of the IQ-m odu lators. An
additional attenuation of 6 dB has been used to reduce the im-
pact of the reections from the IQ-modulator and ampliers,
resulting in 4-PAM signal with a voltage swing of
[2]. The 112-Gb/s 16-QAM m odu lated signals are rst ampli-
ed separately and subsequently multiplexed using a exi-grid
wavelength selective switch (WSS, Waveshaper 4000S) with
37.5-GHz channel spacing. The resulting optical spectrum is
shown in Fig. 1(c). The combined signal is subsequently fed
into a polarization-m ultiplexing (POLM UX) stage. In this stage
the incoming optical signal is split up i n two equally power ed
tributaries, one delayed for de-correlation, and subsequently re-
combined using a polarization beam combiner (PBC), resulting
in 2 24-Gb /s POLMUX-1 6QAM m odulation.
The transmission link consists of a recirculating loop
(Fig. 1(d)) with four spans of 82-km large-
pure-silica-core
ber. The average loss of the ber is 0.161 dB/km, the chro-
matic dispersion coefcient is 21.0 ps/nm/km a t a wavelength
of 1550 nm, and the d ispersion slope equals 0.061
.
Theaverageeffectiveareaofthelarge-
ber is 133 ,
which results in a nonlinear coefcient of 0.6
.The
average total loss of each of the 82-km spans is 14 dB, which
is a combination of the ber loss and the splice losses between
standard single-mode ber and the large-
ber. The span
loss is compensated for by hybrid EDFA /R aman amplication
using backward-pumping with an average ON/OFF Raman
gain of 6.5 dB at the maximum pump power. The Raman gain
is reduced due to the core size of the large-
ber, but the
low ber loss at the Raman pump wavelengths (0.20 dB/km at
1450 nm) ensures that a sufciently high Raman gain is still
obtained at moderate pump level (6 50 mW ).
After transmission, the chan nels of interest are ltered out
using a de-multiplexing lter with a 100-GHz bandwidth and
subsequently fed to a polarization-diversity 90
optical h ybrid.
The s ig nal is mixed with a local oscillato r (LO) with 100-kHz
linewidth, which selects the channel under test by coherent
channel selection (tuned within
60 MHz of the channel
wavelength). The four outputs of the 90
optical hybrid are
converted to the electrical domain using balanced photodiodes,
and subsequently dig iti zed using a 50-Gsamples/s real- time
digital sampling scope (D SA72004B). A sequence of one mil-
lion samples, corresponding to 4.48 million bits is subsequently
used for ofine digital signal processing. Per measurement
point, the average BER of two 224-Gb/s POLMUX-16QAM
channels, selected by the de-multiplexing lter, are presented
in the results section. The average BER is calculated from
eight shots, fo ur shots of both channels separately, at different
time instances. Ofine digital signal processing is used to
demodulate the resulting samples as described in [2]. During
the measurements no cycle slips were observed, and therefore
no differential decoding was used.
III. R
ESULTS
The back-to-back measured BER of the 224-Gb/s POLMUX-
16QAM modulated sig nal, single channel and mu ltiplex ed on
a 37.5-GHz grid, as a functio n of the OSNR is shown in Fig.
2. At a BER of
, the requ ired OSNR for single channel
224-Gb/s POLMUX-16QAM modulation is 2 5.4 dB . A 5.1-dB
penalty with respect to the theoretical O SN R requirement of a
224-Gb/s POLMUX-16QAM modulated signal is observed. At
the FEC-thresh o ld at a BER of
thedifferencebe-
tween the measured and theoretical required OSNR is reduced
to 3.5 dB (19.0 dB versus 22.5 d B). In the WDM co nguration
the required OSNR for a BER of
is equal to 27.1 dB. The
1.7-dB OSNR penalty between the single channel and WDM
conguration is resulting from the linear crosstalk between the
WDM signals, due to the tig ht channel spacing [3].
Fig. 3 shows the measured average BER as a function of the
launch power for the center two 224-Gb/s POLMUX-16QAM
channels (channels 5 and 6) after transmission over 656 km of

