Extremely Advanced Optical Transmission Technologies Issued: June 7, 207 EXAT ROAD MAP TASK FORCE Technical Committee on Extremely Advanced Optical Transmission Technologies (EXAT), The Institute of Electronics, Information and Communication Engineers (IEICE)
Content I. Trend. Terrestrial 2. Submarine 3. Ether II. Recent 3M technology. Application / Future implementation 2. Transmission capacity 3. Optical amplifier 4. Multi-level modulation. Optical fiber and fiber connector 2. Optical amplifier 3. Optical mode related device I. Trend 7 th June 207 Transpacific submarine cable system Transmission capacity per fiber [b/s] 000 P 00T 0T 00G 0. 0.0 0G 0.00 G 0.000 TDM Digital coherent TGN-P WDM-Optical Amp. TPE Unity PC- Japan-US TPC-5 TPC-4 TPC-3 China-US 00M FASTER This figure shows the transmission capacity trend in transpacific optical submarine cable systems as an example of transoceanic submarine cable system. The transmission capacity of transoceanic submarine cable systems has increased by introducing new technologies, such as optical amplifier, wavelength-division-multiplexing (WDM)and digital coherent technologies. The designed capacity of, the latest transpacific submarine cable system, "FASTER", which was placed into service in 206, reached to 0 Tbit/s using 00 channels of l00 Gbit/s WDM signals. Since the international traffic continues to increase,00 Tbit/s class system will be required at the end of 2020's. The transmission capacity in SMF, however, is limited to around 00 Tbit/s due to the limitation of fiber launch power by the fiber fuse as well as fiber non-linearlity. ln addition, the limitation of the power supply as well as the space for optical cable and repeaters must be taken in account for the submarine systems where the electric power for the repeaters is supplied from the power feeding equipment located at the landing stations., The supply voltage in current systems is already reaches around 5 kv, thus it is difficult to drastically increase the supply voltage. ln general the required power is proportionally increased as the transmission capacity increases, therefore, the power efficiency improvement is inevitable for further capacity increase. SDM using Multi-core fiber is expected to solve the above issues. The launch power in the core is decreased by using multiple cores and power and space saving in optical repeaters is expected by the integration. Development velopment and and the the fundamental fundamental technology technology for terrestrial for terrestrial optical communication communication Transmission capacity per fiber [bit/s] 0 P P 00 T 0 T 00 G 0 G G 00 M 0 M Space division multiplexing (SDM) Fiber-Launched Power Limit Nonlinear Shannon Limit Digital coherent WDM Optical amplificationion TDM :Research (with SDM) :Research (without SDM) >20 T/fiber :Commercial system in NTT 400G-bit/s/ch. SMF Limit 00 G 0 G G 00 M Channel capacity [bit/s] This figure shows the development and the fundamental technology of terrestrial optical communication. The first optical communication was implemented in 98 in Japan. Since then, the transmission capacity increases with a rate of.4 times per year which corresponds to 5 digit capacity enhancement. To exploit the maximum transmission capacity of SMF, superior technologies have been implemented including TDM, WDM-Optical amplification, Digital coherent technologies. Recently 00 Gbit/s per wavelength, and 0 Tbit/s system have been appeared by using Digital coherent technology, and further higher capacity system is still expected. Due to 5G cell phone service and IoT (Internet of Things) development, the system capacity trend is still increasing and Pbps class system will be required at the end of 2020 s. SMF transmission capacity is, however, forecasted to be maximum at around 00 Tb/s due power limitation as well as spectral efficiency saturation in SMF Space division multiplexing (SDM) is expected to realize further higher scaling technology beyond the limit, with the cooperation of Digital coherent QAM, and thus it gets high interest world-widely. Ethernet (Short reach transmission) Interface transmission [bit/s] 0000000 0T 000000 00G 00000 0G 0000 G 000 00M 0M 0Gx0 0Gx4 25Gx4 M This figure shows the evolution of Ethernet interface speed standardized in IEEE 802.3. 00GbE is the current fastest standardized interface, and 400GbE standardization is scheduled in 207, and later it will reach to Tbps at around 2020. Parallel transmission technology was introduced for l00gbe due to the difficulty of serial one. After 0 Gbps x l0 parallel interface was standardized in 200, 25Gbps x 4 parallel interface was standardized in 204. There is a possibility that the number of parallel channels increases as the interface speed increases. The ways to realize parallel interface are using ribbon-fiber in addition to WDM. The merit of ribbon-fiber is that it requires single transmitter/receiver specification while it requires several ones in case of WDM. MCF is highly expected to be exploited for more than ltbps Ether due to the high space efficiency compared to ribbon-fiber. 2
