Te Subcom Sets Record for Transmission Capacity Using C+L Technology and Nokia & Facebook Push Undersea Fiber to 32 Tbit/s
TE SubCom, a TE Connectivity Ltd. company and an industry pioneer in undersea communications technology, today announced that it has demonstrated a new transmission record of 70.4 Tbit/s capacity over 7,600 km. The record transmission is made possible utilizing SubCom’s newly available C+L technology, which includes amplifiers with nearly 10THz of usable optical bandwidth, a transmission line designed to minimize noise accumulation and maximize performance, a new transmission format using multi-dimensional coded modulation with hybrid probabilistic and geometrical constellation shaping, and digital signal processing featuring several stages of non-linear compensation. The results were announced at the Optical Fiber Communication (OFC) Conference 2017, which took place March 19-23 in Los Angeles, California.
As previously announced, SubCom will deploy C+L in an undersea cable system which commenced manufacturing in late 2016. SubCom’s C+L effectively doubles the available transmission bandwidth of the repeater by an unprecedented factor of two in supported capacity per fiber pair when compared to the same number of fiber pairs in traditional C-band technology. Since only half the number of fiber pairs are needed compared to a C-band repeater, C+L technology is much more cost competitive and opens the path to the record capacities that SubCom is achieving today.
“70.4 Tbit/s a staggering figure,” said Neal Bergano, chief technology officer, TE SubCom. “With accelerating data center growth, developments in virtual and augmented reality, 4k and 8k video and machine-to-machine technology, bandwidth demands will continue unabated. Cable operators will continue to seek the fastest, highest capacity and most cost-effective systems available to them, and this new record positions C+L technology as the industry standard. We couldn’t be prouder of our research and development team for this tremendous achievement.”.
Nokia and Facebook have claimed to more than double the capacity of existing fiber on a 5,500km trans-Atlantic subsea cable in tests with Nokia Bell Labs' new probabilistic constellation shaping (PCS) technology.
The companies collaborated in order to increase capacity capabilities of subsea fiber and enhance the support for telecommunications infrastructure coping with surging bandwidth-hungry applications, such as video and virtual reality.
According to Nokia, the field trials of new optical digital signal processing technologies, which used shaped 64-QAM, over the subsea link between New York and Ireland achieved a record spectral efficiency of 7.46 b/s/Hz and increased the stated capacity of the system by almost 2.5 times. It indicates the potential to upgrade this cable to 32Tbps per fiber in the future.
The test also included a successful 11,000km round trip submarine transmission using shaped 64-QAM achieving a record spectral efficiency of 5.68 b/s/Hz and a first demonstration of 200 Gbit/s and 250 Gbit/s wavelengths and 16-QAM modulation over a trans-Atlantic submarine route using a real-time coherent DSP.
"Facebook wants to increase the pace of innovation and adoption of next-generation optical technologies,” said Dr. Stephen Grubb, global optical network architect at Facebook.
“This field trial with Nokia demonstrates that the scalable optical technology of PCS together with narrow line-width laser sources can achieve capacities extremely close to the Shannon limit. This ensures that we are both maximising our investment in submarine cable systems, as well as continuing to drive the cost per bit of submarine transport lower."
PCS, an area of active Nokia Bell Labs research, is a novel technique that uses 'shaped' quadrature amplitude modulation (QAM) formats to flexibly adjust transmission capacity to near the physical limits of a given fiber-optic link.
Sam Bucci, head of optical networking at Nokia, added: "We are thrilled to partner with Facebook to promote our common commitment to accelerating innovation in optical transmission. By demonstrating promising areas of Nokia Bell Labs research such as PCS, as well as coherent technologies available today, we hope to chart a path forward for the industry towards higher capacities, greater reach, and more network flexibility."
The news comes after Facebook and Microsoft announced in May 2016 that they had teamed up to build a 6,600km subsea cable system the firm claims “will be the highest-capacity subsea cable to ever cross the Atlantic with speeds of 160 Tbit/s”. The system, which has TE SubCom selected as MAREA’s supply partner, will feature eight fiber-pairs, will also be the first to connect the US to southern Europe: from Virginia Beach, Virginia, to Bilbao, Spain. The deployment of MAREA is being operated by Telxius and is on time to be ready for service by the end of 2017.
Source: TE SubCom Press Release & Capacity Media
ANALYSIS: In January 2016 at the annual PTC conference in Hawaii, Neil Bergano (quoted above) talked about how suppliers are addressing insatiable demand growth. He said that 400 Gbit/s is the absolute limit for a wavelength in the "C" (Common) channel but that using C&L bands together could double the available bandwidth. He also predicted that future submarine cables will be physically larger. Power is a critical but solvable issue. In response, Frank Rey of Microsoft responded, "We have time to work out the roles of suppliers and providers so that technology can stay ahead of demand". Facebook's activities are now reflecting this view.
SubCom has therefore continued to follow the path of looking to get as close as possible to the Shannon Limit and has actually exceeded expectations with this "hero experiment". However, although 32 Tbit/s per fiber pair cables will be sufficient for the next few years, with worldwide demand for international capacity still growing at somewhere close to 50% per year, using C+L only buys two additional years before demand exceeds supply.
Facebook's recruitment of Steve Grubb from Infinera has had a profound impact on the company's public statements on this issue. Facebook is now taking a lead in pushing R&D to find a solution and is working with both TE SubCom and Nokia ASN. It looks like it will be TE SubCom that gets the first opportunity to demonstrate C+L on the transatlantic "Marea" system which should be ready for service in the first quarter of 2018.
It is also interesting to note that, in January 2017, NTT announced a breakthrough in the use of multimode fiber in submarine fiber optic cables. TE SubCom published a paper in the same area at the OFC conference in 2016 entitled "105.1 Tb/s Power-Efficient Transmission over 14,350 km using a 12-Core Fiber" by A. Turukhin, O.V. Sinkin, H. G. Batshon, H. Zhang, Y. Sun, M. Mazurczyk, C. R. Davidson, J.-X. Cai, M. A. Bolshtyansky, D. G. Foursa, and A. Pilipetskii.
The reason that NTT says that it is still 10 years away from commercializing this technology is that an entirely new undersea amplifier would be required. Such an amplifier would have to be smaller than twelve of today's amplifiers (which are in any case getting smaller), consume less power, and be as reliable or better. Furthermore, this technology does not improve the economics of the terminal equipment.
These "hero experiments" are collaborative. Generally, a carrier would work with a fiber maker to make the fiber, an optics supplier to make interface hardware, and other manufacturers to make the terminal equipment. TE SubCom and Nokia ASN perform the integration.
Large carriers like NTT have significant research resources. All they need is one researcher with an activity in this area. They license a promising technology to a supplier to build if they want it commercialized. AT&T also maintains an advanced optics research lab that works on such experiments.
One might imagine that the impact of the introduction of multimode fiber on the submarine fiber optic market, if commercialized, would be to end current talk of hitting Shannon's limit. We would not have to build more cables to meet demand since we would be able to get an order of magnitude more throughput at the same cost per bit.
However, in reality, with or without the NTT experiment, new cables will still be needed because multimode fiber technology would require a whole new family of technology that is not commercially available today, including optical fibers and amplifiers.
Regardless of whether or not NTT can commercialize this technology, today's cables will top out at around 70 Tbit/s per fibre pair due to the Shannon Limit. When the current generation of submarine cables become full, new cables will be needed, either based on technology similar to today's system, or based on some other technology of tomorrow.
Julian Rawle, Author
Thought leadership articles and commentary on developments related to the subsea fibre optic cable industry can be found here.