Future Proofing Cable’s Optical Access Network: “A Coherent Story”
The demand for data network capacity has been growing exponentially year after year with no sign of stopping. If the past is a guide to the future, the cable industry must come up with radically more efficient use of the existing cable infrastructure in order to meet demand.
Today, the most constrained part of the network, and the most costly to upgrade, is the fiber infrastructure between the headend and the fiber node, to the wireless cell radio or to large business customers. Avoiding costly fiber re-trenching requires a fundamentally new approach to this part of the network. This is where coherent technology provides an opportunity.
If you are familiar with coherent optics, then you are aware that the technology has been used in long-haul fiber optic networks for decades. CableLabs has adapted that technology for use in short-haul access networks, and simplified it to reduce the cost. And it has much higher capacity for future growth than the analog optics that are used in many of today’s HFC networks – possibly more than 1,000 times more capacity! We have already demonstrated 50 times more capacity than analog optics can achieve today by using coherent optics on 80 kilometers of fiber, and more improvements are on the way.
Using traditional analog optics, to achieve that high transport medium quality requires increasing the optical transmit power level, which unfortunately reduces the number of optical analog carriers the fiber can support due to fiber non-linearity effects. Figure 1 shows a representation of fiber’s wavelength spectrum with 4 analog optical carriers.
Figure 1 – Fiber spectrum with 4 analog carriers
This limitation on analog optical transport has prompted the cable industry to look into other architectural evolution approaches. One approach that solves the analog limitation problem, while also tackling space limitations that may exist in certain hubs, is the distributed architecture approach. In a distributed architecture no radio frequency (RF) is transported through the optical link. The optical link does not contribute to distortion of the DOCSIS® RF signal, only the coaxial portion of the network is responsible for the degradation of the RF signal.
Today, this digital optical link uses intensity modulated direct detection systems such as the ones found in 10 Gigabit Ethernet links and in passive optical networks (PON). In these non-coherent systems, the signal modulation used is On-Off Keying (OOK). OOK is achieved by simply switching the laser source off and on. Non-coherent systems operate at lower power than analog optics and can therefore make better use of the wavelength spectrum in fiber. Figure 2 depicts the wavelength spectrum of fiber with several non-coherent OOK optical carriers.
Figure 2 – Fiber spectrum with intensity modulated non-coherent carriers
A high-capacity non-coherent (but digital) optical link can carry 100 Gbps using 10 wavelengths (optical carriers) carrying 10 Gbps each. Non-coherent systems are a suitable near term approach, but there are additional fiber resource challenges that have to be considered when evaluating a long-term strategy.
HFC Networks have been typically designed with 6 to 8 fibers connecting the hub to the fiber node. Two of these fibers are used for primary downstream and upstream connection and in some cases two additional fibers are used for redundancy purposes. The rest of the fibers were left for future use. Unfortunately, a large amount of these ‘future use’ fibers, because of an ever-increasing demand for bandwidth, have since been repurposed for business services, cell backhaul, node splits and fiber deep architectures. In some cases, only the two primary fibers that are feeding the fiber node remain available for access transport.
In fact, the architecture migration toward Full Duplex DOCSIS, which strives towards a more symmetric transport, relies on a node plus zero amplifiers (N+0) architecture. A typical node in today’s HFC networks will supply services to 500 households. When converted into an N+0 architecture, the result is the creation of 12 to 18 deeper N+0 nodes. The challenge for the optical portion of the access network becomes supplying enough bit rate capacity to 12-18 N+0 nodes, each capable of supplying 10 Gbps to residential subscribers.
This fiber shortage problem will only intensify as fiber demand for business services and wireless backhaul increases. Assuming that costly fiber re-trenching from hub to original fiber node is to be avoided, a different solution must be found to provide the required capacity. This is where coherent technology provides an opportunity!
Coherent technology has been used to achieve higher speeds than any other optical technology. In coherent optics, both amplitude and phase modulation are used to put information onto an optical carrier. This enables the generation of Quadrature Phase Shift Keying (QPSK) and Quadrature Amplitude Modulation (QAM) constellations carrying information. The nature of the coherent signal also allows the separation of the optical signal in two orthogonal polarizations. Each polarization can independently carry the two-dimensional constellations mentioned above. The signal processing used in coherent systems facilitates shaping of the spectrum of the signal to avoid interference with adjacent optical carriers. Along with the much lower power requirements, coherent technology allows for efficient multiplexing of the optical carriers within the wavelength spectrum of fiber. Figure 3 shows the wavelength spectrum of fiber with coherent optical carriers over 2 polarizations.