II. Recent 3M technology High-Capacity Optical Transport Network Evolution by Dense Space Division Multiplexing Phase 2 SDM ROADM for SMF bundles or SDM fiber Phase Data Center Interconnection #! Data Center Network!"#$%&'()*+,)(!"#$%-'.)*+,)(!"#$%&'()*-"#$%-'.)*+,)( /02#)%&'()*302#)%-'.)*+,)( This page shows the application area and its phase of SDM. At phase, which will be in middle time range (up to B. 2020 s), 00 km or less distance point to point w/o optical amplifier is expected including data center, interconnection, and inter-station wiring. For the implementation, a lower cost scenario is important in comparison with parallel SMF system. At phase 2 (up to M. 2020 s), submarine system across pacific ocean is expected. For the implementation, SDM optical amplification is important to support power conservation in addition to high transmission capacity. Another application is high capacity optical node system for terrestrial core-metro network. Photonic node technology is the key one for it including optical switch and optical amplifier based on SDM to realize 0 Pbps throughput with low power and highly economical system. At phase 3, which is in long time range (E. 2020 s), novel SDM fiber will be installed globally on terrestrial to replace SDM. International standardization regarding SDM fiber in addition to economical merit is desired at this stage. Output / consumption ration in multi-core EDFA Core amp. Clad amp. Scheme Single core EDFA 7 core NTT 2core EYDF Signal output per core 20.6 dbm (4 mw) 22. dbm (62 mw) 8 dbm (63 mw) OFS clad amp. 7 core 2 0 dbm (0 mw) Power consumption All core EYDF: :EYDF: Pros: Efficiency, Cont: Cont: Narrower Narrower band band : Tech. Dig. IEICE Society, B-0-27, 204 : Tech. Dig. IEICE Society, B-0-27, 204 2: Abedin et al., Opt letters, vol.39 Feb. 204. 2: Abedin et al., Opt letters, vol.39 Feb. 204. 3: 3 ECOC206 M.2.A.2 Per a core Power consumption per output power(per a core) - 9.5 W 0.080 W 33 W 4.7W 0.034 W 29 W 2.42W 0.038 W 22 W 3.4 W 0.34 W 0.3 0.0 0.05 core 7 core NTT OFS This figure summarizes recent R&D regarding SDM. Some Japanese research activities have proved the followings already; - Over Ebit/s km transmission by using MCF. - Over Pbps transmission by using Multi-core multi-mode fiber Fiber Fiber Capacity Capacity (Pbit/s) (Pbit/s) #!"#!"!# R&D in dene R&D SDM in dene SDM 00 30 0 3 499*:,0673 8-49*:,0673 8- Core Network Metro Network 4*:,0673 8-9;4*:,0673 8-0 00 000 0000 00000 Transmission Distance (km) (km) 0 4*5,0673 8- Wireless and wireline Access Network Phase 3 SDM transmission media migration Phase 2 Power-efficient submarine system Phase MCF Optical Wiring (Telecom office) Y. Miyamoto, SUM206, ME3., 206 Y. Miyamoto, SUM206, ME3., 206 Long-haul transmission using SDM DSDM 00 Distance [km] マルチコアファイバ Multi-core fiber マルチモードファイバ Few-mode /multi-mode fiber マルチコア マルチモードファイバ Multi-core multi-mode fiber 3 モード modes,000 0,000 *Dense Space Division Multiplexing This figure summarizes recent transmission experiments using SDM. Some Japanese research activities have proved the followings; - Over l Ebit/s km transmission by using MCF. - Over 0 Pbit/s transmission by using Multi-core multi-mode fiber Ether due to the high space efficiency compared to ribbon-fiber. This figure summarizes recent R&D regarding SDM in terms of space channel as a function of distance. Some Japanese research activities have proved the followings already; - 00 space channel 0 km transmission using multi-core and multi-mode - 36 space channel 500 km class transmission using multi-core and multi-mode - 32 space channel 000 km transmission using multi-core 7 space channel over ocean transmission using multi-core Spectral efficiency as a function of distance in multi-level modulation scheme Spectral efficiency as a function of distance in multi-level modulation scheme Spectral Efficiency per pol. [bit/s/hz] Relative Required SNR [db] Long haul/ Metro (backward compatibility) Shot reach (compact and power efficient) This figure shows the spectral efficiency as a function of distance in multi-level modulation scheme. Multi-level QAM modulation is important fundamental technology in terms of highly spectral efficient WDM. Japanese R&D activity has realized 6QAM transmission as a commercially available level, and over 024QAM has been also reported from Japanese EXAT related organization. In QAM modulation format, transmission distance decreases when multi-level increases due to the narrowing of spectral spacing which degrades S/N ratio in general. For this reason, multi-level has a trade off relation between transmission distance, and thus, it is necessary to consider the applicable field while further development is expected in this research field. 3 4