SLEIFFER et al .:10 224-Gb/s POLMUX-16QAM TRANSMISSIO N OVE R 656 km OF LAR GE- PSCF 1429
Fig. 2. Measured back-to-back BER curves for 224-Gb/s POLMU X-16QAM.
Fig. 3. Launch power variation versus average BER for the center two
224-Gb/s POLMUX-16QAM channels after 656 km of transmission.
Fig. 4. Average BER results for the center two 224-Gb/s POLMUX-16QAM
channels versus transmission distance.
large- pure-silica-core ber. The o ptimal launch power is
1 dBm per 224-Gb/s POLMUX-16QAM channel, which con-
rms the signicant benet of large core ber to improve the
nonlinear tolerance of high spectrally efcient modulation f or-
mats such as POLMUX-16Q AM modulatio n. Fig. 4 shows the
measured BER as a function of transmission distance for the
center two 224-Gb/s POLMUX-1 6QA M modulated ch annels,
and at the launch power of
1 dBm per channel. After 656 km
of large-
pure-silica-core ber (2 recirculating loops) an av-
erage BER of
is measured. This translates into a
margin of approximately 0.6 dB i n Q-factor with respect to an
FEC-limit of
. A fter 3 recirculating loops (98 4 km)
the measured BER has degraded to
.
Fig. 5 shows the measured BER for each of the ten 224-Gb/s
WDM channels after 656 km transm ission and at the optimum
launch power. For each of the measurements, the ECL-lasers
are tuned from t he center channel to the channel un der test t o
ensure a low enough laser linewidth for carrier phase estima-
tion. The transmission results depicted in Fig. 5 conrm that all
the 224-Gb/s channels have nearly the same performance, and
Fig. 5. BER results for the ten 224-Gb/s POLMUX-16QAM c hannels after
656 km of tra nsmission.
the worst measured channel has a margin of 0.5 dB in Q-factor
(BER of
) with respect to the FEC-limit.
IV. C
ONCLUSION
We have shown tran smission of ten 224-Gb/s POLMUX-
16QAM modulated channels on a 37.5-GHz exible grid, re-
sultinginaspectralefciency of 5.6 b /s/Hz. A total capacity
of 2 Tb/s has been transmitted over 656 k m of large-
pure-
silica-core ber with a margin of 0.5 dB in Q-factor w ith respect
to the FEC-limit for the worst channel.
R
EFERENCES
[1] M.Birk,P.Gerard,R.Curto,andL.Nelsonet al., “Field trial of a real-
time, single wavelength, coherent 100 Gbit/s PM-QPSK channel up-
grade of an installed 1800 km link,” presented a t the OFC, San Diego,
CA, 2010, Paper PDPD1.
[2] M.S.Alad, M. Kuschnerov, S. L. Jansen, T. Wuth, D. van den Bornea,
andH.deWaardt,“Transmissionof11
224-Gb/s POLMUX-RZ-
16QAM over 1500 km of LongLine and pure-silica SMF,” presented
at the ECOC, Torino, Italy, 2010, Paper WE.8.C.2.
[3] A. H. Gnauck, P. J. Winzer, S. C handrasekhar, X. Liua, B. Zhu, and D.
W. Peckham, “Spectrally efcient long-haul WDM transmission using
224-GB/s polarization-multip lex ed 16-QA M, J. Lightw. Technol.,
vol. 29, no. 4, pp. 373–377, 2011.
[4] X . Zhou, L. Nelson, P. Magill, R. Isa ac, B. Zhu, D. W. Peckham, P.
Borel, and K. Ca rlso n, “8
450-Gb/s, 50-GHz-spaced, PDM-32QAM
transmission over 400 km an d one 50 GHz-g rid ROADM,” p resented
at the OFC, Los Angeles, CA, 2011, Paper PDPB3.
[5] A. Sano, T. Kobayashi, A . Matsuura, and S. Yamamoto et al., “100
120-Gb/s PDM 64-QAM transmission over 160 km using linewidth-
tolerant pi lotless digita l coherent detection,” presented at the E COC,
Torino, Italy, 2010, Paper PD2.4.
[6] A.H.Gnauck,P.J.Winzer,A.Konczykowska,andF.Jorgeet al.,
“Generation and transmission of 21.4-Gbaud PDM 64-QAM using a
high-power DAC driving a single I/Q modulator, presented at the
OFC, Los Angeles, CA, 2011, Paper PDPB2.
[7] D.Qian,M.-F.Huang,E.Ip,andY.-K.Huanget al., “101.7-Tb/s (370
294-Gb/s) PDM-128QAM-OFDM transmission over 3 55-km
SSMF using pilot-based phase noise mitigation, presented at the
OFC, Los Angeles, CA, 2011, Paper PDPB5.
[8] R.-J. Essiambre, G. Kramer, P . J. Winzer, G. J. Foschini, a nd B. Goebel,
“Capacity limits of optical ber networks,” J. Lightw. Technol., vol. 28,
no. 4, pp. 662–701, 2010.
[9] X. L iu, S. Chandrasekhar, B. Zhu, P. J. Winzer, A. H. Gnauck, and
D. W. Peckham, “Transmission of a 448-Gb/s reduced-guard-interval
CO-OFDM signal with a 60-GHz optical bandwidth over 2000 km of
ULAF and ve 80-GHz-Grid ROADMs,” presented at the OFC, San
Diego, CA, 2010, Paper PDPC2.
[10] S. Gringeri, B . Basch, V. Shukla, R. Egorov, and T. J. Xia, “Flexible ar-
chitectures for optical transport nodes and networks,” IEEE Commun.
Mag., vol. 48, no. 7, pp. 40–50, Jul. 2010.
[11] Y. Yamamoto, M. Hirano, and T . Sasaki, “A new class of optical ber
to su pp o rt large capacity transmission,” presented at the OFC, Los An-
geles, CA, 2011, Paper OWA6.
Citations
More filters
Journal ArticleDOI