Figure 3 – Efficiently packed coherent optical carriers over orthogonal polarizations
Coherent optics has been used in the long-haul environment for over 30 years. The long-haul environment is a harsh environment that consists of very long distances, sometimes up to 3000 km. In a long-haul environment, significant channel compensation is employed to correct the long distance related impairments, making long haul solutions expensive.
The access network environment is very different from long haul networks in one key respect, optical links in access networks are typically no longer than 30 km. That is two orders of magnitude shorter than long haul. The complex and expensive system implementation that long haul is known for no longer applies to access implementation. The shorter fiber lengths result in minimal dispersion of the optical signal. Furthermore, since no in-line amplification is needed, non-linear distortion and noise are significantly reduced. This increases the link margin and enables much lower implementation costs. It is NOT your father’s coherent implementation!
Here at CableLabs® we have re-engineered the coherent link to meet the special conditions of the access network. We have developed technology that is higher performance and much lower cost when compared to long-haul or metro environments.
In the laboratory, we have achieved 256 Gbps over 80 km on a single wavelength with minimal dispersion compensation. That is ~26 times the capacity of what can be achieved over an analog optical carrier fully loaded with 1.2 GHz worth of DOCSIS 3.1 signals. We have achieved that using a symbol rate of 32 GBaud (32 GHz), using 16QAM modulation (4 bits per symbol) over 2 polarizations (32*4*2=256 Gbps). In addition, we have multiplexed eight of these wavelengths to achieve 2048 Gbps. That is 50 times more than what can be achieved over 4 analog optical carriers each with 10 Gbps of DOCSIS 3.1 payload!
The optical access environment could lend itself to further improvement in capacity per wavelength by further increasing symbol rate and/or modulation order. A future achievement of 64 QAM modulation could represent the pinnacle in efficiency and capacity per wavelength of our optical access environment. One can only dream of such transport efficiencies in the long-haul environment.
Coherent optics is extremely flexible. Capacities per wavelength greater than 256 Gbps may not be needed at each target end-point in the near future. Maybe 100 or 200 Gbps will do. The fact that modulation order, polarization and symbol rate can be varied enables significant flexibility in the type of supported services. Lower symbol rates allow for multiplexing 100 or 200 Gbps wavelengths to end points. In the access network, it makes sense to dedicate a single wavelength to a target end point (subscriber). In the access, since wavelength spectrum is a precious commodity, higher speed should not be wasted on multiple wavelengths but used to reach a greater diversity of target end-points. This avoids retrenching from the hub to original fiber node in order to lay additional fiber strands. Ideally, operators would only have to deploy more fiber from the original fiber node to deeper end-points in their networks.
As the industry evolves toward Node+0 architectures, the volume of optical connections to intelligent nodes will increase substantially compared to traditional architectures. Interoperability and a robust vendor ecosystem are therefore key to providing a low-cost solution using coherent optics.
With key goals being interoperability and vendor diversity, CableLabs intends to develop specifications which leverage the aforementioned benefits of coherent optics in the access network. Similar to previous specification development efforts, the coherent optics specifications will focus on interface requirements, signal integrity requirements, configuration, and management. As usual, CableLabs welcomes involvement from the vendor community to develop these specifications. In the near future, look for announcements related to the establishment of a coherent optics working group to develop the specifications.
At CableLabs, we are developing and specifying technology that allows the cable industry to support the growing requirements of broadband access. Come and join us in developing tomorrow’s high capacity network solutions!
Dr. Curtis Knittle, VP of Wired Technologies, also contributed to this article.
The Analog Monkey on Our Backs: Wrong-Thinking about Data over Cable
We inherit a lot of things from the past – not all of them good. When the telephone was invented by Elisha Grey, it evolved from wired telegraph lines. This was a natural evolution since the telegraph lines already existed and were proven, whereas radio technology was still in its infancy. When television was first introduced, it was radiated through the air to receivers in homes, as there were no cable systems initially, but radio broadcast by then had become a proven technology. However, this was not how the customers wanted to consume these services. When you are being entertained with pictures and sound you usually prefer to be seated. But when you want to talk to someone, you frequently are on the go. So due to technical constraints at the time, we ended up using a wired media for an ideally mobile product, and a wireless media for ideally sedentary product. This helps explains the marketing success of both cell phones and cable television.
As service providers, we are still adjusting the wired and wireless mix, and current technology is still influencing that adjustment.
Claude Shannon (1916-2001)
In 1947 Claude Shannon came along with a remarkable paper "Communication in the Presence of Noise" on the data capacity of a communications channel in the presence of random noise. He stated that the data capacity (C, in bits per second) of the channel is proportional to the bandwidth (B, in Hz) of the channel times log2(1 + signal power/noise power).