A scenario of fiber This figure shows a future scenario of spatial channel per single fiber. Blue: MCF, Red: FMF, Green: FM-MCF. SDM fiber R&D has been active since 200, and around 30 channels has already been demonstrated in MCF or FMF, and maximum 4 channels was reported in FM-MCF. Getting mature in fabrication as well as transmission technology toward actual implementation. ln general, fiber reliability degrades as cladding diameter increases, therefore, one possible scenario is to utilize existing standard diameter (i.e. 25 μm cladding diameter or 250 μm coating diameter) with limited spatial channels of around 0 for the early stage implementation of MCF. Later, implementation of much thicker diameter rather than present standard will realize more than 0 channels MCF, and over 30 channels FM-MCF. Short distance application (up to 0 km intra/inter building with connector manufactured MCF) is the early exploitable area in terms of cost and fiber-splicing ease. Later on, over 00 km application will be expected. The required crosstalk level in fiber will be changed as follows; -25 db or less for 0 km, -35 db or less for 40 km, and -40 db or less for 00 km. While in submarine system, 200 km application will be realized faster due introducing remote pumping or distributed Raman amplification, and then SDM amplification and mode control matureness enables 000 km class transmission system. SDM amplifier based submarine system may be implemented in advance depending on clad pumping technology development. A scenario of connector Larger clad SM- MCF connector w/o center connector 25 m clad SM-MCF connector High accuracy FMF Connector FM-MCF connector <SM-MCF+FMF> or <FM-MCF> <FM connector technology> This figure shows a scenario of fiber connector. SM-MCF connector is the present R&D topic. MCF rotation control is a key to realize low connecting loss. Higher accuracy in rotation will be required as core number increases in SM-MCF because of the affection of outer layout core position error. In 25 um cladding MCF, center core less core arrangement is investigated. Low connection loss has been realized based on center-core alignment, and later, center core less core connection will be required. To realize low loss connection in FMF, highly accurate connection against higher order mode is required in addition to the evaluation technology of FMF. For this reason in this roadmap, R&D in FMF connector is also take in account simultaneously with SM-MCF connector. The combination of SM-MCF connector and FMF connector, or new FM-MCF connector technologies enable higher channel FM-MCF connector. A scenario of fiber 2 A scenario of amplifier Relative core density against SMF This figure shows a future scenario of relative space channel density per single fiber. Blue: MCF, Red: FMF, Green: FM-MCF. Similar content with the previous figure, however, the vertical axis is replaced to relative core density. lt is defined as total summation of effective area of each core in MCF against cladding area size. ln case of FM-MCF, each mode numbers are also taken in account. Relative core density is calculated against SMF core density, therefore, relative core density of 20 corresponds to 20 times space utilization efficiency against SMF. Energy efficiency / Core This amplifier scenario offers a view in terms of power consumption. At phase, short distance SDM is expected. In this case, EDFA will not be used due to the distance. At phase 2, long-haul such as submarine SDM is expected. 2 times high effectiveness in power is assumed, thus batch amplification for SDM fiber is expected. At phase 3, terrestrial and flexible network SDM are expected. Based on higher core numbers and fiber optimization, 0 times higher efficiency is assumed in this case. Furthermore, with the combination of mode-division multiplexing, 00 times higher efficiency will be realized at the end. 5 6
A scenario of mode-division multiplexing device A scenario of mode-division multiplexing device Mode Crosstalk [db] 30 0 2020 ~ Short distance (DC) MIMO less ns switch 2025 ~ Short distance (DC) MIMO less Sub ns switch 2030 ~ This figure shows a scenario of mode-division multiplexing device. At phase, short distance SDM will be realized in advance. Later on, similar technology will be penetrated into also long-haul transmission. For mode-division multiplexing device, one important characteristics is the crosstalk, therefore, this figure is summarized in terms of it. 0 00 000 Channel / Mode Number A scenario of mode combiner Mode Mode Number Number 00 0 C: + mode multiplexing ~0km No MIMO ~40km MIMO/no MIMO ~00km MIMO 970 2040 This figure shows a scenario of mode-division multiplexing device. One important point of view is no MIMO enabling technology because latency in addition to cost is a key issue in short distance application, especially in data center. The present data in this figure includes optics, phase plates, and waveguides. Superior crosstalk characteristic is the most important specification for no MIMO mode-division multiplexing device. For this reason, another mode candidate itself will be also important research topic including single dimensional mode system (including OAM (orbital angular momentum, super-mode, and others) apart from LP modes in regular FMF. Acknowledgement The EXAT roadmap has been discussed and arranged by the task force member in EXAT technical group in Japan during FY205-206. TF member Kiichi Hamamoto, Yutaka Miyamaoto 2, Kaxuhide Nakajima 2, Itsuro Morita 3, Toshiharu Ito 4, Kazuhiko Aikawa 5, Takashi Sugihara 6, Ryuichi Sugizaki 7. Kyushu Univ., 2. NTT, 3. KDDI, 4. NEC, 5. Fujikura, 6. Mitsubishi Electric, 7. Furukawa Electric Technical Committee on Extremely Advanced Optical Transmission Technologies (EXAT), The Institute of Electronics, Information and Communication Engineers (IEICE)