Open Access
Add to BibSonomyAbstract
We demonstrate the first 7-core multicore erbium-doped fiber amplified (MC-EDFA) transmission of 40 x 128-Gbit/s PDM-QPSK signals over 6,160-km 7-core multicore fiber (MCF). The crosstalk (XT) from all of the other 6 cores of a MC-EDFA and a 55-km length MCF are about −46.5 dB and −45.6 dB at center core, respectively. The core-to-core rotation approach at every amplified span is used to average the XT of all cores. The averaged optical signal-to-noise ratio (OSNR) after 6,160-km transmission is 15.6 dB with 0.1 nm resolution bandwidth. The Q-factor of all 40 channels surpasses the threshold of the forward-error-correction of 6.4 dB with 1 dB margin after 6,160 km. The total net capacity is 28.8 Tbit/s per fiber and achieved capacity-distance product is 177 Pbit/s.km per fiber. We confirmed the feasibility of MC-EDFA repeatered systems for trans-oceanic transmission.
©2013 Optical Society of America
1. Introduction
To fulfill the endless demand for higher-capacity transmissions, diverse high-capacity transmissions have been demonstrated over traditional single-core fiber. So far, the highest capacity-distance product (CDP) of 203 Pbit/s.km has been reported with 30-Tb/s transmission over 6,630 km [1] of single-core fiber (SCF) with spectral efficiency (SE) of 6.1bit/s/Hz per fiber. However, further enhancement in either the capacity or distance is very challenging due to the limited optical signal-to-noise ratio (OSNR). Facing such a bottleneck, one promising solution to overcome the capacity limit is the use of space-division-multiplexing (SDM) based on multicore fiber (MCF) [2–4] or multi-mode fiber [5]. Indeed, several SDM transmissions have been reported: the record capacity of 1 Pb/s over 52-km distance has been demonstrated over 12-core MCF [2] with aggregated SE of 91.4 bit/s/Hz per fiber, and the first demonstration of repeatered MCF transmission for 2688 km with 1.28-Tb/s capacity has been achieved over 7-core fiber [3] with aggregated SE of 15 bit/s/Hz per fiber. The reported longest distance with MCF was 4200 km with 3-core MCF [4] with single-channel. These experiments relied on single-core (SC-) EDFA.
One of the remaining key milestones to the realization of optical transmission systems based on MCF, is the introduction of multicore EDFA (MC-EDFA) [6,7]. The MC-EDFA has an attractive possibility for downsizing and low power consumption compared to the multiple SC-EDFAs [6]. However, MC-EDFA intrinsically generates XT between cores, which impairs long-distance transmission. Therefore the feasibility of long-haul transmission using MC-EDFA still needs to be demonstrated. In this paper, we demonstrated experimentally trans-oceanic class transmission using 55-km spans of 7-core MC-EDFA [6] and 7-core MCF. We confirmed a reachable distance of 6,160 km with 40 x 128-Gbit/s PDM-QPSK signals per core (total net capacity: 28.8 Tbit/s) with aggregated SE of 14.4 bit/s/Hz per fiber. To our best knowledge, this is the world first transmission experiment using MC-EDFA and MCF, and the achieved CDP is 177 Pbit/s.km per fiber.
2. Experimental setup
Figure 1 shows the experimental setup with MC-EDFA and MCF [8]. The transmission line consists in a loop composed of a span of a 55-km length 7-core MCF, a 7-core MC-EDFA, gain-flattening filters (GFF) and two single-core EDFAs. The cores of the MCF are connected in series using the core-to-core rotation approach [3], i.e. the output of core #1 of MCF is connected to core #1 of MC-EDFA, and the output is connected to core #2 of MCF, and so on for the seven cores, forming 7 spans. In our loop structure, the XT is averaged between the 7 cores. The GFFs are inserted between the cores and two single-core EDFAs are used to compensate the excess loss of the GFFs at core #4 and the optical switch for the loop operation. The gain of SC-EDFA at core #4 is about 5 dB. The output of core #7 of MC-EDFA is connected to the loop switch through a GFF.
The transmission span is composed of 55-km length 7-core MCF and fan-in (FI) and fan-out (FO) devices. Figure 2a , 2b and 2c show the cross-section of MCF, FI/FO, and the end face of fiber bundle of FI/FO. In order to obtain both high SDM density and low XT, simultaneously, the cladding diameter and the core pitch of the MCF are designed to be less than 200 μm and 56 μm, respectively. The main characteristics of the MCF at 1550 nm are an effective area (Aeff) of 99 μm2, a cable cut-off wavelength of 1390 nm, an attenuation loss of 0.188-0.200 dB/km and a chromatic dispersion of 18.4-18.7 ps/nm/km. The total span loss between the FI input and the FO output port including the 55-km MCF and two fusion splice points ranged from 11.4 to 12.3 dB at 1550-nm.