Reduced Complexity Digital Back-Propagation Methods for Optical Communication Systems

TL;DR: Two proposals to reduce the hardware complexity required by digital back-propagation are discussed, one confirms and extends published results for non-dispersion managed link, while the second introduces a novel method applicable to dispersion managed links, showing complexity reductions in the order of 50% and up to 85%, respectively.
Proceedings ArticleDOI

Adaptive digital back-propagation for optical communication systems

TL;DR: An adaptive digital back-propagation method to self-determine unknown fiber nonlinear coefficient gamma, and optimal DBP performance, without knowledge of gamma, is obtained by A-DBP.
Journal ArticleDOI

The validity of ``Odd and Even'' channels for testing all-optical OFDM and Nyquist WDM long-haul fiber systems

TL;DR: Numerical simulations demonstrate that using odd and even channels generates stronger nonlinear distortions during transmission, causing an unrealistically large penalty in the nonlinearity-limited region.
Journal ArticleDOI

Nonlinearity Compensation of 224 Gb/s Dual-Polarization 16-QAM Transmission Over 2700 km

TL;DR: In this article, a low-pass filter assisted digital back propagation (DBP) algorithm was proposed for long-haul transmission for a 224 Gb/s dual-polarization 16-ary quadrature-amplitude-modulation signal, using ultralarge effective area fiber with span lengths of 75.5 km and distributed Raman amplification.
Journal ArticleDOI

Generation of square or hexagonal 16-QAM signals using a dual-drive IQ modulator driven by binary signals.

TL;DR: Polarization-multiplexed (PM)-16-QAM signals at 28 Gbuad are experimentally demonstrated and stable performance is achieved with simple bias control.
References
More filters
Journal ArticleDOI

Capacity Limits of Optical Fiber Networks

TL;DR: In this article, the capacity limit of fiber-optic communication systems (or fiber channels?) is estimated based on information theory and the relationship between the commonly used signal to noise ratio and the optical signal-to-noise ratio is discussed.
Journal ArticleDOI

Flexible architectures for optical transport nodes and networks

TL;DR: The architectures and various ROADM implementations including colorless, directionless, and contentionless add/drop structures are presented and the effect of scaling bit rates beyond 100 Gb/s onROADM architectures is reviewed including providing variable channel bandwidth depending on bit rate.
Proceedings ArticleDOI

101.7-Tb/s (370×294-Gb/s) PDM-128QAM-OFDM transmission over 3×55-km SSMF using pilot-based phase noise mitigation

TL;DR: In this paper, a record capacity transmission of 101.7-Tb/s (370×294-Gb/s) is performed over 3×55 km SSMF using PDM-128QAM-OFDM modulation and pilot-based phase noise mitigation.
Journal ArticleDOI

Spectrally Efficient Long-Haul WDM Transmission Using 224-Gb/s Polarization-Multiplexed 16-QAM

TL;DR: In this paper, the authors discuss the generation, wavelength-division-multiplexed (WDM) longhaul transmission, and coherent detection of 224-Gb/s PIMO 16-ary quadrature amplitude modulation (16-QAM) at a line rate of 28 Gbaud.
Proceedings ArticleDOI

Transmission of a 448-Gb/s reduced-guard-interval CO-OFDM signal with a 60-GHz optical bandwidth over 2000 km of ULAF and five 80-GHz-Grid ROADMs

TL;DR: In this paper, the authors demonstrate the generation and detection of a reduced-guard-interval CO-OFDM signal at 448 Gb/s, occupying 60 GHz optical bandwidth.
Related Papers (5)