This brings up the question of what is the constraint on data capacity for any medium? If you want to increase the data capacity of a channel, is the channel lacking bandwidth, or is signal power constrained? The answer varies by medium. In wireless networks generally bandwidth is scarce. The US government sells it by the Hz, and it is very expensive. In fiber optic cable, signal power is usually required to be relatively low. A single mode fiber starts to go nonlinear at only about 10 milliwatts of power, but it has an enormous amount of bandwidth. On coaxial cable both power and bandwidth are constrained, and increased cable loss at high frequencies increases the noise power at higher frequencies. This is also true for telephony twisted pair, although the attenuation of twisted pair is much greater than coax. Cable’s capacity is not as constrained by Shannon’s Capacity Theorem as it is by common wrong assumptions, some of which are:
We need to deliver analog TV signals at any frequency that the cable plant carries.
Not true. There is no requirement for analog TV delivery at 1.8GHz, which is the highest frequency supported in the DOCSIS® 3.1 specification. In fact, the requirement to deliver ANY analog TV signals is disappearing, or is already gone.
4096 QAM is four times better than 1024 QAM.
Not true again. 4096 QAM transports 12 bits per symbol, and 1024 QAM transports 10 bits per symbol. So, 4096 has only 20% more capacity than 1024. But this 20% comes at an enormous price of 400% more required signal power. On the other hand, if you could somehow find 4X more bandwidth, you could use the same 4096 QAM signal power to transmit 400% more data. (Hint: look above 1GHz)
In RF line amplifiers, steep up-tilts are required.
Not true again. This is a carry-over from analog TV delivery, and is no longer required in DOCSIS 3.1 networks. In fact, the optimal way to spend your signal power is explained in Shannon’s classic paper, in particular Fig. 8. It is now referred to as the water-pour method. For the water-pour method, basically, in bands with low noise, it is optimal to use a higher percentage of the available signal power, and in bands with high noise, it is optimal to use lower percentage of the available signal power. Besides, it is just not practical from a signal power standpoint to extend the up-tilt to 1.8GHz for analog TV delivery.
A technical paper on this topic has been published on the CableLabs website.
In conclusion, our networks are going all digital, the old requirements have changed, and there is a lot of additional data capacity available on cable networks. Data capacity can be accessed both by using higher frequencies and reallocating signal power more wisely.
Tom Williams is a Principal Architect at CableLabs.
DOCSIS 3.1 Technology: Spec to Product in One Year
Around this time last year, CableLabs kicked off the first DOCSIS® 3.1 interoperability event. Looking back, it is amazing how far we have come.
Over the last year, CableLabs has held seven DOCSIS 3.1 interoperability events that provide manufacturers with the opportunity to work together on interoperability, development, and compliance. A total of 27 vendors across cable modem, head-end, and test equipment manufacturers have participated so far. Over the course of these events, vendors have continued to demonstrate improvements in product maturity.
Additionally, CableLabs has completed seven dry run events which provided vendors with additional development opportunities and evaluation of product readiness for certification and field trials through joint test execution between CableLabs and the vendors. This progress has paved the way for CableLabs to officially open the door for device certification submission and for cable operators to plan their field trials. While we can’t say much about what is going on in our certification labs, we can say it is like a beehive in there! CableLabs will be continuing the interoperability and dry run events in 2016 to support the industry and accelerate device availability.
One of the highlights of the year was the DOCSIS 3.1 technology demonstration day, where vendors flexed their muscles and showed the potential of the DOCSIS 3.1 technology. Demonstrations showcased the multi-Gbps capabilities of DOCSIS 3.1 technology, even while the products were still in development stage. Also, the ability to deploy DOCSIS 3.1 services in today’s networks through a fluid transition was demonstrated. Some vendors even showed how some of today’s cable modem termination systems, which are already deployed in the field, can be upgraded to support DOCSIS 3.1 technology while simultaneously supporting DOCSIS 3.1 and DOCSIS 3.0 devices. The state of the art in spectral efficiency was showcased with the support of 4096 QAM which is a 50% increase from what is possible in today’s DOCSIS 3.0 networks. All indicators are pointing in the right direction that DOCSIS 3.1 technology, as promised, will enable the cable operator to deliver great user experience to the customer.
In addition to holding several training workshops, CableLabs has been working closely with SCTE and NCTI to develop the required training courses to educate and prepare the workforce for the deployment of DOCSIS 3.1 services.
And as the year comes to an end, Comcast puts the icing on the cake by announcing the kickoff of their DOCSIS 3.1 field trials signaling that their customers will soon be enjoying the great experiences that can be offered by DOCSIS 3.1 technology. Congratulations to Comcast for such a big achievement, and ending the year on a high note for the industry and for its customers as a whole.
Belal Hamzeh is Director of Network Technologies at CableLabs.