Figure 3a , 3b and 3c show the cross-section, composition, and gain and noise figure of the 7-core MC-EDFA which is prepared to compensate the span loss as shown in Fig. 1. Since the mode field diameter (MFD) of MC-EDF is about 7.3 μm which is much smaller than that of the MCF, a core pitch of 45 μm is sufficient to reduce the crosstalk. The MC-EDF has an attenuation coefficient of ~3.4 dB/m and a small-signal gain of ~4.3 dB/m at 1550 nm. Figure 3c shows the gain and noise figure with 980 nm forward and backward pumping as a function of wavelength. The gain and noise figure of the 7-core MC-EDFA as shown in Fig. 3b were estimated by considering the insertion loss of the WDM coupler, isolator, and FI/FO. Especially the insertion loss of FI/FO was estimated to be up to 1.0 dB because of the MFD mismatch between MC-EDF and thin fiber (MFD is about 10.0 μm). The solid curves and dashed curves correspond to the gain and the noise figure, respectively. As shown in this figure, this MC-EDFA has a maximum gain of about 15 dB and typical noise figure of 7 dB over the C-band.
Fig. 3 7-core MC-EDFA, (a) cross-section of MC-EDF, (b) composition, (c) gain and noise figure of MC-EDFA.
Figure 4a and 4b show the XT between two cores of MCF and MC-EDFA, respectively. The core number 1 represents the center core of MCF and MC-EDFA. The XT denoted as “A-B” is defined by the power ratio between the signal power transferred to the B core and the power remaining in the A core. XT occurs mainly between adjacent cores, therefore the evaluated XT at respective outer and center core are comes from the respective 6 neighboring cores and 3 neighboring cores. For the confirmation, the XT coming from the non-adjacent 3 cores is also measured for outer cores of MCF. The XT per span of the MCF is less than −50 dB for all channels. For geometrical reasons, the center core is the worst channels, as it is surrounded by 6 cores. The total XT of MCF and MC-EDFA at center core are estimated to be −45.6 dB and −46.5 dB per span, respectively. Accordingly, the center core of total XT is −43 dB per span. The XT per loop, i.e. after transmission through all 7 cores, is −36.9 dB. For example, accumulated XT after 6,160 km transmission with 112 spans would be about −22.5 dB at center core without core-to-core rotation approach; however, the XT is averaged and reduced to be −24.8 dB with the core-to-core rotation approach.
In the transmitter, the 40 channels are generated form external cavity lasers (ECLs) with linewidths in the order of 100 kHz and DFB-LDs. DFB-LDs are replaced by ECL when the corresponding channel is measured. The lasers are aligned in 50 GHz spacing from 1550.116 nm to 1565.905 nm. The even and odd channels are multiplexed with arrayed waveguide grating (AWG) individually. The channels are modulated by optical IQ-modulators (IQMs). The baseband signal of 32 Gbit/s with pseudo-random bit sequence (PRBS, 215-1) is generated from a pulse pattern generator (PPG). The IQMs modulate even and odd channels into QPSK individually. Even and odd channels are combined with a 100-50 GHz wavelength interleaver (IL). These signals are fed into a polarization multiplexing emulator (PME). The nominal data-rate after PDM is 128 Gbit/s. This assumes 20% overhead for soft-decision forward-error-correction (SD-FEC) with OTU4 framing over 103.125-Gbit/s Ethernet payload [9].
At the receiver, the measured channel is extracted by optical band-pass filters and converted by four balanced photo-diodes (balanced PDs) after the polarization-diversity 90 degree hybrid. The electrical output signals from balanced PDs are digitized by the analog-digital converters (ADC) of a realtime oscilloscope, at a sampling-rate of 50 GSa/s and with a bandwidth of 16 GHz. The received signals are demodulated by offline digital signal processing. The procedure is as follows: 1) deskew and orthogonalization, 2) digital filtering with (0.6/T) 3-dB bandwidth, where T is the symbol duration, 3) re-sampling at two samples per symbol, 4) frequency-domain chromatic dispersion compensation, 5) clock recovery, 6) polarization-dependent (2x2), 31-tap, T/2-spaced adaptive butterfly finite impulse response filter (FIR) with constant modulus algorithm, 7) carrier recovery using decision-directed phase locked loop, 8) decision and BER evaluation based on direct error counting.
3. Generation of the SDM signals in the loop structure
Figure 5 shows the schematic explanation for the generation of 7 parallel signals and for the path of each signal in the loop. The loop transmission line should simultaneously have 7 independent paths named as path I to VII consisting in the 7 cyclic permutations of [#1,#2,...,#7] to [#7,#1,...,#6]. One way to evaluate all possible paths would be changing the position of the loop switch to generate all cyclic permutations and evaluate the characteristics in each configuration.
Fig. 5 (a) The schematic explanation to generate 7 parallel signals and path for each signal at loop, (b) signal transmission of each path.
Here, we keep the position of the loop switch fixed between cores #7 and #1, as shown in Fig. 5a, and the signal transition of each path is shown in Fig. 5b. We use the fact that N + 1 laps of [#1,#2,...,#7] include N laps of any of the other permutations [#2,#3,...,#1] to [#7,#1,...,#6] as shown in Fig. 5b. Therefore, the single evaluation of (N + 1) laps in the configuration [#1,#2,...,#7] is a sufficient criterion to characterize the transmission through N laps with any other permutation as it is more severe. For example, an actual distance of 6,545 km with 17 laps is considered as an effective transmission distance of 6,160 km with 16 laps for the 7 paths.
4. Experimental results
Figure 6 shows measured back-to-back Q-factor of the 128-Gbit/s PDM-QPSK signal as a function of optical signal-to-noise ratio (OSNR) with/without IL at Tx. The inset shows the constellation without additional amplified-spontaneous-emission (ASE) noise. The Q-factor is calculated from the BER. OSNR is defined with a noise bandwidth of 0.1 nm. No penalty is observed with the filtering effect of IL.
The dependency of fiber launch power is measured after an effective transmission of 6,160 km as shown in Fig. 7a . The optimal performance is observed around −1 dBm/ch. The relationship between Q-factor and distance is shown in Fig. 7b. The OSNR and estimated Q-factor from OSNR after transmission are also shown. The fiber launch power is set to −1 dBm/ch. The OSNR after 6,160-km transmission is about 15.6 dB with 0.1 nm resolution bandwidth. The transmission performance of the near center channel is measured with the wavelength of 1557.768 nm. As mentioned in the previous section, one lap of loop transmission is subtracted to derive the transmission distance. The Q-penalty caused by nonlinear effect and XT after 6,160-km transmission is estimated to about 1.4 dB by referring to the OSNR. This indicates that the impact of XT is very low even after 6,160-km transmission of MCF with more than 100 MC-EDFAs.
Fig. 7 Transmission performances, (a) The dependency of fiber launch power, (b) Q-factor vs Effective transmission distance.
Figure 8 shows the optical spectrum of 40 x 128 Gbit/s PDM-QPSK signals after 6,160- km transmission. Figure 9 summarizes the constellation maps and results of performance measurement after 6,160-km transmission. The worst channels are observed around the channel with the shortest wavelength because of the limitation of gain bandwidth. The Q-difference between the best and worst channel is about 1.1 dB. Q-factors of all of the received channels have over 1-dB margin regarding to the limit of SD-FEC with 20% overhead, namely 6.4 dB [9].
5. Further discussions
In this experiment, core-to-core rotation approach was used. However, the possible practical configuration may be the direct coupling between MCF and MC-EDFA without FI/FO. In this case, the accumulated XT after 6,160 km transmission with 112 spans would be about −22.5 dB (0.0056 in linear) at center core, while the XT is averaged and reduced to be −24.8 dB (0.0033 in linear) with the core-to-core rotation approach as mentioned at section 2. The one of important points is the ratio between accumulated XT and OSNR. The OSNR after 6,160 km is 15.6 dB with 0.1 nm resolution bandwidth, therefore the power ratio between ASE noise and signal per channel spacing of 50 GHz is estimated as −9.6 dB (0.109 in linear). From these values, the ratio between signal and ASE noise with XT will be 9.4 dB at center core with direct coupling, and 9.5 dB with core-to-core rotation approach, respectively. It means that even if the MCF and MC-EDFA is directly connected, the reduction of reachable distance will not be so significant.
6. Conclusions
We demonstrated the world-first experiment with 7-core MC-EDFA repeatered trans-oceanic class transmission of 40 x 128-Gbit/s PDM-QPSK signals over 6,160-km 7-core MCF. The estimated total XT after 6,160 km is about −24.8 dB with core-to-core rotation approach in this experiment. The Q-factor after transmission of 6,160 km is over 7.4 dB which is 1 dB above the SD-FEC limit of 6.4 dB. The total net capacity and achieved capacity-distance product are 28.8 Tbit/s and 177 Pbit/s.km per fiber, respectively. The estimated transmission Q-penalty caused by nonlinearity and/or XT is about 1.4 dB. This result indicates that the XT does not cause a significant reduction of the reachable distance in this experiment. From these results, we confirmed that MC-EDFA repeatered trans-oceanic transmissions are feasible. The transmission system with MCF and MC-EDFA is a promising candidate for next generation high capacity transmission systems.
Acknowledgments
The research results have been achieved by “Research on Innovative Optical Fiber Technology” and “R&D of Innovative Optical Communication Infrastructure”, the Commissioned Research of National Institute of Information and Communications Technology (NICT), JAPAN.
References and links
1. M. Mazurczyk, D. G. Foursa, H. G. Batshon, H. Zhang, C. R. Davidson, J.-X. Cai, A. Pilipetskii, G. Mohs, and Neal S. Bergano, “30 Tb/s Transmission over 6,630 km Using 16QAM Signals at 6.1 bits/s/Hz Spectral Efficiency,” ECOC2012, Th.3.C.2.
2. H. Takara, A. Sano, T. Kobayashi, H. Kubota, H. Kawakami, A. Matsuura, Y. Miyamoto, Y. Abe, H. Ono, K. Shikama, Y. Goto, K. Tsujikawa, Y. Sasaki, I. Ishida, K. Takenaga, S. Matsuo, K. Saitoh, M. Koshiba, and T. Morioka, “1.01-Pb/s (12 SDM/222 WDM/456 Gb/s) Crosstalk-managed Transmission with 91.4-b/s/Hz Aggregate Spectral Efficiency,” ECOC2012, Th.3.C.1.
3. S. Chandrasekhar, A. Gnauck, X. Liu, P. Winzer, Y. Pan, E. C. Burrows, B. Zhu, T. Taunay, M. Fishteyn, M. Yan, J. M. Fini, E. Monberg, and F. Dimarcello, “WDM/SDM Transmission of 10 x 128-Gb/s PDM-QPSK over 2688-km 7-Core Fiber with a per-Fiber Net Aggregate Spectral-Efficiency Distance Product of 40,320 km.b/s/Hz,” ECOC2011, Th.13.C.4.
4. R.Ryf, R. Essiambre, A. Gnauck, S. Randel, M. A. Mestre, C. Schmidt, P. Winzer, R. Delbue, P. Pupalaikis, A. Sureka, T. Hayashi, T. Taru, and T. Sasaki, “Space-Division Multiplexed Transmission over 4200 km 3-Core Microstructured Fiber,” OFC/NFOEC2012, PDP5C.2.
5. E. Ip, N. Bai, Y. Huang, E. Mateo, F. Yaman, S. Bickham, H. Tam, C. Lu, M . Li, S. Ten, A. P. T. Lau, V. Tse, G. Peng, C. Montero, X. Prieto, and G. Li, “88×3×112-Gb/s WDM Transmission over 50 km of Three-Mode Fiber with Inline Few-Mode Fiber Amplifier,” ECOC2011, Th.13.C.2.
6. Y. Mimura, Y. Tsuchida, K. Maeda, R. Miyabe, K. Aiso, H. Matsuura, and R. Sugizaki, “Batch Multicore Amplification with Cladding-Pumped Multicore EDF,” ECOC2012, Tu.4.F.1.
7. Y. Tsuchida, K. Maeda, K. Watanabe, T. Saito, S. Matsumoto, K. Aiso, Y. Mimura, and R. Sugizaki, “Simultaneous 7-Core Pumped Amplification in Multicore EDF Through Fibre Based Fan-In/Out,” ECOC2012, Tu.4.F.2.
8. H. Takahashi, T. Tsuritani, E. L. T. de Gabory, T. Ito, W. R. Peng, K. Igarashi, K. Takeshima, Y. Kawaguchi, I. Morita, Y. Tsuchida, Y.Mimura, K. Maeda, T. Saito, K. Watanabe, K.Imamura, R. Sugizaki, and M. Suzuki, “First Demonstration of MC-EDFA-Repeatered SDM Transmission of 40 x 128-Gbit/s PDM-QPSK Signals per Core over 6,160-km 7-core MCF,” ECOC2012, Th.3.C.3.
9. T. Mizuochi, Y. Miyata, K. Kubo, T. Sugihara, K. Onohara, and H. Yoshida, “Progress in Soft-Decision FEC,” OFC/NFOEC2011, NWC2.
